diff --git a/.github/workflows/regression.yml b/.github/workflows/regression.yml new file mode 100644 index 00000000..3810cc53 --- /dev/null +++ b/.github/workflows/regression.yml @@ -0,0 +1,41 @@ +name: Build + +on: [push, pull_request] + +jobs: + build: + runs-on: ubuntu-latest + + defaults: + run: + shell: sh + + strategy: + matrix: + pgversion: + - 16 + - 15 + - 14 + - 13 + - 12 + - 11 + - 10 + + env: + PGVERSION: ${{ matrix.pgversion }} + + steps: + - name: checkout + uses: actions/checkout@v3 + + - name: install pg + run: | + sudo /usr/share/postgresql-common/pgdg/apt.postgresql.org.sh -v $PGVERSION -p -i + sudo apt-get install -f libdbd-pg-perl + sudo -u postgres createuser -s "$USER" + + - name: test + run: | + # 02_same_schema test currently broken + rm -fv t/02_same_schema.t + LC_ALL=C PERL_USE_UNSAFE_INC=1 PGBINDIR=/usr/lib/postgresql/$PGVERSION/bin prove t diff --git a/.gitignore b/.gitignore index 4489c953..318c2a28 100644 --- a/.gitignore +++ b/.gitignore @@ -14,3 +14,5 @@ test_database_check_postgres* cover_db/ check_postgres- MYMETA.* +/Vagrantfile +.vagrant diff --git a/.travis.yml b/.travis.yml new file mode 100644 index 00000000..eb1333ad --- /dev/null +++ b/.travis.yml @@ -0,0 +1,41 @@ +# run the check_postgres testsuite on travis-ci.com +--- +# versions to run on +env: + - PGVERSION=13 + - PGVERSION=12 + - PGVERSION=11 + - PGVERSION=10 + - PGVERSION=9.6 + - PGVERSION=9.5 + - PGVERSION=9.4 + - PGVERSION=9.3 + +dist: focal +sudo: required +language: perl +perl: + - '5.30.0' + - '5.14' # 5.14 is shipped with Ubuntu precise (12.04), also oldest version supported by Travis on focal + +before_install: + - sudo apt-get -qq update + +install: + - sudo apt-get install curl ca-certificates gnupg + - curl https://www.postgresql.org/media/keys/ACCC4CF8.asc | sudo apt-key add - + - sudo sh -c 'echo "deb http://apt.postgresql.org/pub/repos/apt $(lsb_release -cs)-pgdg main" > /etc/apt/sources.list.d/pgdg.list' + - sudo apt-get update + - sudo apt-get install postgresql-$PGVERSION + - pg_lsclusters + - dpkg -l postgresql\* | cat + - printenv | sort + +script: + - rm -rf test_database_check_postgres* + - perl Makefile.PL + - cpanm --quiet --installdeps --notest . + - PGBINDIR=/usr/lib/postgresql/$PGVERSION/bin make test TEST_VERBOSE=1 + +after_script: + - tail -n 200 test_database_check_postgres*/pg.log diff --git a/LICENSE b/LICENSE new file mode 100644 index 00000000..95cadcac --- /dev/null +++ b/LICENSE @@ -0,0 +1,21 @@ +Copyright 2007 - 2025 Greg Sabino Mullane + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + 1. Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. + 2. Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND ANY EXPRESS OR IMPLIED +WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF +MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO +EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, +EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT +OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING +IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY +OF SUCH DAMAGE. diff --git a/MANIFEST b/MANIFEST index c320ac26..ec6658ac 100644 --- a/MANIFEST +++ b/MANIFEST @@ -1,17 +1,20 @@ check_postgres.pl check_postgres.pl.asc check_postgres.pl.html -README +LICENSE +README.md SIGNATURE Makefile.PL MANIFEST MANIFEST.SKIP perlcriticrc META.yml +MYMETA.json MYMETA.yml TODO t/00_basic.t +t/00_release.t t/00_signature.t t/00_test_tester.t t/01_validate_range.t @@ -47,6 +50,7 @@ t/02_query_runtime.t t/02_query_time.t t/02_relation_size.t t/02_replicate_row.t +t/02_replication_slots.t t/02_same_schema.t t/02_sequence.t t/02_settings_checksum.t @@ -61,4 +65,7 @@ t/03_translations.t t/04_timeout.t t/05_docs.t t/99_cleanup.t +t/99_perlcritic.t +t/99_pod.t +t/99_spellcheck.t t/CP_Testing.pm diff --git a/MANIFEST.SKIP b/MANIFEST.SKIP index 63d276f2..8f2691c0 100644 --- a/MANIFEST.SKIP +++ b/MANIFEST.SKIP @@ -16,6 +16,7 @@ t/99_perlcritic.t t/99_pod.t t/00_release.t t/99_spellcheck.t +.travis.yml README.dev .check_postgresrc ^blame diff --git a/META.yml b/META.yml index 0a6605d1..0cf8dcde 100644 --- a/META.yml +++ b/META.yml @@ -1,16 +1,16 @@ --- #YAML:1.0 name : check_postgres.pl -version : 2.21.0 +version : 2.26.0 abstract : Postgres monitoring script author: - - Greg Sabino Mullane + - Greg Sabino Mullane license : bsd distribution_type : script dynamic_config : 0 requires: - perl : 5.006001 + perl : 5.10.0 build_requires: Test::More : 0.61 recommends: @@ -30,7 +30,7 @@ recommends: provides: check_postgres: file : check_postgres.pl - version : 2.21.0 + version : 2.26.0 keywords: - Postgres @@ -44,7 +44,7 @@ resources: homepage : http://bucardo.org/check_postgres/ license : http://bucardo.org/check_postgres/ bugtracker : https://github.com/bucardo/check_postgres/issues - MailingList : https://mail.endcrypt.com/mailman/listinfo/check_postgres + MailingList : https://bucardo.org/mailman/listinfo/check_postgres Repository : git://bucardo.org/check_postgres.git meta-spec: diff --git a/Makefile.PL b/Makefile.PL index b5fd73a9..607a71cc 100644 --- a/Makefile.PL +++ b/Makefile.PL @@ -4,9 +4,9 @@ use ExtUtils::MakeMaker qw/WriteMakefile/; use Config; use strict; use warnings; -use 5.006001; +use 5.10.0; -my $VERSION = '2.21.0'; +my $VERSION = '2.26.0'; if ($VERSION =~ /_/) { print "WARNING! This is a test version ($VERSION) and should not be used in production!\n"; @@ -54,9 +54,9 @@ print "Configuring check_postgres $VERSION\n"; my %opts = ( NAME => 'check_postgres', ABSTRACT => 'Postgres monitoring script', - AUTHOR => 'Greg Sabino Mullane ', + AUTHOR => 'Greg Sabino Mullane ', PREREQ_PM => { - 'ExtUtils::MakeMaker' => '6.11', + 'ExtUtils::MakeMaker' => '6.64', 'Test::More' => '0.61', 'version' => '0', }, @@ -67,6 +67,11 @@ my %opts = ( NEEDS_LINKING => 0, NORECURS => 1, PM => {}, + TEST_REQUIRES => { + 'DBD::Pg' => '2.0', + 'DBI' => '1.51', + 'Date::Parse' => '0', + }, clean => { FILES => join ' ' => @cleanfiles }, ); @@ -100,10 +105,9 @@ sub clean { ## no critic (RequireArgUnpacking) $string .= qq{\t@ gpg --verify check_postgres.pl.asc\n}; $string .= qq{\n\nhtml : \n\t}; $string .= <<'EOM'; - pod2html check_postgres.pl > check_postgres.pl.html + pod2html --title check_postgres.pl check_postgres.pl > check_postgres.pl.html @ perl -pi -e "s///" check_postgres.pl.html @ perl -pi -e "s~ git clone.*~ git clone git://bucardo.org/check_postgres.git~" check_postgres.pl.html - @ perl -pi -e "s~\S+(.+)~<title>check_postgres.pl\\1~" check_postgres.pl.html @ perl -pi -e "s~\`\`(.+?)''~"\\1"~g" check_postgres.pl.html @ rm -f pod2htmd.tmp pod2htmi.tmp EOM diff --git a/README.dev b/README.dev index 744cf849..f93edf62 100644 --- a/README.dev +++ b/README.dev @@ -8,18 +8,23 @@ For testing PostgreSQL 9.2 and later, DBD::Pg 2.19.3 is required. Running the testsuite: +* env -i make test * LC_ALL=C make test * initdb and friends not in $PATH: LC_ALL=C make test PGBINDIR=/usr/lib/postgresql/9.2/bin * Run a single test: LC_ALL=C make test TEST_FILES=t/02_database_size.t * Skip network tests: LC_ALL=C make test SKIP_NETWORK_TESTS=1 +The testsuite is sensitive to locale and PG environment variables such as LANG +and PGDATABASE. Using `env -i` will unset all variables from the user +environment for running the tests. + ** RELEASE PROCESS ** Quick notes on the current release process: * Make sure all changes are recorded in the relevant POD section. * Add a release date next to the new version number -* Change the version number everywhere (use prove -v t/00_release.t to verify) +* Change the version number everywhere (use `RELEASE_TESTING=1 prove -v t/00_release.t` to verify) * git commit as needed * Run: perl Makefile.PL; make html * Run: make signature_asc @@ -47,7 +52,7 @@ Login to the bucardo.org box, and then: * edit latest_version.txt * edit index.html -* Edit the bucardo.org page and bump the version: http://bucardo.org/wiki/Check_postgres +* Edit the bucardo.org page and bump the version: http://bucardo.org/check_postgres * Email to check_postgres-announce with summary of changes diff --git a/README b/README.md similarity index 61% rename from README rename to README.md index 82028cae..72adeb84 100644 --- a/README +++ b/README.md @@ -1,103 +1,104 @@ -check_postgres is Copyright (C) 2007-2013, Greg Sabino Mullane +check_postgres +============== + +[![Build Status](https://travis-ci.com/bucardo/check_postgres.svg?branch=master)](https://travis-ci.com/bucardo/check_postgres) This is check_postgres, a monitoring tool for Postgres. The most complete and up to date information about this script can be found at: -http://bucardo.org/check_postgres/ - -You should go check there right now to make sure you are installing -the latest version! +https://bucardo.org/check_postgres/ This document will cover how to install the script. -* Quick method: +Quick method +------------ -For the impatient Nagios admin, just copy the "check_postgres.pl" file -to your Nagios scripts directory, and perhaps symlink entries to that +For the impatient Nagios admin, just copy the "check_postgres.pl" file +to your Nagios scripts directory, and perhaps symlink entries to that file by: -cd <the directory you just copied the file to> -mkdir postgres -cd postgres -perl ../check_postgres.pl --symlinks + cd <the directory you just copied the file to> + mkdir postgres + cd postgres + perl ../check_postgres.pl --symlinks Then join the announce mailing list (see below) - -* Complete method: +Complete method +--------------- The better way to install this script is via the standard Perl process: -perl Makefile.PL -make -make test -make install + perl Makefile.PL + make + env -i make test + make install -The last step usually needs to be done as the root user. You may want to -copy the script to a place that makes more sense for Nagios, if using it +The last step usually needs to be done as the root user. You may want to +copy the script to a place that makes more sense for Nagios, if using it for that purpose. See the "Quick" instructions above. -For 'make test', please report any failing tests to check_postgres@bucardo.org. -The tests need to have some standard Postgres binaries available, such as -'initdb', 'psql', and 'pg_ctl'. If these are not in your path, or you want to -use specific ones, please set the environment variable PGBINDIR first. +For `make test`, please report any failing tests to check_postgres@bucardo.org. +The tests need to have some standard Postgres binaries available, such as +`initdb`, `psql`, and `pg_ctl`. If these are not in your path, or you want to +use specific ones, please set the environment variable `PGBINDIR` first. More +details on running the testsuite are available in `README.dev`. -Once 'make install' has been done, you should have access to the complete +Once `make install` has been done, you should have access to the complete documentation by typing: -man check_postgres + man check_postgres The HTML version of the documentation is also available at: -http://bucardo.org/check_postgres/check_postgres.pl.html +https://bucardo.org/check_postgres/check_postgres.pl.html + +Development happens via git. You can check out the repository by doing: + + https://github.com/bucardo/check_postgres + git clone https://github.com/bucardo/check_postgres.git -* Mailing lists + +Mailing lists +------------- The final step should be to subscribe to the low volume check_postgres-announce mailing list, so you learn of new versions and important changes. Information on joining can be found at: -https://mail.endcrypt.com/mailman/listinfo/check_postgres-announce +https://bucardo.org/mailman/listinfo/check_postgres-announce General questions and development issues are discussed on the check_postgres list, which we recommend people join as well: -https://mail.endcrypt.com/mailman/listinfo/check_postgres - -Development happens via git. You can check out the repository by doing: - -git clone git://bucardo.org/check_postgres.git - -All changes are sent to the commit list: - -https://mail.endcrypt.com/mailman/listinfo/check_postgres-commit +https://bucardo.org/mailman/listinfo/check_postgres -COPYRIGHT: ----------- - Copyright (c) 2007-2013 Greg Sabino Mullane +COPYRIGHT +--------- + Copyright 2007 - 2025 Greg Sabino Mullane -LICENSE INFORMATION: --------------------- +LICENSE INFORMATION +------------------- -Redistribution and use in source and binary forms, with or without +Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: - 1. Redistributions of source code must retain the above copyright notice, + 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation + 2. Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. -THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND ANY EXPRESS OR IMPLIED -WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF -MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO +THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND ANY EXPRESS OR IMPLIED +WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF +MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, -EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT -OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING -IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY +EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT +OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING +IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. diff --git a/SIGNATURE b/SIGNATURE index 866791d0..e856b0fb 100644 --- a/SIGNATURE +++ b/SIGNATURE @@ -1,5 +1,5 @@ This file contains message digests of all files listed in MANIFEST, -signed via the Module::Signature module, version 0.68. +signed via the Module::Signature module, version 0.83. To verify the content in this distribution, first make sure you have Module::Signature installed, then type: @@ -14,71 +14,78 @@ not run its Makefile.PL or Build.PL. -----BEGIN PGP SIGNED MESSAGE----- Hash: RIPEMD160 -SHA1 3daf641677071210a1702e075f9837becd504d9d MANIFEST -SHA1 98fcae49e2adf1d73e0687998ca33992451984d6 MANIFEST.SKIP -SHA1 221581ac5b59c2cf634f6a521f02f5e4659e048d META.yml -SHA1 2c4d6ce8e44d088b6068dc07bfe902b7f3f350b3 MYMETA.yml -SHA1 6d4a6c15891a90886fdb9a4f37e80c3a1ccbc492 Makefile.PL -SHA1 0382a8d7bd1e6c8e2e85be867173ab231f78a66b README -SHA1 e4e41aec52387ddfb9ebe9257a5eee40700adb00 TODO -SHA1 49424a5f4aae35cda48709bc42b47a3bea94c345 check_postgres.pl -SHA1 f34b99a45f1670576d28eacb42b0e8c56c3245a0 check_postgres.pl.asc -SHA1 0bda94edc7eeab34fc73bfe105f566a6f6def2f5 check_postgres.pl.html -SHA1 3dc431ab18171fa977f919cddcee504b4b74392f perlcriticrc -SHA1 ef43082a685d993fdd151de16590ce0f6832de7a t/00_basic.t -SHA1 29700e8331d1780e87b858581054cd190e4f4ecb t/00_signature.t -SHA1 439053254ee01e2ec406f9d1605dc5e28257b8bd t/00_test_tester.t -SHA1 02f59725d968d85d855b0bcc771c99003c6e0389 t/01_validate_range.t -SHA1 7b9ef508bc0fc1624dc81129b4014a1b9704db7e t/02_autovac_freeze.t -SHA1 c50f069d276e44b95408fe16aefa46c24745a038 t/02_backends.t -SHA1 923acca525a528ccefd630eef8e015007a1bc5d0 t/02_bloat.t -SHA1 58f04f98539371d9469ca93890b3f61a990fa343 t/02_checkpoint.t -SHA1 c3e31bedea2a55b49d61874a7e4b8f6900746b22 t/02_cluster_id.t -SHA1 e4a26b2713f0a52f31a38fb9b522a7f6e8bb9d8e t/02_commitratio.t -SHA1 4c403c8d733fff18e8ef5b198c0f0837419b4e77 t/02_connection.t -SHA1 3ec5348a4125429a16e486a3850baa43954d278a t/02_custom_query.t -SHA1 6e255ba83d65c41c9f27ec05b3c66b57f531b42e t/02_database_size.t -SHA1 4d5d6b521ae8ec45cb14feea155858379ec77028 t/02_dbstats.t -SHA1 f01542845070822b80cac7aaa3038eae50f8d3ae t/02_disabled_triggers.t -SHA1 cee0a31024e2f6340a8419d71df2691ac1005e3c t/02_disk_space.t -SHA1 316cd71675cb133e0ff97e1ec27e8ab9211330af t/02_fsm_pages.t -SHA1 770c9c0293746f10e5f48b1251e9ea5e91ee2210 t/02_fsm_relations.t -SHA1 6294bc1dcd57a8db469deb9a69aebd1eb5998ae9 t/02_hitratio.t -SHA1 f83838ef9da1d67e026f88b833522a73f3ae6699 t/02_last_analyze.t -SHA1 80de470784f910b1775adbf616ac620f30d8fdcc t/02_last_vacuum.t -SHA1 5fca9f38305fd0c8c5d1726d08f0a95ae8755b8e t/02_listener.t -SHA1 876b469168626ae733eea9999937e94dcdac2c9b t/02_locks.t -SHA1 a780920848536059af81bc77140ed70051630d00 t/02_logfile.t -SHA1 36e9e9e9a1cfb4ad8ae0793ed66b2750cbcf9888 t/02_new_version_bc.t -SHA1 4e329549179fc8a7e3228ee9383fc4978641d6ff t/02_new_version_box.t -SHA1 76eb1a2d2bf4f13ad5530ec6ec1431feabf2bf34 t/02_new_version_cp.t -SHA1 61b60c4ba4b93d7813786fcec6c66c0825f08858 t/02_new_version_pg.t -SHA1 276df93bed5f6147351dc577706dc3ab30ba0555 t/02_new_version_tnm.t -SHA1 02091d75f8371671d85cc0be37bd76b26edcfc3e t/02_pgagent_jobs.t -SHA1 73ef3b85a5557de783c756f3eebaeac70c1bbed1 t/02_pgbouncer_checksum.t -SHA1 5eac7fd171c118d2c358ae719f8e6397883c5a12 t/02_prepared_txns.t -SHA1 2706f51b92149330f123e174ad009ada0124a538 t/02_query_runtime.t -SHA1 033bb1162ad990161da1c591d5fb9f7c452ee4c9 t/02_query_time.t -SHA1 bdd9a4a73bbce6a5450cbc04e1fe357f3bc587a7 t/02_relation_size.t -SHA1 94fd92c8a3453e86d80114009f8a40ccb709704b t/02_replicate_row.t -SHA1 e36d0632a32b22ebdded6ab91fd46aca7f8b2b31 t/02_same_schema.t -SHA1 23df28d99d8799d85e7edb55ccbdcfc68fbfa22f t/02_sequence.t -SHA1 65ea5ff56452e71fefde682bf5e874d30800bd37 t/02_settings_checksum.t -SHA1 17ab792f132cd6fabd6fa33598b5a4d502428543 t/02_slony_status.t -SHA1 64493100381abd82aa1eb06a46d67456f4e8486d t/02_timesync.t -SHA1 221d3c78eece2a5fc796611e3133e5ccc8f8fad0 t/02_txn_idle.t -SHA1 7c77be0301c936974778c7c92186088ef9b06526 t/02_txn_time.t -SHA1 24ff2b5b0690557e1a45227d5c1131226c9ff30a t/02_txn_wraparound.t -SHA1 2270e466a5761256be6b69cc0c7e8c03f2414e3b t/02_version.t -SHA1 93c5da727c39f3dbb1953d061fa6f96578ba7952 t/02_wal_files.t -SHA1 dad138868393ead7c170f24017a7c04b80fe5f87 t/03_translations.t -SHA1 eb66336f915f77a1cd82c1b19b44acc7dc2d5038 t/04_timeout.t -SHA1 e01ec73ad338765ee20afed29718b12c2ed83d82 t/05_docs.t -SHA1 96e6e5da3e1f6753a469e7afc8cf070024b33b23 t/99_cleanup.t -SHA1 095c6b9fdc944faf7890f13d5330b179f41710a1 t/CP_Testing.pm +SHA256 b903bc1c84af2abee7ea553beb9d85cc49031478b32dcb9d8245ed61b1611cab LICENSE +SHA256 95b887fc1e39944b6056c809f5910b8df76026bd6d837b37dc0d40bc9c8d27df MANIFEST +SHA256 0083cbd79dcd094b2b37014ebba34bb77f95f854f6f56f48db2c77e16ddcec3f MANIFEST.SKIP +SHA256 b03bdcd32c37f8352a2bc9833a2885217f953d8c7d6b441d73fc30a90836520e META.yml +SHA256 230f47a73ed0f6106d2c9a79f583c8ba66a2ce283180902d42de1e4c125100c9 MYMETA.json +SHA256 8facd1a45236a702d8c7464f9254fe3b40d993303d12386553d4b0f473561a87 MYMETA.yml +SHA256 5bdc278ad01b4eea46d4d017b30b70d8bde3c387777624d7e71d8c4ce8142a68 Makefile.PL +SHA256 074851520b04909ab15f3e64015ba19be8791124c9927a34c58e28a194092a35 README.md +SHA256 b32a224f3752f4a608dc6959d614626a2bc054daa7c086dae77af727e19623d6 TODO +SHA256 bde647425d1ead756b47170f378876af9c52a694801199a581571095136c3cb0 check_postgres.pl +SHA256 23920745377364d3e175b4dacece36495072b24fedace08bed83f2efc03f61d4 check_postgres.pl.asc +SHA256 f980b970e772001084e308294574dce800dcb6cfd2c74b689b55810e1b44fab1 check_postgres.pl.html +SHA256 ac0bf76395788f0f80acb21cd44c9fe642f81ef9398367cd2daf0cd498875f64 perlcriticrc +SHA256 9fcca73430b5197ebda1034292297e22b22f1a2d0bbc560f69c5881893c79da8 t/00_basic.t +SHA256 1ba0aa962c0a3030b84f07efc2037617b689a81bb8062d27e7b9151f5eea0772 t/00_release.t +SHA256 3f1437e52fd7822b3cb643abbbbdf996102f8309ec21821d7d04943eb0ae6613 t/00_signature.t +SHA256 09d6eba96ad41a2afd5d7ab44ccd0cf2e63721ba77a5e2f848af079e84013abd t/00_test_tester.t +SHA256 612a12f11c5d2f8f5a189d3b9ba4bcbe42db6c34956674a64c248c1673fbf757 t/01_validate_range.t +SHA256 0c41f4bef5c24ec4bd142f97794d20e63a71c22ca59fe4ede3a0da6438ab3a2a t/02_autovac_freeze.t +SHA256 640d63df6f1ed15dff2dd30e9c0af903817a89c9527bc2f36c7b9cd4919cb8d8 t/02_backends.t +SHA256 326278182077ed03b0162f6272413fdb7fc8b03b497e0f6b119eaf93ca9bc447 t/02_bloat.t +SHA256 ec05dd611c31478ad934d583fe4e1733e45beb8d7df8a86dce45a5d5c391a669 t/02_checkpoint.t +SHA256 8845a57f1341f9b114f5f89cdfbb531b6d6ca0243ec1c9f3b9b7b5be826055fe t/02_cluster_id.t +SHA256 5577c0c9cf0ccede67bb46bc07840b7d133cbc684908ee2e10923f30fa2101b6 t/02_commitratio.t +SHA256 1c94f9e5e4ada7db941f8dfb155727f94f087d993423d5ff92323660f52b1c2c t/02_connection.t +SHA256 d5ae165b2ecabce5bf5c3ef8ec1f1a2969389b4c1de158e619b2c5ecf2c9a058 t/02_custom_query.t +SHA256 f695bfe8898289d314ba3e983834719569a4033d89b92bf0410817d5d2a98d60 t/02_database_size.t +SHA256 25c4b66dd04d03dcaf49bfee7354ae03b456713b7bca024c1948cdddd9d5e0f9 t/02_dbstats.t +SHA256 01b0f8863d56ee11a6fa51a23926c0ec6c594617c2420c6a952ba25be9df015f t/02_disabled_triggers.t +SHA256 adb114dd1db5c6643bc51d9b8dbdae04ee54f7549d99a455e0524a708dce92a4 t/02_disk_space.t +SHA256 91ca4f95397785a752eae9eea09c5130b61a1adbd8e7bebd3cac99b36c4b8c8b t/02_fsm_pages.t +SHA256 04c2cf961e8857c1c43370b37201292454985962d05fcf0c6129536610f54706 t/02_fsm_relations.t +SHA256 d8a19295044d5bb7bb17c1d7b607adcb24e00d79b3befe55d63a44e004d4a92f t/02_hitratio.t +SHA256 b92cb79e99a6267b1f54ae4b2e6d12110dd30d7c3af2b22c5abae857e096a0a5 t/02_last_analyze.t +SHA256 43918b18d44847fdf71e6136c6194fef29cddb51a32626fef6fc5602f338ca0e t/02_last_vacuum.t +SHA256 173cc863f6fd95ead67e78673b656ef01b480bf8f18598e063a1238f00e0d394 t/02_listener.t +SHA256 1ab0543da8ede0f8916dd3ffc40864ac5429e168ea244817bbc5b0ef5642e455 t/02_locks.t +SHA256 a47ce9b00e47cd81089a932e1d6d3628c9e77ff59d3383520e1870979b1aa986 t/02_logfile.t +SHA256 aac8818c41e43d419d29bb7681208a6c84896c6e0369d5412acf20e6f5c1e195 t/02_new_version_bc.t +SHA256 5c6fa6b5b770deb53336543c0c3a64aeba3dbec3ea2d25576bdae0b807c289c4 t/02_new_version_box.t +SHA256 a6a74a835a9ded9ba5cc1e32d9ffa7abc5a012426572c807bacb5f0ba95606ec t/02_new_version_cp.t +SHA256 95b7353a1bf11319d1ead30ff57bc1f59c7a55c788ca625e62b1d26a4f5aa4e8 t/02_new_version_pg.t +SHA256 f8e679f9863acafaa32bd399687dbce4230088800fc6f14d6b7fe49668f13188 t/02_new_version_tnm.t +SHA256 5fa86842561eb8c5f6fb53e70bbcfde7dd6cfc67666117e85e2aa65013c98859 t/02_pgagent_jobs.t +SHA256 a044428fe50a1090dd32dc840a0e6450f2bc39d1c99e27465b57a2520bb75d48 t/02_pgbouncer_checksum.t +SHA256 b6678acbbe5d0ca8d53bdf968ac9d35b864b6d7820bca08c8f06b6fc4060ca40 t/02_prepared_txns.t +SHA256 742f2c1139df5adb389b7ef1fa4a7fddcab74575e4f6064608bf49e80021d3fb t/02_query_runtime.t +SHA256 a274c07e334ab778592e40614bba8c96521396ff8ed5f979a8c9f8c4e8af2d13 t/02_query_time.t +SHA256 54ab8c6465cfbe8d1a70bdbd9b38d65ff1dcdaa39019efbf7f5aba3c9f320670 t/02_relation_size.t +SHA256 2bbeda7b00699b1dee34782660e59fd85b8439efc7157045dfbf674dc43ccba1 t/02_replicate_row.t +SHA256 20b06267816c0b0743d6ae08ec52554d51664daa2649a57afafd497478fa679d t/02_replication_slots.t +SHA256 91d9c85ddb272583d27d07092eee0ae0f05c6ed8a5fc1dfd858f30b4347ce3bb t/02_same_schema.t +SHA256 6ddc739944702f159e184d8e76b7f674c47ea7a02dfcfe4c462fa580113dc1f6 t/02_sequence.t +SHA256 0f95ed94e593b500f5f355e7e5e08c8ec2f396736ce6c15f39f10daa37abccf7 t/02_settings_checksum.t +SHA256 34463c9bc4fc4a5cbe042ebbdb65485c3846718bdf8d17cce7b45f2f5774bb02 t/02_slony_status.t +SHA256 6c2bf940066e9c02c692d047db6c86cb6a21be2dff3d1f233da35a355a835ac3 t/02_timesync.t +SHA256 d098d67860ef4e0597b7c06b02c4472b8ec4a17ea76c7742e600fbdfa5a3262c t/02_txn_idle.t +SHA256 d682e42340d57b4b367b602fe6600fc418aa357fba2332941a287654b32bfd30 t/02_txn_time.t +SHA256 9b8ab57fda660eb514888fc909bf398c07c7a18f4d5bf5704f512c8d0b4c5053 t/02_txn_wraparound.t +SHA256 38c4a9462bcc41f6cddef3f6775f2f5fdf4d7532b4a089d34863e431892e0be3 t/02_version.t +SHA256 5b1008817f231630eb72982745b7667f267abf85d27aff3fb2acce717cd0eb7a t/02_wal_files.t +SHA256 4385a1b1734a885996c9d3ce8b1224adbb8a6f960480fd43a2f13ccf99fac9f8 t/03_translations.t +SHA256 f6595585ba848b022915adb012d644a1f94ee7bfadab7f2499b29b0d2573049a t/04_timeout.t +SHA256 8edd2b72aa8905a6719adbf8deb20e45d07129ebb3bfcb0f0dc815e92a4aacb0 t/05_docs.t +SHA256 f93b083dadcef109b65dc6b1a508df58283714535a4bdb2a728d3a42261253bc t/99_cleanup.t +SHA256 6ef7226a9573b0b3b3d2b9e8da188797a89e9ad76e14df1828152415ce984e8e t/99_perlcritic.t +SHA256 c42cd73f37e4aaf149cd4dbd62508863b33e78d47e9182c38ac3e42ea961016b t/99_pod.t +SHA256 e59846d99d3501bcc7ed7b857bd432685a195c76a12cec309eec3771e2e8f987 t/99_spellcheck.t +SHA256 2de6d4d74f7c4d9e3140cd3ea1d7beb72759d6316ff995ac758d10e29998b7a9 t/CP_Testing.pm -----BEGIN PGP SIGNATURE----- -iEYEAREDAAYFAlJB46QACgkQvJuQZxSWSshIPACgwFPjhdZ2REkyKirLaMYSwznp -ptIAn0ttmsPV4cKBfUp/wFUumCMi/A/y -=+efS +iEYEAREDAAYFAl44wHsACgkQvJuQZxSWSsjS/wCZAYUvoGzvbxIL6YdEuKsm7FQI +1vYAn3vjQ5x6FEqjSMR/bB1+jCj3gAii +=zw5E -----END PGP SIGNATURE----- diff --git a/TODO b/TODO index bd0e61c0..586eb55a 100644 --- a/TODO +++ b/TODO @@ -2,6 +2,10 @@ Quick list of outstanding items / bugs / feature requests for CP: NOTE: All bugzilla items are now on github +* Add --setup and --reset flags to replicate_row to automate things. + +* Add documentation on privs required for each action, and recommended grants for roles, etc. + * The same_schema action does not check indexes. See bugzilla #54 * Perform automatic creation of views and function to allow all actions to be run diff --git a/check_postgres.pl b/check_postgres.pl index 60e86c3b..928f39a6 100755 --- a/check_postgres.pl +++ b/check_postgres.pl @@ -1,27 +1,29 @@ #!/usr/bin/env perl -# -*-mode:cperl; indent-tabs-mode: nil-*- +# -*-mode:cperl; indent-tabs-mode: nil; cperl-indent-level: 4 -*- ## Perform many different checks against Postgres databases. ## Designed primarily as a Nagios script. ## Run with --help for a summary. ## -## Greg Sabino Mullane <greg@endpoint.com> +## Greg Sabino Mullane <greg@turnstep.com> +## ## End Point Corporation http://www.endpoint.com/ ## BSD licensed, see complete license at bottom of this script ## The latest version can be found at: -## http://www.bucardo.org/check_postgres/ +## https://bucardo.org/check_postgres/ ## ## See the HISTORY section for other contributors package check_postgres; -use 5.006001; +use 5.10.0; use strict; use warnings; use utf8; use Getopt::Long qw/GetOptions/; Getopt::Long::Configure(qw/ no_ignore_case pass_through /); use File::Basename qw/basename/; +use File::Spec::Functions; use File::Temp qw/tempfile tempdir/; File::Temp->safe_level( File::Temp::MEDIUM ); use Cwd; @@ -32,7 +34,8 @@ package check_postgres; binmode STDOUT, ':encoding(UTF-8)'; -our $VERSION = '2.21.0'; +our $VERSION = '2.26.0'; +our $COMMA = ','; use vars qw/ %opt $PGBINDIR $PSQL $res $COM $SQL $db /; @@ -87,6 +90,7 @@ package check_postgres; ## Items without a leading tab still need translating ## no critic (RequireInterpolationOfMetachars) our %msg = ( +## English 'en' => { 'address' => q{address}, 'age' => q{age}, @@ -111,6 +115,7 @@ package check_postgres; 'checkpoint-baddir2' => q{pg_controldata could not read the given data directory: "$1"}, 'checkpoint-badver' => q{Failed to run pg_controldata - probably the wrong version ($1)}, 'checkpoint-badver2' => q{Failed to run pg_controldata - is it the correct version?}, + 'checkpoint-nobin' => q{Could not find a suitable pg_controldata executable}, 'checkpoint-nodir' => q{Must supply a --datadir argument or set the PGDATA environment variable}, 'checkpoint-nodp' => q{Must install the Perl module Date::Parse to use the checkpoint action}, 'checkpoint-noparse' => q{Unable to parse pg_controldata output: "$1"}, @@ -127,7 +132,6 @@ package check_postgres; 'custom-nostring' => q{Must provide a query string}, 'database' => q{database}, 'dbsize-version' => q{Target database must be version 8.1 or higher to run the database_size action}, - 'depr-pgcontroldata' => q{PGCONTROLDATA is deprecated, use PGBINDIR instead.}, 'die-action-version' => q{Cannot run "$1": server version must be >= $2, but is $3}, 'die-badtime' => q{Value for '$1' must be a valid time. Examples: -$2 1s -$2 "10 minutes"}, 'die-badversion' => q{Invalid version string: $1}, @@ -135,7 +139,7 @@ package check_postgres; 'die-nosetting' => q{Could not fetch setting '$1'}, 'diskspace-fail' => q{Invalid result from command "$1": $2}, 'diskspace-msg' => q{FS $1 mounted on $2 is using $3 of $4 ($5%)}, - 'diskspace-nodata' => q{Could not determine data_directory: are you connecting as a superuser?}, + 'diskspace-nodata' => q{Could not determine data_directory: are you running as a superuser?}, 'diskspace-nodf' => q{Could not find required executable /bin/df}, 'diskspace-nodir' => q{Could not find data directory "$1"}, 'file-noclose' => q{Could not close $1: $2}, @@ -159,6 +163,7 @@ package check_postgres; 'listening' => q{listening}, 'locks-msg' => q{total "$1" locks: $2}, 'locks-msg2' => q{total locks: $1}, + 'lockwait-msg' => q{$1: $2($3) blocking $4($5) for $6 blocked statement "$7"}, 'logfile-bad' => q{Invalid logfile "$1"}, 'logfile-debug' => q{Final logfile: $1}, 'logfile-dne' => q{logfile $1 does not exist!}, @@ -186,7 +191,10 @@ package check_postgres; 'no-match-set' => q{No matching settings found due to exclusion/inclusion options}, 'no-match-table' => q{No matching tables found due to exclusion/inclusion options}, 'no-match-user' => q{No matching entries found due to user exclusion/inclusion options}, + 'no-match-slot' => q{No matching replication slots found due to exclusion/inclusion options}, + 'no-match-slotok' => q{No replication slots found}, 'no-parse-psql' => q{Could not parse psql output!}, + 'no-role' => q{Need psql 9.6 or higher to use --role}, 'no-time-hires' => q{Cannot find Time::HiRes, needed if 'showtime' is true}, 'opt-output-invalid' => q{Invalid output: must be 'nagios' or 'mrtg' or 'simple' or 'cacti'}, 'opt-psql-badpath' => q{Invalid psql argument: must be full path to a file named psql}, @@ -195,12 +203,18 @@ package check_postgres; 'opt-psql-nofind' => q{Could not find a suitable psql executable}, 'opt-psql-nover' => q{Could not determine psql version}, 'opt-psql-restrict' => q{Cannot use the --PGBINDIR or --PSQL option when NO_PSQL_OPTION is on}, + 'partman-premake-ok' => q{All premade partitions are present}, + 'partman-conf-tbl' => q{misconfigured in partman.part_config}, + 'partman-conf-mis' => q{missing table in partman.part_config}, 'pgagent-jobs-ok' => q{No failed jobs}, 'pgbouncer-pool' => q{Pool=$1 $2=$3}, 'pgb-backends-mrtg' => q{DB=$1 Max connections=$2}, 'pgb-backends-msg' => q{$1 of $2 connections ($3%)}, 'pgb-backends-none' => q{No connections}, 'pgb-backends-users' => q{$1 for number of users must be a number or percentage}, + 'pgb-maxwait-msg' => q{longest wait: $1s}, + 'pgb-maxwait-nomatch'=> q{No matching rows were found}, + 'pgb-maxwait-skipped'=> q{No matching rows were found (skipped rows: $1)}, 'PID' => q{PID}, 'port' => q{port}, 'preptxn-none' => q{No prepared transactions found}, @@ -245,6 +259,7 @@ package check_postgres; 'relsize-msg-reli' => q{largest relation is index "$1": $2}, 'relsize-msg-relt' => q{largest relation is table "$1": $2}, 'relsize-msg-tab' => q{largest table is "$1": $2}, + 'relsize-msg-indexes' => q{table with largest indexes is "$1": $2}, 'rep-badarg' => q{Invalid repinfo argument: expected 6 comma-separated values}, 'rep-duh' => q{Makes no sense to test replication with same values}, 'rep-fail' => q{Row not replicated to slave $1}, @@ -256,7 +271,8 @@ package check_postgres; 'rep-sourcefail' => q{Source update failed}, 'rep-timeout' => q{Row was not replicated. Timeout: $1}, 'rep-unknown' => q{Replication check failed}, - 'rep-wrongvals' => q{Cannot test replication: values are not the right ones ($1 not $2 nor $3)}, + 'rep-wrongvals' => q{Cannot test replication: values are not the right ones ('$1' not '$2' nor '$3')}, + 'repslot-version' => q{Database must be version 9.4 or higher to check replication slots}, 'runcommand-err' => q{Unknown error inside of the "run_command" function}, 'runcommand-nodb' => q{No target databases could be found}, 'runcommand-nodupe' => q{Could not dupe STDERR}, @@ -268,6 +284,7 @@ package check_postgres; 'runtime-badname' => q{Invalid queryname option: must be a simple view name}, 'runtime-msg' => q{query runtime: $1 seconds}, 'schema' => q{Schema}, + 'ss-badobject' => q{Unrecognized object types: $1}, 'ss-createfile' => q{Created file $1}, 'ss-different' => q{"$1" is different:}, 'ss-existson' => q{Exists on:}, @@ -343,6 +360,271 @@ package check_postgres; 'wal-numfound' => q{WAL files found: $1}, 'wal-numfound2' => q{WAL "$2" files found: $1}, }, + +## Spanish +'es' => { + 'address' => q{dirección}, + 'age' => q{edad}, + 'backends-fatal' => q{No es posible conectar: demasiadas conexiones}, + 'backends-mrtg' => q{DB=$1 Max conexiones=$2}, + 'backends-msg' => q{$1 de $2 conexiones ($3%)}, + 'backends-nomax' => q{No es posible determinar max_connections}, + 'backends-oknone' => q{No hay conexiones}, + 'backends-po' => q{lo siento, hay demasiados clientes}, + 'backends-users' => q{$1 debe ser un número o porcentaje de usuarios}, + 'bloat-index' => q{(bd $1) índice $2 filas:$3 páginas:$4 deberíaser:$5 ($6X) bytes desperdiciados:$7 ($8)}, + 'bloat-nomin' => q{ninguna relación cumple el criterio de bloat mínimo}, + 'bloat-table' => q{(bd $1) tabla $2.$3 filas:$4 páginas:$5 deberíaser:$6 ($7X) bytes desperdiciados:$8 ($9)}, + 'bug-report' => q{Por favor reporte estos detalles a check_postgres@bucardo.org:}, + 'checkcluster-id' => q{Identificador de la base de datos:}, +'checkcluster-msg' => q{cluster_id: $1}, + 'checkcluster-nomrtg'=> q{Debe proporcionar un número con la opción --mrtg}, + 'checkmode-prod' => q{en producción}, + 'checkmode-recovery' => q{en recuperación de archivo}, + 'checkmode-state' => q{Estado del clúster de base de datos:}, + 'checkpoint-baddir' => q{Directorio de datos inválido: "$1"}, + 'checkpoint-baddir2' => q{pg_controldata no pudo leer el directorio de datos: "$1"}, + 'checkpoint-badver' => q{Fallo al ejecutar pg_controldata - probable que la versión sea incorrecta ($1)}, + 'checkpoint-badver2' => q{Fallo al ejecutar pg_controldata - verifique que es la versión correcta}, +'checkpoint-nobin' => q{Could not find a suitable pg_controldata executable}, + 'checkpoint-nodir' => q{Debe especificar el argumento --datadir o definir la variable de ambiente PGDATA}, + 'checkpoint-nodp' => q{Debe instalar el módulo Perl Date::Parse para usar la acción checkpoint}, + 'checkpoint-noparse' => q{No se pudo interpretar la salida de pg_controldata: "$1"}, + 'checkpoint-noregex' => q{No se pudo encontrar la expresion regular para este chequeo}, + 'checkpoint-nosys' => q{No es posible ejecutar pg_controldata: $1}, + 'checkpoint-ok' => q{El último checkpoint fue hace 1 segundo}, + 'checkpoint-ok2' => q{El último checkpoint fue hace $1 segundo}, + 'checkpoint-po' => q{Instante del último checkpoint:}, + 'checksum-msg' => q{checksum: $1}, + 'checksum-nomd' => q{Debe instalar el módulo Perl Digest::MD5 para usar la acción checksum}, + 'checksum-nomrtg' => q{Debe proporcionar un checksum con la opción --mrtg}, + 'custom-invalid' => q{Formato devuelto por la consulta personalizada es inválido}, + 'custom-norows' => q{No se obtuvieron filas}, + 'custom-nostring' => q{Debe proporcionar el texto con la consulta}, + 'database' => q{base de datos}, + 'dbsize-version' => q{La base de datos debe ser version 8.1 o superior para utilizar la acción database_size}, + 'die-action-version' => q{No es posible ejecutar "$1": versión del servidor debe ser >= $2, pero es $3}, + 'die-badtime' => q{El valor de '$1' debe ser un tiempo valido. Ejemplos: -$2 1s -$2 "10 minutes"}, + 'die-badversion' => q{Cadena de versión no válida: $1}, + 'die-noset' => q{No puede ejecutar "$1": la opción $2 no esta activa}, + 'die-nosetting' => q{No fue posible obtener configuración '$1'}, + 'diskspace-fail' => q{Resultado inválido en el comando "$1": $2}, + 'diskspace-msg' => q{FS $1 montado en $2 esta usando $3 de $4 ($5%)}, + 'diskspace-nodata' => q{No fue posible determinar el data_directory: se está conectando como superusuario?}, + 'diskspace-nodf' => q{No se encuentra el ejecutable requerido /bin/df}, + 'diskspace-nodir' => q{No fue posible encontrar el directorio de datos "$1"}, + 'file-noclose' => q{No puedo cerrar $1: $2}, + 'files' => q{archivos}, + 'fsm-page-highver' => q{No se puede comprobar fsm_pages en servidores con versión 8.4 o posterior}, + 'fsm-page-msg' => q{ranuras de páginas FSM utilizadas: $1 de $2 ($3%)}, + 'fsm-rel-highver' => q{No se puede comprobar fsm_relations en servidores con versión 8.4 o posterior}, + 'fsm-rel-msg' => q{relaciones FSM utilizadas: $1 de $2 ($3%)}, + 'hs-future-replica' => q{El esclavo reporta que el reloj del maestro está adelantado, comprobar sincronización de tiempo}, + 'hs-no-role' => q{No es un par maestro/esclavo}, + 'hs-no-location' => q{No pude obtener ubicación actual de xlog en $1}, +'hs-receive-delay' => q{receive-delay}, +'hs-replay-delay' => q{replay_delay}, +'hs-time-delay' => q{time_delay}, + 'hs-time-version' => q{La base de datos debes ser versión 9.1 or superior para comprobar el tiempo de retraso del esclavo}, + 'index' => q{Indice}, + 'invalid-option' => q{Opción inválida}, + 'invalid-query' => q{Se devolvió una consulta inválida: $1}, + 'language' => q{Idioma}, + 'listener-msg' => q{escuchas encontrados: $1}, + 'listening' => q{escuchado}, + 'locks-msg' => q{total "$1" bloqueos: $2}, + 'locks-msg2' => q{total bloqueos: $1}, + 'logfile-bad' => q{Archivo de registro inválido "$1"}, + 'logfile-debug' => q{Archivo de registro final: $1}, + 'logfile-dne' => q{El archivo de registo $1 no existe!}, + 'logfile-fail' => q{Fallo al registrar en: $1}, + 'logfile-ok' => q{registrando en: $1}, + 'logfile-openfail' => q{El archivo de registro "$1" no se pudo abrir: $2}, + 'logfile-opt-bad' => q{Opción de registro inválida}, + 'logfile-seekfail' => q{Fallo al buscar en $1: $2}, + 'logfile-stderr' => q{La salida de registro ha sido redirigida a STDERR: por favor proporcione un nombre de archivo}, + 'logfile-syslog' => q{La base de datos está usando syslog, por favor especifique una ruta con la opción --logfile (fac=$1)}, + 'mode-standby' => q{Servidor en modo "standby"}, + 'mode' => q{modo}, + 'mrtg-fail' => q{La acción $1 fracasó: $2}, + 'new-ver-nocver' => q{No fue posible descargar la información de versión para $1}, + 'new-ver-badver' => q{No fue posible interpretar la información de versión para $1}, + 'new-ver-dev' => q{No es posible comparar versiones de desarrollo: usted tiene $1 versión $2}, + 'new-ver-nolver' => q{No fue posible determinar información de versión local para $1}, + 'new-ver-ok' => q{La versión $1 es la última para $2}, + 'new-ver-warn' => q{Por favor actualice a la versión $1 de $2. Usted esta ejecutando $3}, + 'new-ver-tt' => q{Su versión de $1 ($2) parece ser más nueva que la versión actual! ($3)}, + 'no-db' => q{No hay bases de datos}, + 'no-match-db' => q{No hay bases de datos coincidentes debido a las opciones de exclusión/inclusión}, + 'no-match-fs' => q{No se encuentran sistemas de archivos coincidentes debido a las opciones de exclusión/inclusión}, + 'no-match-rel' => q{No se encuentran relaciones coincidentes debido a las opciones de exclusión/inclusión}, + 'no-match-set' => q{No se encuentran opciones de configuración coincidentes debido a las opciones de exclusión/inclusión}, + 'no-match-table' => q{No se encuentran tablas coincidentes debido a las opciones de exclusión/inclusión}, + 'no-match-user' => q{No se encuentran entradas coincidentes debido a las opciones de exclusión/inclusión}, + 'no-match-slot' => q{No se encuentran ranuras de replicación coincidentes debido a las opciones de exclusión/inclusión}, + 'no-match-slotok' => q{No se encuentran ranuras de replicación}, + 'no-parse-psql' => q{No se pudo interpretar la salida de psql!}, + 'no-time-hires' => q{No se encontró Time::HiRes, necesario si 'showtime' es verdadero}, + 'opt-output-invalid' => q{Formato de salida inválido: debe ser 'nagios' o 'mrtg' o 'simple' o 'cacti'}, + 'opt-psql-badpath' => q{Argumento psql es inválido: debe ser una ruta absoluta a un archivo con nombre psql}, + 'opt-psql-noexec' => q{El archivo "$1" no parece ser ejecutable}, + 'opt-psql-noexist' => q{No se puede encontrar el ejecutable psql: $1}, + 'opt-psql-nofind' => q{No fue posible encontrar un ejecutable psql}, + 'opt-psql-nover' => q{No fue posible determinar la versión de psql}, + 'opt-psql-restrict' => q{No puede usar la opción --PGBINDIR o --PSQL si NO_PSQL_OPTION está habilitado}, + 'pgagent-jobs-ok' => q{No hay trabajos fallidos}, + 'pgbouncer-pool' => q{Pool=$1 $2=$3}, + 'pgb-backends-mrtg' => q{BD=$1 Max conexiones=$2}, + 'pgb-backends-msg' => q{$1 de $2 conexiones ($3%)}, + 'pgb-backends-none' => q{No nay conexiones}, + 'pgb-backends-users' => q{$1 debe ser un número o porcentaje de usuarios}, +'PID' => q{PID}, + 'port' => q{puerto}, + 'preptxn-none' => q{No se encontraron transacciones preparadas}, + 'psa-disabled' => q{No hay consultas - están deshabilitados stats_command_string o track_activities?}, + 'psa-noexact' => q{Error desconocido}, + 'psa-nosuper' => q{No hay coincidencias - por favor ejecute como superusuario}, + 'qtime-count-msg' => q{Total de consultas: $1}, + 'qtime-count-none' => q{no mas que $1 consultas}, + 'qtime-for-msg' => q{$1 consultas más largas que $2s, más larga: $3s$4 $5}, + 'qtime-msg' => q{consulta más larga: $1s$2 $3}, + 'qtime-none' => q{no hay consultas}, + 'query' => q{consulta}, + 'queries' => q{consultas}, +'query-time' => q{query_time}, + 'range-badcs' => q{Opción '$1' inválida: debe ser un "checksum"}, + 'range-badlock' => q{Opción '$1' inválida: debe ser el número de bloqueos, o "type1=#:type2=#"}, + 'range-badpercent' => q{Opción '$1' inválida: debe ser un porcentaje}, + 'range-badpercsize' => q{Opción '$1' inválida: debe ser un tamaño o un porcentaje}, + 'range-badsize' => q{Tamaño inválido para la opción '$1'}, + 'range-badtype' => q{validate_range solicitado con tipo desconocido '$1'}, + 'range-badversion' => q{Cadena inválida para la opción '$1': $2}, + 'range-cactionly' => q{Esta acción es solo para uso en cacti y no acepta parámetros warning o critical}, + 'range-int' => q{Valor inválido para la opción '$1': debe ser un entero}, + 'range-int-pos' => q{Valor inválido para la opción '$1': debe ser un entero positivo}, + 'range-neg-percent' => q{No puede especificar un porcentaje negativo!}, + 'range-none' => q{No se requieren opciones de warning o critical}, + 'range-noopt-both' => q{Debe especificar ambas opciones de 'warning' y 'critical'}, + 'range-noopt-one' => q{Debe especificar una opción 'warning' o 'critical'}, + 'range-noopt-only' => q{Puede especifiar solo una de las opciónes 'warning' o 'critical'}, + 'range-noopt-orboth' => q{Debe especificar un valor de 'warning', de 'critical', o ambos}, + 'range-noopt-size' => q{Debe especificar un tamaño de warning y/o critical}, + 'range-nosize' => q{Debe especificar un tamaño de warning y/o critical}, + 'range-notime' => q{Debe especificar un tamaño de warning y/o critical}, + 'range-seconds' => q{Valor inválido para la opción '$1': debe ser número de segundos}, + 'range-version' => q{debe ser en el formato X.Y o X.Y.Z, donde X es el número mayor de la versión, }, + 'range-warnbig' => q{El valor de la opción 'warning' no puede ser mayor que el de 'critical'}, + 'range-warnbigsize' => q{El valor de la opción 'warning' ($1 bytes) no puede ser mayor que el de 'critical ($2 bytes)'}, + 'range-warnbigtime' => q{El valor de la opción 'warning' ($1 s) no puede ser mayor que el de 'critical ($2 s)'}, + 'range-warnsmall' => q{El valor de la opción 'warning' no puede ser menor que el de 'critical'}, + 'range-nointfortime' => q{Valor inválido para las opciones '$1': debe ser un entero o un valor de tiempo}, + 'relsize-msg-ind' => q{el índice más grande es "$1": $2}, + 'relsize-msg-reli' => q{la relación más grande es el índice "$1": $2}, + 'relsize-msg-relt' => q{la relación más grande es la tabla "$1": $2}, + 'relsize-msg-tab' => q{la tabla más grande es "$1": $2}, +'relsize-msg-indexes' => q{table with largest indexes is "$1": $2}, + 'rep-badarg' => q{Parámetro invalido para repinfo: se esperan 6 valores separados por coma}, + 'rep-duh' => q{No tiene sentido probar la replicación con los mismos valores}, + 'rep-fail' => q{Fila no replicada en el esclavo $1}, + 'rep-noarg' => q{Necesita un valor para repinfo}, + 'rep-norow' => q{La fila del origen de la replicación no fue encontrada: $1}, + 'rep-noslaves' => q{No se encontraron esclavos}, + 'rep-notsame' => q{No se puede probar la replicación: los valores no coinciden}, + 'rep-ok' => q{La fila fue replicada}, + 'rep-sourcefail' => q{Falló la actualización del origen}, + 'rep-timeout' => q{La fila no fue replicada. Timeout: $1}, + 'rep-unknown' => q{Chequeo de replicación fallido}, + 'rep-wrongvals' => q{No puedo verificar la replicación: los valores no son correctos ('$1' no '$2' ni '$3')}, + 'repslot-version' => q{La base de datos debe ser version 9.4 o superior para ver las ranuras de replicación}, + 'runcommand-err' => q{Error desconocido en la función "run_command"}, + 'runcommand-nodb' => q{No se encontró ninguna base de datos buscada}, + 'runcommand-nodupe' => q{No fue posible duplicar STDERR}, + 'runcommand-noerr' => q{No fue posible abrir STDERR?!}, + 'runcommand-nosys' => q{Llamada al sistema falló con $1}, + 'runcommand-pgpass' => q{Se creo archivo temporal pgpass $1}, + 'runcommand-timeout' => q{Tiempo de espera del comando agotado! Considere incrementar --timeout a valor mayor que $1}, + 'runtime-badmrtg' => q{valor de queryname inválido?}, + 'runtime-badname' => q{Opción queryname inválida: debe ser el nombre de una única vista}, + 'runtime-msg' => q{tiempo de la consulta: $1 seg}, + 'schema' => q{Esquema}, +'ss-badobject' => q{Unrecognized object types: $1}, + 'ss-createfile' => q{Archivo creado $1}, + 'ss-different' => q{"$1" es diferente:}, + 'ss-existson' => q{Existe en:}, + 'ss-failed' => q{Las bases de datos son diferentes. Elementos no coincidentes: $1}, + 'ss-matched' => q{Todas las bases tienen los mismos elementos}, + 'ss-missingon' => q{Desaparecido en:}, + 'ss-noexist' => q{$1 "$2" no existe en todas las bases de datos:}, + 'ss-notset' => q{"$1" no está definido en todas las bases de datos:}, + 'ss-suffix' => q{Error: no puede usar sufijos a menos que este buscando esquemas basados en tiempo}, + 'seq-die' => q{No es posible obtener información sobre la secuencia $1}, + 'seq-msg' => q{$1=$2% (llamadas pendientes=$3)}, + 'seq-none' => q{No se encontraron secuencias}, + 'size' => q{tamaño}, + 'slony-noschema' => q{No fue posible determinar el esquema para Slony}, + 'slony-nonumber' => q{El llamado a sl_status no devolvió un número}, +'slony-lagtime' => q{Slony lag time: $1}, + 'symlink-create' => q{Creado "$1"}, + 'symlink-done' => q{No se crea "$1": El enlace $2 ya apunta a "$3"}, + 'symlink-exists' => q{No se crea "$1": El archivo $2 ya existe}, + 'symlink-fail1' => q{No fue posible desvincular "$1": $2}, + 'symlink-fail2' => q{No pude crear enlace simbolico de $1 a $2: $3}, + 'symlink-name' => q{Este comando no va a funcionar a menos que el programa tenga la palabra "postgres" en su nombre}, + 'symlink-unlink' => q{Desvinculando "$1":$2 }, + 'table' => q{Tabla}, +'testmode-end' => q{END OF TEST MODE}, + 'testmode-fail' => q{Connexión fallida: $1 $2}, + 'testmode-norun' => q{No puede ejecutar "$1" en $2: versión debe ser >= $3, pero es $4}, + 'testmode-noset' => q{No puede ejecutar "$1" en $2: la opción $3 no está activa}, +'testmode-nover' => q{Could not find version for $1}, +'testmode-ok' => q{Connection ok: $1}, +'testmode-start' => q{BEGIN TEST MODE}, + 'time-day' => q{día}, + 'time-days' => q{días}, + 'time-hour' => q{hora}, + 'time-hours' => q{horas}, + 'time-minute' => q{minuto}, + 'time-minutes' => q{minutos}, + 'time-month' => q{mes}, + 'time-months' => q{meses}, + 'time-second' => q{segundo}, + 'time-seconds' => q{segundos}, + 'time-week' => q{semana}, + 'time-weeks' => q{semanas}, + 'time-year' => q{año}, + 'time-years' => q{años}, +'timesync-diff' => q{diff}, + 'timesync-msg' => q{timediff=$1 BD=$2 Local=$3}, + 'transactions' => q{transacciones}, + 'trigger-msg' => q{Disparadores desactivados: $1}, +'txn-time' => q{transaction_time}, + 'txnidle-count-msg' => q{Total transacciones inactivas: $1}, + 'txnidle-count-none' => q{no más de $1 transacciones inactivas}, + 'txnidle-for-msg' => q{$1 transacciones inactivas mayores que $2s, más larga: $3s$4 $5}, + 'txnidle-msg' => q{transacción inactiva más larga: $1s$2 $3}, + 'txnidle-none' => q{no hay transacciones inactivas}, + 'txntime-count-msg' => q{Total de transacciones: $1}, + 'txntime-count-none' => q{no más que $1 transacciones}, + 'txntime-for-msg' => q{$1 transacciones más largas que $2s, más larga: $3s$4 $5}, + 'txntime-msg' => q{transacción más larga: $1s$2 $3}, + 'txntime-none' => q{No hay transacciones}, + 'txnwrap-cbig' => q{El valor 'critical' debe ser menor a 2000 millones}, + 'txnwrap-wbig' => q{El valor 'warning' debe ser menor a 2000 millones}, + 'unknown-error' => q{Error desconocido}, + 'usage' => qq{\nUso: \$1 <opciones>\n Pruebe "\$1 --help" para obtener lista completa de opciones\n Pruebe "\$1 --man" para acceder al manual completo\n}, + 'user' => q{Usuario}, + 'username' => q{nombre-de-usuario}, + 'vac-nomatch-a' => q{No hay tablas coincidentes que hayan sido analizadas}, + 'vac-nomatch-v' => q{No hay tablas coincidentes que hayan sido vaciadas}, + 'version' => q{versión $1}, + 'version-badmrtg' => q{Argumento de versión mrtg inválido}, + 'version-fail' => q{versión $1, pero se esperaba $2}, + 'version-ok' => q{versión $1}, + 'wal-numfound' => q{Archivos WAL encontrados: $1}, + 'wal-numfound2' => q{Archivos WAL "$2" encontrados: $1}, +}, + +## French 'fr' => { 'address' => q{adresse}, 'age' => q{âge}, @@ -367,6 +649,7 @@ package check_postgres; 'checkpoint-baddir2' => q{pg_controldata n'a pas pu lire le répertoire des données indiqué : « $1 »}, 'checkpoint-badver' => q{Échec lors de l'exécution de pg_controldata - probablement la mauvaise version ($1)}, 'checkpoint-badver2' => q{Échec lors de l'exécution de pg_controldata - est-ce la bonne version ?}, +'checkpoint-nobin' => q{Could not find a suitable pg_controldata executable}, 'checkpoint-nodir' => q{Vous devez fournir un argument --datadir ou configurer la variable d'environnement PGDATA}, 'checkpoint-nodp' => q{Vous devez installer le module Perl Date::Parse pour utiliser l'action checkpoint}, 'checkpoint-noparse' => q{Incapable d'analyser le résultat de la commande pg_controldata : "$1"}, @@ -383,7 +666,6 @@ package check_postgres; 'custom-nostring' => q{Vous devez fournir une requête}, 'database' => q{base de données}, 'dbsize-version' => q{La base de données cible doit être une version 8.1 ou ultérieure pour exécuter l'action database_size}, -'depr-pgcontroldata' => q{PGCONTROLDATA is deprecated, use PGBINDIR instead.}, 'die-action-version' => q{Ne peut pas exécuter « $1 » : la version du serveur doit être supérieure ou égale à $2, alors qu'elle est $3}, 'die-badtime' => q{La valeur de « $1 » doit être une heure valide. Par exemple, -$2 1s -$2 « 10 minutes »}, 'die-badversion' => q{Version invalide : $1}, @@ -400,10 +682,13 @@ package check_postgres; 'fsm-page-msg' => q{emplacements de pages utilisés par la FSM : $1 sur $2 ($3%)}, 'fsm-rel-highver' => q{Ne peut pas vérifier fsm_relations sur des serveurs en version 8.4 ou ultérieure}, 'fsm-rel-msg' => q{relations tracées par la FSM : $1 sur $2 ($3%)}, +'hs-future-replica' => q{Slave reporting master server clock is ahead, check time sync}, 'hs-no-role' => q{Pas de couple ma??tre/esclave}, 'hs-no-location' => q{N'a pas pu obtenir l'emplacement courant dans le journal des transactions sur $1}, 'hs-receive-delay' => q{délai de réception}, 'hs-replay-delay' => q{délai de rejeu}, +'hs-time-delay' => q{time_delay}, +'hs-time-version' => q{Database must be version 9.1 or higher to check slave lag by time}, 'index' => q{Index}, 'invalid-option' => q{Option invalide}, 'invalid-query' => q{Une requête invalide a renvoyé : $1}, @@ -439,6 +724,8 @@ package check_postgres; 'no-match-set' => q{Aucun paramètre trouvé à cause des options d'exclusion/inclusion}, 'no-match-table' => q{Aucune table trouvée à cause des options d'exclusion/inclusion}, 'no-match-user' => q{Aucune entrée trouvée à cause options d'exclusion/inclusion}, + 'no-match-slot' => q{Aucune fentes de réplication trouvée à cause options d'exclusion/inclusion}, + 'no-match-slotok' => q{Pas de fentes de réplication trouvé}, 'no-parse-psql' => q{N'a pas pu analyser la sortie de psql !}, 'no-time-hires' => q{N'a pas trouvé le module Time::HiRes, nécessaire quand « showtime » est activé}, 'opt-output-invalid' => q{Sortie invalide : doit être 'nagios' ou 'mrtg' ou 'simple' ou 'cacti'}, @@ -498,6 +785,7 @@ package check_postgres; 'relsize-msg-reli' => q{la plus grosse relation est l'index « $1 » : $2}, 'relsize-msg-relt' => q{la plus grosse relation est la table « $1 » : $2}, 'relsize-msg-tab' => q{la plus grosse table est « $1 » : $2}, + 'relsize-msg-indexes' => q{la table avec les index les plus volumineux est « $1 » : $2}, 'rep-badarg' => q{Argument repinfo invalide : 6 valeurs séparées par des virgules attendues}, 'rep-duh' => q{Aucun sens à tester la réplication avec les mêmes valeurs}, 'rep-fail' => q{Ligne non répliquée sur l'esclave $1}, @@ -509,7 +797,8 @@ package check_postgres; 'rep-sourcefail' => q{Échec de la mise à jour de la source}, 'rep-timeout' => q{La ligne n'a pas été répliquée. Délai dépassé : $1}, 'rep-unknown' => q{Échec du test de la réplication}, - 'rep-wrongvals' => q{Ne peut pas tester la réplication : les valeurs ne sont pas les bonnes (ni $1 ni $2 ni $3)}, + 'rep-wrongvals' => q{Ne peut pas tester la réplication : les valeurs ne sont pas les bonnes (ni '$1' ni '$2' ni '$3')}, + 'repslot-version' => q{Base de données doit être la version 9.4 ou ultérieure pour vérifier fentes de réplication}, 'runcommand-err' => q{Erreur inconnue de la fonction « run_command »}, 'runcommand-nodb' => q{Aucune base de données cible trouvée}, 'runcommand-nodupe' => q{N'a pas pu dupliqué STDERR}, @@ -521,6 +810,7 @@ package check_postgres; 'runtime-badname' => q{Option invalide pour queryname option : doit être le nom d'une vue}, 'runtime-msg' => q{durée d'exécution de la requête : $1 secondes}, 'schema' => q{Schéma}, +'ss-badobject' => q{Unrecognized object types: $1}, 'ss-createfile' => q{Création du fichier $1}, 'ss-different' => q{"$1" est différent:}, 'ss-existson' => q{Existe sur :}, @@ -596,82 +886,374 @@ package check_postgres; 'wal-numfound' => q{Fichiers WAL trouvés : $1}, 'wal-numfound2' => q{Fichiers WAL "$2" trouvés : $1}, }, -'af' => { -}, + +## Czech 'cs' => { 'checkpoint-po' => q{�as posledn�ho kontroln�ho bodu:}, }, + +## German 'de' => { - 'backends-po' => q{tut mir leid, schon zu viele Verbindungen}, - 'checkpoint-po' => q{Zeit des letzten Checkpoints:}, -}, -'es' => { - 'backends-po' => q{lo siento, ya tenemos demasiados clientes}, - 'checkpoint-po' => q{Instante de �ltimo checkpoint:}, + 'address' => q{Adresse}, + 'age' => q{Alter}, + 'backends-fatal' => q{Kann nicht verbinden: zu viele Verbindungen}, + 'backends-mrtg' => q{DB=$1 Max. Anzahl Verbindungen (max_connections)=$2}, + 'backends-msg' => q{$1 of $2 Verbindungen ($3%)}, + 'backends-nomax' => q{Kann Wert für max_connections nicht bestimmen}, + 'backends-oknone' => q{Keine Verbindungen}, + 'backends-po' => q{Tut mir leid, schon zu viele Clients}, + 'backends-users' => q{$1 als Anzahl Benutzer muss eine Zahl oder ein Prozentwert sein}, + 'bloat-index' => q{(db $1) Index $2 Zeilen:$3 Seiten:$4 sollte sein:$5 ($6X) verschwendete Bytes:$7 ($8)}, + 'bloat-nomin' => q{Keine Relation entsprichten den Kriterien für minimalen Bloat}, + 'bloat-table' => q{(db $1) Tabelle $2.$3 Zeilen:$4 Seiten:$5 sollte sein:$6 ($7X) verschwendete Größe:$8 ($9)}, + 'bug-report' => q{Bitte berichte über diese Details an check_postgres@bucardo.org:}, + 'checkcluster-id' => q{Datenbank-Systembezeichner:}, + 'checkcluster-msg' => q{cluster_id: $1}, + 'checkcluster-nomrtg'=> q{Es muss eine Zahl per Option --mrtg mitgegeben werden}, + 'checkmode-prod' => q{in Produktion}, + 'checkmode-recovery' => q{in der Wiederherstellung aus dem Archiv}, + 'checkmode-state' => q{Zustand des Datenbank-Clusters:}, + 'checkpoint-baddir' => q{Ungültiges Datenverzeichnis (data_directory): "$1"}, + 'checkpoint-baddir2' => q{pg_controldata konnte das angebene Verzeichnis lesen: "$1"}, + 'checkpoint-badver' => q{Kann pg_controldata nicht starten - vielleicht die falsche Version ($1)}, + 'checkpoint-badver2' => q{Fehler beim Start von pg_controldata - ist es die richtige Version?}, +'checkpoint-nobin' => q{Could not find a suitable pg_controldata executable}, + 'checkpoint-nodir' => q{Entweder muss die Option --datadir als Argument angegebn werden, oder die Umgebungsvariable PGDATA muss gesetzt sein}, + 'checkpoint-nodp' => q{Das Perl-Modul Date::Parse muss installiert sein für die Verwendung der checkpoint-Aktion}, + 'checkpoint-noparse' => q{Kann die Ausgabe von pg_controldata nicht lesen: "$1"}, + 'checkpoint-noregex' => q{Kann den regulären Ausdruck für diese Prüfung nicht finden}, + 'checkpoint-nosys' => q{Konnte pg_controldata nicht aufrufen: $1}, + 'checkpoint-ok' => q{Letzter Checkpoint war vor 1 Sekunde}, + 'checkpoint-ok2' => q{Letzter Checkpoint war von $1 Sekunden}, + 'checkpoint-po' => q{Zeit des letzten Checkpoints:}, + 'checksum-msg' => q{Prüfsumme: $1}, + 'checksum-nomd' => q{Das Perl-Modul Digest::MD5 muss installiert sein für die Verwendung der checksum-Aktion}, + 'checksum-nomrtg' => q{Es muss eine Prüfsummer per Option --mrtg mitgegeben werden}, + 'custom-invalid' => q{Ungültiges Format von der benutzerdefinierten Abfrage zurückgeliefert}, + 'custom-norows' => q{Keine Zeilen erhalten}, + 'custom-nostring' => q{Es muss eine Abfrage (Query) mitgegeben werde}, + 'database' => q{Datenbank}, + 'dbsize-version' => q{Die Zieldatenbank muss Version 8.1 oder höher sein für die Verwendung der database_size-Aktion}, + 'die-action-version' => q{Kann "$1" nicht starten: Die Serverversion muss >= $2 sein, ist aber $3}, + 'die-badtime' => q{Wert für '$1' muss eine gültige Zeit sein. Beispiele: -$2 1s -$2 "10 minutes"}, + 'die-badversion' => q{Ungültiger Versionsstring: $1}, + 'die-noset' => q{Kann "$1" nicht starten, denn $2 ist nicht eingeschaltet (on)}, + 'die-nosetting' => q{Kann die Einstellung '$1' nicht lesen}, + 'diskspace-fail' => q{Ungültiges Ergebnis des Kommandos "$1": $2}, + 'diskspace-msg' => q{Dateisystem $1 gemountet auf $2 verwendet $3 von $4 ($5%)}, + 'diskspace-nodata' => q{Kann das Datenverzeichnis (data_directory) nicht bestimmen: Bist du als superuser verbunden?}, + 'diskspace-nodf' => q{Kann die nötige ausführbare Datei nicht finden /bin/df}, + 'diskspace-nodir' => q{Kann das Datenverzeichnis "$1" nicht finden}, + 'file-noclose' => q{Kann $1 nicht schließen $1: $2}, + 'files' => q{Dateien}, + 'fsm-page-highver' => q{Kann fsm_pages auf Servern mit Version 8.4 oder höher nicht prüfen}, + 'fsm-page-msg' => q{Anzahl benutzter fsm-Seiten: $1 von $2 ($3%)}, + 'fsm-rel-highver' => q{Kann fsm_relations auf Servern mit Version 8.4 oder höher nicht prüfen}, + 'fsm-rel-msg' => q{Anzahl benutzter fsm-Relationen: $1 von $2 ($3%)}, + 'hs-future-replica' => q{Der Slave-Server berichtet, dass die Uhr des Masters vorgeht, Zeitsynchronisation prüfen}, + 'hs-no-role' => q{Dies ist kein Master/Slave-Paar}, + 'hs-no-location' => q{Kann die aktuelle xlog-Position (WAL) auf $1 nicht bestimmen}, + 'hs-receive-delay' => q{Empfangsverzögerung}, + 'hs-replay-delay' => q{Wiederherstellungsverzögerung}, + 'hs-time-delay' => q{Zeitverzug}, + 'hs-time-version' => q{Datenbank muss Version 9.1 oder höher sein um die Verzögerung des Slaves in Zeit anzugeben}, + 'index' => q{Index}, + 'invalid-option' => q{Ungültige Option}, + 'invalid-query' => q{Ungültige Abfrage geliefert: $1}, + 'language' => q{Sprache}, + 'listener-msg' => q{Gefundene Lauscher: $1}, + 'listening' => q{lausche}, + 'locks-msg' => q{Insgesamt "$1" Sperren: $2}, + 'locks-msg2' => q{Sperren insgesamt: $1}, + 'logfile-bad' => q{Ungültige Log-Datei "$1"}, + 'logfile-debug' => q{Letzte Log-Datei: $1}, + 'logfile-dne' => q{Log-Datei $1 existiert nicht!}, + 'logfile-fail' => q{Kann nicht nach $1 loggen}, + 'logfile-ok' => q{logge nach: $1}, + 'logfile-openfail' => q{Kann Log-Datei "$1" nicht öffnen: $2}, + 'logfile-opt-bad' => q{Ungültige Log-Datei-Option}, + 'logfile-seekfail' => q{Positionieren in Datei $1 fehlgeschlagen: $2}, + 'logfile-stderr' => q{Log-Ausgabe wurde umgelenkt auf stderr: bitte einen Dateinamen angeben}, + 'logfile-syslog' => q{Datenbank verwendet syslog, bitte einen Pfad angeben mit der Option --logfile (fac=$1)}, + 'mode-standby' => q{Server im Standby-Modus}, + 'mode' => q{Modus}, + 'mrtg-fail' => q{Aktion $1 fehlgeschlagen: $2}, + 'new-ver-nocver' => q{Konnte die Versionsangabe für $1 nicht downloaden}, + 'new-ver-badver' => q{Konnte die Versionsangabe für $1 nicht verstehen}, + 'new-ver-dev' => q{Kann auf Entwicklungsversionen keinen Versionsvergleich durchführen: Du hast $1 Version $2}, + 'new-ver-nolver' => q{Konnte die lokale Versionsangabe für $1 nicht bestimmen}, + 'new-ver-ok' => q{Version $1 ist die letzte für $2}, + 'new-ver-warn' => q{Bitte aktualisieren auf $1 von $2. Derzeit läuft $3}, + 'new-ver-tt' => q{Deine Datenbank der Version $1 ($2) scheint der aktuellen Version vorauszugehen! ($3)}, + 'no-db' => q{Keine Datenbanken}, + 'no-match-db' => q{Keine passende Datenbank gefunden gemäß den Ausschluss-/Einschluss-Optionen}, + 'no-match-fs' => q{Keine passenden Dateisysteme gefunden gemäß den Ausschluss-/Einschluss-Optionen}, + 'no-match-rel' => q{Keine passenden Relationen gefunden gemäß den Ausschluss-/Einschluss-Optionen}, + 'no-match-set' => q{Keine passenden Einstellungen gefunden gemäß den Ausschluss-/Einschluss-Optionen}, + 'no-match-table' => q{Keine passenden Tabellen gefunden gemäß den Ausschluss-/Einschluss-Optionen}, + 'no-match-user' => q{Keine passenden Einträge gefunden gemäß den Ausschluss-/Einschluss-Optionen}, + 'no-match-slot' => q{Keine passenden Replikationen gefunden gemäß den Ausschluss-/Einschluss-Optionen}, + 'no-match-slotok' => q{Keine passenden Replikations-Slots gefunden gemäß den Ausschluss-/Einschluss-Optionen}, + 'no-parse-psql' => q{Konnte die Ausgabe von psql nicht verstehen!}, + 'no-time-hires' => q{Kann Time::HiRes nicht finden, ist aber nötig wenn 'showtime' auf 'wahr' gesetzt ist (true)}, + 'opt-output-invalid' => q{Ungültige Ausgabe: Muss eines sein von 'nagios' oder 'mrtg' oder 'simple' oder 'cacti'}, + 'opt-psql-badpath' => q{Ungültiges Argument für psql: Muss ein vollständiger Pfad zu einer Datei namens psql}, + 'opt-psql-noexec' => q{Die Datei "$1" scheint nicht ausführbar zu sein}, + 'opt-psql-noexist' => q{Kann angegebene ausführbare Datei psql nicht finden: $1}, + 'opt-psql-nofind' => q{Konnte keine geeignete ausführbare Datei psql finden}, + 'opt-psql-nover' => q{Konnte die Version von psql nicht bestimmen}, + 'opt-psql-restrict' => q{Kann die Optionen --PGBINDIR und --PSQL nicht verwenden, wenn NO_PSQL_OPTION eingeschaltet ist (on)}, + 'pgagent-jobs-ok' => q{Keine fehlgeschlagenen Jobs}, + 'pgbouncer-pool' => q{Pool=$1 $2=$3}, + 'pgb-backends-mrtg' => q{DB=$1 max. Anzahl Verbindungen=$2}, + 'pgb-backends-msg' => q{$1 von $2 Verbindungen ($3%)}, + 'pgb-backends-none' => q{Keine Verbindungen}, + 'pgb-backends-users' => q{$1 für die Anzahl Benutzer muss eine Zahl oder ein Prozentwert sein}, + 'PID' => q{PID}, + 'port' => q{Port}, + 'preptxn-none' => q{Keine prepared Transactions gefunden}, + 'psa-disabled' => q{Keine Anfragen - ist stats_command_string oder track_activities ausgeschaltet?}, + 'psa-noexact' => q{Unbekannter Fehler}, + 'psa-nosuper' => q{Keine Treffer - Bitte als superuser ausführen}, + 'qtime-count-msg' => q{Gesamtanzahl Abfragen: $1}, + 'qtime-count-none' => q{Nicht mehr als $1 Abfragen}, + 'qtime-for-msg' => q{$1 Abfragen länger als $2s, längste: $3s$4 $5}, + 'qtime-msg' => q{längste Abfrage: $1s$2 $3}, + 'qtime-none' => q{Keine Abfragen}, + 'query' => q{Abfrage}, + 'queries' => q{Abfragen}, + 'query-time' => q{Abfragezeit (query_time)}, + 'range-badcs' => q{Ungültige Option '$1': Es muss eine Prüfsumme sein}, + 'range-badlock' => q{Ungültige Option '$1': Es muss eine Anzahl Sperren oder "type1=#:type2=#" sein}, + 'range-badpercent' => q{Ungültige Option '$1': Es muss ein Prozentwert sein}, + 'range-badpercsize' => q{Ungültige Option '$1': Es muss eine Größe oder ein Prozentwert sein}, + 'range-badsize' => q{Ungültige Größe für die Option '$1'}, + 'range-badtype' => q{validate_range wurde mit unbekanntem Typen '$1' aufgerufen}, + 'range-badversion' => q{Ungültige Zeichenkette '$2' für die Option '$1'}, + 'range-cactionly' => q{Diese Aktion ist nur für die Benutzung mit cacti und kennt keine warning oder critical Argumente}, + 'range-int' => q{Ungültiges Argument für die Option '$1': Muss eine Ganzzahl sein}, + 'range-int-pos' => q{Ungültiges Argument für die Option '$1': Muss eine natürliche Zahl sein}, + 'range-neg-percent' => q{Negativer Prozentwert ist nicht zulässig!}, + 'range-none' => q{Es werden keine Optionen für warning oder critical benötigt}, + 'range-noopt-both' => q{Sowohl die Option 'warning' als auch 'critical' ist nötig}, + 'range-noopt-one' => q{Es muss eine Option 'warning' oder 'critical' angegeben werden}, + 'range-noopt-only' => q{Es darf nur eine Option 'warning' ODER 'critical' angegeben werden}, + 'range-noopt-orboth' => q{Es muss eine Option 'warning', 'critical' oder beide angegeben werden}, + 'range-noopt-size' => q{Es muss eine Größe für warning und/oder critical angegeben werden}, + 'range-nosize' => q{Es muss eine Größe für warning und/oder critical angegeben werden}, + 'range-notime' => q{Es muss eine Zeit für warning und/oder critical angegeben werden}, + 'range-seconds' => q{Ungültiges Argument für die Option '$1': Es muss eine Sekundenanzahl sein}, + 'range-version' => q{Muss im Format X.Y oder X.Y.Z sein, wobei X die Hauptversionsnummer ist, }, + 'range-warnbig' => q{Der Wert für die Option 'warning' kann nicht größer sein als der für 'critical'}, + 'range-warnbigsize' => q{Der Wert für die Option 'warning' ($1 Bytes) kann nicht größer sein als der für 'critical' ($2 Bytes)}, + 'range-warnbigtime' => q{Der Wert für die Option 'warning' ($1 s) kann nicht größer sein als der für 'critical' ($2 s)}, + 'range-warnsmall' => q{Der Wert für die Option 'warning' darf nicht kleiner sein als der für 'critical'}, + 'range-nointfortime' => q{Ungültiges Argument für die Option '$1': Es muss eine ganze Zahl, eine Zeit oder eine Sekundenanzahl sein}, + 'relsize-msg-ind' => q{Größter Index ist "$1": $2}, + 'relsize-msg-reli' => q{Größte Relation ist der Index "$1": $2}, + 'relsize-msg-relt' => q{Größte Relation ist die Tabelle "$1": $2}, + 'relsize-msg-tab' => q{Größte Tabelle ist "$1": $2}, + 'relsize-msg-indexes' => q{Tabelle mit den größten Indexen ist "$1": $2}, + 'rep-badarg' => q{Ungültiges Argument für repinfo: Es werden 6 durch Komma getrennte Werte erwartet}, + 'rep-duh' => q{Es hat keinen Sinn, die Replikation mit denselben Werten zu testen}, + 'rep-fail' => q{Zeile nicht auf Slave $1 repliziert}, + 'rep-noarg' => q{Benötige ein Argument für repinfo}, + 'rep-norow' => q{Zeile für die Replikation nicht gefunden: $1}, + 'rep-noslaves' => q{Keine Slaves gefunden}, + 'rep-notsame' => q{Kann Replikation nicht testen: Werte stimmen nicht überein}, + 'rep-ok' => q{Zeile wurde repliziert}, + 'rep-sourcefail' => q{Aktualisierung der Quelle fehlgeschlagen}, + 'rep-timeout' => q{Zeile wurde nicht repliziet. Zeitüberschreitung: $1}, + 'rep-unknown' => q{Überprüfung der Replikation fehlgeschlagen}, + 'rep-wrongvals' => q{Kann Replikation nicht testen: Werte stimmen nicht ('$1' weder '$2' noch '$3')}, + 'repslot-version' => q{Datenbank muss in Version 9.4 oder höher vorliegen für die Prüfung von Replikationsslots}, + 'runcommand-err' => q{Unbekannte Fehler innerhalb der Funktion "run_command"}, + 'runcommand-nodb' => q{Keine Zieldatenbanken gefunden}, + 'runcommand-nodupe' => q{Konnte STDERR nicht duplizieren}, + 'runcommand-noerr' => q{Konnte STDERR nicht öffnen?!}, + 'runcommand-nosys' => q{Systemaufruf fehlgeschlagen mit $1}, + 'runcommand-pgpass' => q{Temporäre pgpass-Datei $1 erzeugt}, + 'runcommand-timeout' => q{Zeitüberschreitung bei Kommand! Vielleicht sollte --timeout höher als $1 eingestellt werden}, + 'runtime-badmrtg' => q{Ungültiger Abfragename?}, + 'runtime-badname' => q{Ungültige Option queryname: Es muss ein einfacher Name einer Sicht (View) sein}, + 'runtime-msg' => q{Laufzeit der Abfrage: $1 Sekunden}, + 'schema' => q{Schema}, + 'ss-createfile' => q{Habe Datei $1 erzeugt}, + 'ss-different' => q{"$1" unterscheidet sich:}, + 'ss-existson' => q{Existiert auf:}, + 'ss-failed' => q{Datenbanken waren verschiden. Nicht übereinstimmend: $1}, + 'ss-matched' => q{Alle Datenbanken enthalten dasselbe}, + 'ss-missingon' => q{Fehlt in:}, + 'ss-noexist' => q{$1 "$2" existiert nicht in allen Datenbanken:}, + 'ss-notset' => q{"$1" ist nicht auf allen Datenbanken eingestellt:}, + 'ss-suffix' => q{Fehler: Kann das Suffix nicht verwenden, sofern keine zeitbasierten Schemas verwendet werden}, + 'seq-die' => q{Kann keine Information über die Sequenz $1 finden}, + 'seq-msg' => q{$1=$2% (Aufrufe übrig=$3)}, + 'seq-none' => q{Keine Sequenzen gefunden}, + 'size' => q{Größe}, + 'slony-noschema' => q{Konnte das Schema für Slony nicht finden}, + 'slony-nonumber' => q{Aufruf von sl_status hat keine Zahl zurückgeliefert}, + 'slony-lagtime' => q{Verzögerung von Slony: $1}, + 'symlink-create' => q{Habe "$1" erzeugt}, + 'symlink-done' => q{Erzeuge "$1" nicht: $2 ist bereits verlinkt mit "$3"}, + 'symlink-exists' => q{Erzeuge "$1" nicht: Datei $2 existiert bereits}, + 'symlink-fail1' => q{Kann Verlinkung nicht lösen (unlink) "$1": $2}, + 'symlink-fail2' => q{Kann symbolischen Link $1 auf $2 nicht erzeugen: $3}, + 'symlink-name' => q{Dieses Kommando wird nicht funktionieren, sofern das Wort "postgres" nicht im Namen enthalten ist}, + 'symlink-unlink' => q{Löse Verlinkung "$1":$2 }, + 'table' => q{Tabelle}, + 'testmode-end' => q{ENDE DES TEST-MODUS}, + 'testmode-fail' => q{Verbindung fehlgeschlagen: $1 $2}, + 'testmode-norun' => q{Kann "$1" nicht auf $2 laufen lassen: Version muss >= $3 sein, ist aber $4}, + 'testmode-noset' => q{Kann "$1" nicht auf $2 laufen lassen: $3 ist nicht eingeschaltet (on)}, + 'testmode-nover' => q{Kann Versionsinformation für $1 nicht finden}, + 'testmode-ok' => q{Verbindung OK: $1}, + 'testmode-start' => q{BEGINN DES TEST-MODUS}, + 'time-day' => q{Tag}, + 'time-days' => q{Tage}, + 'time-hour' => q{Stunde}, + 'time-hours' => q{Stunden}, + 'time-minute' => q{Minute}, + 'time-minutes' => q{Minuten}, + 'time-month' => q{Monat}, + 'time-months' => q{Monate}, + 'time-second' => q{Sekunde}, + 'time-seconds' => q{Sekunden}, + 'time-week' => q{Woche}, + 'time-weeks' => q{Wochen}, + 'time-year' => q{Jahr}, + 'time-years' => q{Jahre}, + 'timesync-diff' => q{Differenz}, + 'timesync-msg' => q{Zeitdifferenz=$1 DB=$2 lokal=$3}, + 'transactions' => q{Transactionen}, + 'trigger-msg' => q{Deaktivierte Trigger: $1}, + 'txn-time' => q{Tranaktionszeit (transaktion_time)}, + 'txnidle-count-msg' => q{Insgesamt untätig (idle) in Transaktion: $1}, + 'txnidle-count-none' => q{Nicht mehr als $1 untätig (idle) in Transaktion}, + 'txnidle-for-msg' => q{$1 untätige (idle) Transactionen länger als $2s, längste: $3s$4 $5}, + 'txnidle-msg' => q{Längste untätige Transaktion (idle): $1s$2 $3}, + 'txnidle-none' => q{Keine Transaktionen untätig (idle)}, + 'txntime-count-msg' => q{Transaktionen insgesamt: $1}, + 'txntime-count-none' => q{Nicht mehr als $1 Transaktionen}, + 'txntime-for-msg' => q{$1 Transaktionen länger als $2s, längste: $3s$4 $5}, + 'txntime-msg' => q{Längste Transaktion: $1s$2 $3}, + 'txntime-none' => q{Keine Transaktionen}, + 'txnwrap-cbig' => q{Der Wert für 'critical' muss unter 2 Billionen liegen}, + 'txnwrap-wbig' => q{Der Wert für 'warning' muss unter 2 Billionen liegen}, + 'unknown-error' => q{Unbekannter Fehler}, + 'usage' => qq{\nAnwendung: \$1 <Optionen>\n Versuche "\$1 --help" für eine komplette Liste der Optionen\n Versuche "\$1 --man" für ein komplettes Handbuch\n}, + 'user' => q{Benutzer}, + 'username' => q{Benutzernname}, + 'vac-nomatch-a' => q{Keine passenden Tabellen wurden jemals analyisiert}, + 'vac-nomatch-v' => q{Keine passenden Tabellen wurden jemals vakuumiert}, + 'version' => q{Version $1}, + 'version-badmrtg' => q{Ungültiges Argument für die mrtg Version}, + 'version-fail' => q{Version $1, aber es wurde $2 erwartet}, + 'version-ok' => q{Version $1}, + 'wal-numfound' => q{WAL-Dateien gefunden: $1}, + 'wal-numfound2' => q{WAL "$2" Dateien gefunden: $1}, }, + +## Persian 'fa' => { 'checkpoint-po' => q{زمان آخرین وارسی:}, }, + +## Croation 'hr' => { 'backends-po' => q{nažalost, već je otvoreno previše klijentskih veza}, }, + +## Hungarian 'hu' => { 'checkpoint-po' => q{A legut�bbi ellen�rz�pont ideje:}, }, + +## Italian 'it' => { 'checkpoint-po' => q{Orario ultimo checkpoint:}, }, + +## Japanese 'ja' => { 'backends-po' => q{現在クライアント数が多すぎます}, 'checkpoint-po' => q{最終チェックポイント時刻:}, }, + +## Korean 'ko' => { 'backends-po' => q{최대 동시 접속자 수를 초과했습니다.}, 'checkpoint-po' => q{������ üũ����Ʈ �ð�:}, }, + +## Norwegian bokmål 'nb' => { 'backends-po' => q{beklager, for mange klienter}, 'checkpoint-po' => q{Tidspunkt for nyeste kontrollpunkt:}, }, + +## Dutch 'nl' => { }, + +## Polish 'pl' => { 'checkpoint-po' => q{Czas najnowszego punktu kontrolnego:}, }, + +## Portuguese 'pt_BR' => { 'backends-po' => q{desculpe, muitos clientes conectados}, 'checkpoint-po' => q{Hora do último ponto de controle:}, }, + +## Romanian 'ro' => { 'checkpoint-po' => q{Timpul ultimului punct de control:}, }, + +## Russian 'ru' => { 'backends-po' => q{��������, ��� ������� ����� ��������}, 'checkpoint-po' => q{����� ��������� checkpoint:}, }, + +## Slovak 'sk' => { 'backends-po' => q{je mi ��to, je u� pr�li� ve�a klientov}, 'checkpoint-po' => q{Čas posledného kontrolného bodu:}, }, + +## Slovenian 'sl' => { 'backends-po' => q{povezanih je �e preve� odjemalcev}, 'checkpoint-po' => q{�as zadnje kontrolne to�ke ............}, }, + +## Swedish 'sv' => { 'backends-po' => q{ledsen, f�r m�nga klienter}, 'checkpoint-po' => q{Tidpunkt f�r senaste kontrollpunkt:}, }, + +## Tamil 'ta' => { 'checkpoint-po' => q{நவீன சோதனை மையத்தின் நேரம்:}, }, + +## Turkish 'tr' => { 'backends-po' => q{üzgünüm, istemci sayısı çok fazla}, 'checkpoint-po' => q{En son checkpoint'in zamanı:}, }, + +## Chinese (simplified) 'zh_CN' => { 'backends-po' => q{�Բ���, �Ѿ���̫���Ŀͻ�}, 'checkpoint-po' => q{���¼�������ʱ��:}, }, + +## Chinese (traditional) 'zh_TW' => { 'backends-po' => q{對不起，用戶端過多}, 'checkpoint-po' => q{最新的檢查點時間:}, @@ -715,57 +1297,187 @@ package check_postgres; user => { SQL => q{ -SELECT *, usename AS name, quote_ident(usename) AS safeusename +SELECT *, usename AS name, quote_ident(usename) AS saferolname FROM pg_user}, deletecols => [ qw{ passwd } ], }, schema => { SQL => q{ -SELECT n.oid, quote_ident(nspname) AS name, quote_ident(usename) AS owner, nspacl +SELECT n.oid, quote_ident(nspname) AS name, quote_ident(rolname) AS owner, nspacl FROM pg_namespace n -JOIN pg_user u ON (u.usesysid = n.nspowner)}, +JOIN pg_roles r ON (r.oid = n.nspowner)}, deletecols => [ ], exclude => 'temp_schemas', }, language => { SQL => q{ -SELECT l.*, lanname AS name, quote_ident(usename) AS owner +SELECT l.*, lanname AS name, quote_ident(rolname) AS owner FROM pg_language l -JOIN pg_user u ON (u.usesysid = l.lanowner)}, +JOIN pg_roles r ON (r.oid = l.lanowner)}, SQL2 => q{ SELECT l.*, lanname AS name FROM pg_language l }, }, + aggregate => { + SQL => q{ +SELECT a.*, aggfnoid AS name +FROM pg_aggregate a}, + }, + cast => { + SQL => q{ SELECT c.*, FORMAT('%s AS %s', format_type(castsource,NULL), format_type(casttarget,NULL)) AS name FROM pg_cast c}, + }, + comment => { + SQL => q{ +SELECT CASE WHEN so.name IS NULL THEN 'Unknown:' || sd.classoid || ':' || sd.objoid + ELSE so.object || ';' || so.name +END AS name, sd.description AS comment +FROM pg_shdescription sd +LEFT JOIN ( + SELECT 'database' AS object, + oid, tableoid, datname AS name FROM pg_database + UNION ALL SELECT 'role' AS object, + oid, tableoid, rolname AS name FROM pg_authid + UNION ALL SELECT 'tablespace' AS object, + oid, tableoid, spcname AS name FROM pg_tablespace +) AS so ON (so.oid = sd.objoid AND so.tableoid = sd.classoid) + +UNION ALL ( +SELECT CASE WHEN o.name IS NULL THEN 'Unknown:' || d.classoid || ':' || d.objoid + WHEN objsubid > 0 THEN 'column:' || o.name || ':' || (SELECT attname FROM pg_attribute WHERE attrelid::regclass::text = o.name AND attnum = d.objsubid) + ELSE COALESCE(o.object,'?') || ';' || o.name END AS name, d.description AS comment +FROM pg_description d +LEFT JOIN ( + + SELECT CASE WHEN relkind = 'q' THEN 'sequence' WHEN relkind = 'r' THEN 'table' + WHEN relkind IN ('r','p','T') THEN 'table' + WHEN relkind = 'f' THEN 'foreign table' + WHEN relkind IN ('i','I') THEN 'index' + WHEN relkind = 'v' THEN 'view' + WHEN relkind = 'm' THEN 'materialized view' + WHEN relkind = 'S' THEN 'sequence' + WHEN relkind = 'c' THEN 'type' + END AS object, + oid, tableoid, FORMAT('%s.%s', relnamespace::regnamespace, relname) AS name + FROM pg_class + + UNION ALL SELECT 'access method' AS object, + oid, tableoid, amname AS name FROM pg_am + + UNION ALL SELECT 'aggregate' AS object, + oid, tableoid, FORMAT('%s.%s(%s) ', pronamespace::regnamespace, proname, pg_get_function_arguments(oid) ) AS name FROM pg_proc WHERE prokind='a' + + UNION ALL SELECT 'cast' AS object, + oid, tableoid, FORMAT('%s AS %s', format_type(castsource,NULL), format_type(casttarget,NULL)) AS name FROM pg_cast + + UNION ALL SELECT 'collation' AS object, oid, tableoid, FORMAT('%s.%s', collnamespace::regnamespace, collname) AS name FROM pg_collation + + UNION ALL SELECT 'constraint' AS object, + oid, tableoid, FORMAT('%s.%s on %s', connamespace::regnamespace, conname, conrelid::regclass) AS name FROM pg_constraint + + UNION ALL SELECT 'conversion' AS object, + oid, tableoid, FORMAT('%s.%s', connamespace::regnamespace, conname) AS name FROM pg_conversion + + UNION ALL SELECT 'domain' AS object, + oid, tableoid, FORMAT('%s.%s', typnamespace::regnamespace, typname) AS name FROM pg_type WHERE typtype = 'd' + + UNION ALL SELECT 'extension' AS object, oid, tableoid, FORMAT('%s.%s', extnamespace::regnamespace, extname) AS name FROM pg_extension + + UNION ALL SELECT 'event trigger' AS object, oid, tableoid, evtname AS name FROM pg_event_trigger + + UNION ALL SELECT 'foreign data wrapper' AS object, + oid, tableoid, fdwname AS name FROM pg_foreign_data_wrapper + + UNION ALL SELECT 'function' AS object, + oid, tableoid, FORMAT('%s.%s(%s)', pronamespace::regnamespace, proname, pg_get_function_arguments(oid)) AS name FROM pg_proc WHERE prokind IN ('f','p') + + UNION ALL SELECT 'large object' AS object, + loid, tableoid, loid::text AS name FROM pg_largeobject + + UNION ALL SELECT 'operator' AS object, + oid, tableoid, FORMAT('%s.%s(%s,%s)', oprnamespace::regnamespace, oprname, format_type(oprleft,NULL), format_type(oprright,NULL)) AS name FROM pg_operator + + UNION ALL SELECT 'operator class' AS object, + oid, tableoid, FORMAT('%s.%s using %s', opcnamespace::regnamespace, opcname, (SELECT amname FROM pg_am WHERE oid = opcmethod)) AS name FROM pg_opclass + + UNION ALL SELECT 'operator family' AS object, + oid, tableoid, FORMAT('%s.%s using %s', opfnamespace::regnamespace, opfname, (SELECT amname FROM pg_am WHERE oid = opfmethod)) AS name FROM pg_opfamily + + UNION ALL SELECT 'policy' AS object, oid, tableoid, FORMAT('%s ON %s', polname, polrelid::regclass) AS name FROM pg_policy + + UNION ALL SELECT 'language' AS object, + oid, tableoid, lanname AS name FROM pg_language + + UNION ALL SELECT 'rule' AS object, + oid, tableoid, FORMAT('%s ON %s', rulename, ev_class::regclass) AS name FROM pg_rewrite + + UNION ALL SELECT 'schema' AS object, + oid, tableoid, nspname AS name FROM pg_namespace + + UNION ALL SELECT 'server' AS object, + oid, tableoid, srvname AS name FROM pg_foreign_server + + UNION ALL SELECT 'text search configuration', + oid, tableoid, FORMAT('%s.%s', cfgnamespace::regnamespace, cfgname) AS name FROM pg_ts_config + + UNION ALL SELECT 'text search dictionary', + oid, tableoid, FORMAT('%s.%s', dictnamespace::regnamespace, dictname) AS name FROM pg_ts_dict + + UNION ALL SELECT 'text search parser', + oid, tableoid, FORMAT('%s.%s', prsnamespace::regnamespace, prsname) AS name FROM pg_ts_parser + + UNION ALL SELECT 'text search template', + oid, tableoid, FORMAT('%s.%s', tmplnamespace::regnamespace, tmplname) AS name FROM pg_ts_template + + UNION ALL SELECT 'trigger' AS object, + oid, tableoid, FORMAT('%s on %s', tgname, tgrelid::regclass) AS name FROM pg_trigger + + UNION ALL SELECT 'type' AS object, + oid, tableoid, FORMAT('%s.%s', typnamespace::regnamespace, typname) AS name FROM pg_type + +) AS o ON (o.oid = d.objoid AND o.tableoid = d.classoid))}, + }, + domain => { + SQL => q{ SELECT *, FORMAT('%s.%s', typnamespace::regnamespace, typname) AS name FROM pg_type WHERE typtype = 'd'}, + }, + + extension => { + SQL => q{ +SELECT e.*, extname AS name, quote_ident(rolname) AS owner +FROM pg_extension e +JOIN pg_roles r ON (r.oid = e.extowner)}, + }, + type => { SQL => q{ -SELECT t.oid AS oid, t.*, quote_ident(usename) AS owner, quote_ident(nspname) AS schema, +SELECT t.oid AS oid, t.*, quote_ident(rolname) AS owner, quote_ident(nspname) AS schema, nspname||'.'||typname AS name FROM pg_type t -JOIN pg_user u ON (u.usesysid = t.typowner) +JOIN pg_roles r ON (r.oid = t.typowner) JOIN pg_namespace n ON (n.oid = t.typnamespace) WHERE t.typtype NOT IN ('b','c')}, exclude => 'system', }, sequence => { SQL => q{ -SELECT c.*, nspname||'.'||relname AS name, quote_ident(usename) AS owner, +SELECT c.*, nspname||'.'||relname AS name, quote_ident(rolname) AS owner, (quote_ident(nspname)||'.'||quote_ident(relname)) AS safename, quote_ident(nspname) AS schema FROM pg_class c -JOIN pg_user u ON (u.usesysid = c.relowner) +JOIN pg_roles r ON (r.oid = c.relowner) JOIN pg_namespace n ON (n.oid = c.relnamespace) WHERE c.relkind = 'S'}, innerSQL => 'SELECT * FROM ROWSAFENAME', + SQL10 => q{SELECT schemaname||'.'||sequencename AS name, * FROM pg_sequences}, }, view => { SQL => q{ -SELECT c.*, nspname||'.'||relname AS name, quote_ident(usename) AS owner, - quote_ident(relname) AS safename, quote_ident(nspname) AS schema, +SELECT c.*, nspname||'.'||relname AS name, quote_ident(rolname) AS owner, + quote_ident(relname) AS safename, quote_ident(nspname) AS schemaname, TRIM(pg_get_viewdef(c.oid, TRUE)) AS viewdef, spcname AS tablespace FROM pg_class c -JOIN pg_user u ON (u.usesysid = c.relowner) +JOIN pg_roles r ON (r.oid = c.relowner) JOIN pg_namespace n ON (n.oid = c.relnamespace) LEFT JOIN pg_tablespace s ON (s.oid = c.reltablespace) WHERE c.relkind = 'v'}, @@ -773,11 +1485,11 @@ package check_postgres; }, table => { SQL => q{ -SELECT c.*, nspname||'.'||relname AS name, quote_ident(usename) AS owner, - quote_ident(relname) AS safename, quote_ident(nspname) AS schema, +SELECT c.*, nspname||'.'||relname AS name, quote_ident(rolname) AS owner, + quote_ident(relname) AS safename, quote_ident(nspname) AS schemaname, spcname AS tablespace FROM pg_class c -JOIN pg_user u ON (u.usesysid = c.relowner) +JOIN pg_roles r ON (r.oid = c.relowner) JOIN pg_namespace n ON (n.oid = c.relnamespace) LEFT JOIN pg_tablespace s ON (s.oid = c.reltablespace) WHERE c.relkind = 'r'}, @@ -785,12 +1497,12 @@ package check_postgres; }, index => { SQL => q{ -SELECT c.*, i.*, nspname||'.'||relname AS name, quote_ident(usename) AS owner, - quote_ident(relname) AS safename, quote_ident(nspname) AS schema, +SELECT c.*, i.*, nspname||'.'||relname AS name, quote_ident(rolname) AS owner, + quote_ident(relname) AS safename, quote_ident(nspname) AS schemaname, spcname AS tablespace, amname, - pg_get_indexdef(c.oid) AS indexdef + pg_get_indexdef(c.oid) AS indexdef, indrelid::regclass::text AS tablename FROM pg_class c -JOIN pg_user u ON (u.usesysid = c.relowner) +JOIN pg_roles r ON (r.oid = c.relowner) JOIN pg_namespace n ON (n.oid = c.relnamespace) JOIN pg_index i ON (c.oid = i.indexrelid) LEFT JOIN pg_tablespace s ON (s.oid = c.reltablespace) @@ -798,52 +1510,55 @@ package check_postgres; WHERE c.relkind = 'i'}, exclude => 'system', }, - operator => { - SQL => q{ -SELECT o.*, o.oid, nspname||'.'||o.oprname AS name, quote_ident(o.oprname) AS safename, - usename AS owner, nspname AS schema, - t1.typname AS resultname, - t2.typname AS leftname, t3.typname AS rightname -FROM pg_operator o -JOIN pg_user u ON (u.usesysid = o.oprowner) -JOIN pg_namespace n ON (n.oid = o.oprnamespace) -JOIN pg_proc p1 ON (p1.oid = o.oprcode) -JOIN pg_type t1 ON (t1.oid = o.oprresult) -LEFT JOIN pg_type t2 ON (t2.oid = o.oprleft) -LEFT JOIN pg_type t3 ON (t3.oid = o.oprright)}, - exclude => 'system', +operator => { SQL => q{ + SELECT FORMAT('%s.%s(%s,%s)', oprnamespace::regnamespace, oprname, format_type(oprleft,NULL), format_type(oprright,NULL)) AS name, + rolname AS owner, quote_ideNT(oprnamespace::regnamespace::text) AS schemaname + FROM pg_operator o + JOIN pg_roles r ON (r.oid = o.oprowner) + WHERE oprnamespace::regnamespace::text <> 'pg_catalog' +}, }, trigger => { SQL => q{ -SELECT t.*, n1.nspname||'.'||t.tgname AS name, quote_ident(t.tgname) AS safename, quote_ident(usename) AS owner, - n1.nspname AS tschema, c1.relname AS tname, +SELECT t.*, n1.nspname||'.'||c1.relname||'.'||t.tgname AS name, quote_ident(t.tgname) AS safename, quote_ident(rolname) AS owner, + quote_ident(n1.nspname) AS schemaname, + n1.nspname||'.'||c1.relname AS tablename, n2.nspname AS cschema, c2.relname AS cname, - n3.nspname AS procschema, p.proname AS procname + n3.nspname AS procschema, p.proname AS procname, + pg_get_triggerdef(t.oid) AS triggerdef, + ( WITH nums AS (SELECT unnest(tgattr) AS poz) + SELECT string_agg(attname,',') FROM pg_attribute a JOIN nums ON + (nums.poz = a.attnum AND tgrelid = a.attrelid) + ) AS trigger_columns FROM pg_trigger t JOIN pg_class c1 ON (c1.oid = t.tgrelid) -JOIN pg_user u ON (u.usesysid = c1.relowner) +JOIN pg_roles r ON (r.oid = c1.relowner) JOIN pg_namespace n1 ON (n1.oid = c1.relnamespace) JOIN pg_proc p ON (p.oid = t.tgfoid) JOIN pg_namespace n3 ON (n3.oid = p.pronamespace) LEFT JOIN pg_class c2 ON (c2.oid = t.tgconstrrelid) LEFT JOIN pg_namespace n2 ON (n2.oid = c2.relnamespace) -WHERE t.tgconstrrelid = 0 AND tgname !~ '^pg_'}, +WHERE t.tgconstrrelid = 0 AND t.tgconstrindid = 0 AND tgname !~ '^pg_'}, }, function => { SQL => q{ SELECT p.*, p.oid, nspname||'.'||p.proname AS name, quote_ident(p.proname) AS safename, md5(prosrc) AS source_checksum, - usename AS owner, nspname AS schema + rolname AS owner, quote_ident(nspname) AS schemaname, + pg_get_function_arguments(p.oid) AS function_arguments FROM pg_proc p -JOIN pg_user u ON (u.usesysid = p.proowner) +JOIN pg_roles r ON (r.oid = p.proowner) JOIN pg_namespace n ON (n.oid = p.pronamespace)}, exclude => 'system', }, constraint => { SQL => q{ -SELECT c.*, c.oid, n.nspname||'.'||c.conname AS name, quote_ident(c.conname) AS safename, - n.nspname AS schema, relname AS tname +SELECT c.*, c.oid, n.nspname||'.'||c1.relname||'.'||c.conname AS name, quote_ident(c.conname) AS safename, + quote_ident(n.nspname) AS schemaname, + n.nspname||'.'||r.relname AS tablename, + pg_get_constraintdef(c.oid) AS constraintdef, translate(c.confmatchtype,'u','s') AS confmatchtype_compat FROM pg_constraint c +JOIN pg_class c1 ON (c1.oid = c.conrelid) JOIN pg_namespace n ON (n.oid = c.connamespace) JOIN pg_class r ON (r.oid = c.conrelid) JOIN pg_namespace n2 ON (n2.oid = r.relnamespace)}, @@ -852,8 +1567,8 @@ package check_postgres; column => { SQL => q{ SELECT a.*, n.nspname||'.'||c.relname||'.'||attname AS name, quote_ident(a.attname) AS safename, - n.nspname||'.'||c.relname AS tname, - typname, quote_ident(nspname) AS schema, + n.nspname||'.'||c.relname AS tablename, + typname, quote_ident(nspname) AS schemaname, pg_get_expr(d.adbin, a.attrelid, true) AS default FROM pg_attribute a JOIN pg_type t ON (t.oid = a.atttypid) @@ -865,6 +1580,25 @@ package check_postgres; postSQL => q{ORDER BY n.nspname, c.relname, a.attnum}, exclude => 'system', }, + + foreign_server => { + SQL => q{ +SELECT f.*, srvname AS name, quote_ident(rolname) AS owner +FROM pg_foreign_server f +JOIN pg_roles r ON (r.oid = f.srvowner)}, + }, + foreign_table => { + SQL => q{ +SELECT srvname||':'||ftrelid::regclass AS name, ftoptions +FROM pg_foreign_table +JOIN pg_foreign_server f on (f.oid = ftserver)}, + }, + foreign_data_wrapper => { + SQL => q{ +SELECT f.*, fdwname AS name, quote_ident(rolname) AS owner +FROM pg_foreign_data_wrapper f +JOIN pg_roles r ON (r.oid = f.fdwowner)}, + }, ); my $rcfile; @@ -889,7 +1623,7 @@ package check_postgres; RCLINE: while (<$rc>) { next if /^\s*#/; - next unless /^\s*(\w+)\s*=\s*(.+?)\s*$/o; + next unless /^\s*(\w+)\s*=\s*(.+?)\s*$/; my ($name,$value) = ($1,$2); ## no critic (ProhibitCaptureWithoutTest) ## Map alternate option spellings to preferred names if ($name eq 'dbport' or $name eq 'p' or $name eq 'dbport1' or $name eq 'p1' or $name eq 'port1') { @@ -905,16 +1639,16 @@ package check_postgres; $name = 'dbuser'; } ## Now for all the additional non-1 databases - elsif ($name =~ /^dbport(\d+)$/o or $name eq /^p(\d+)$/o) { + elsif ($name =~ /^dbport(\d+)$/ or $name eq /^p(\d+)$/) { $name = "port$1"; } - elsif ($name =~ /^dbhost(\d+)$/o or $name eq /^H(\d+)$/o) { + elsif ($name =~ /^dbhost(\d+)$/ or $name eq /^H(\d+)$/) { $name = "host$1"; } - elsif ($name =~ /^db(\d)$/o) { + elsif ($name =~ /^db(\d)$/) { $name = "dbname$1"; } - elsif ($name =~ /^u(\d+)$/o) { + elsif ($name =~ /^u(\d+)$/) { $name = "dbuser$1"; } @@ -953,12 +1687,16 @@ package check_postgres; 'no-check_postgresrc', 'assume-standby-mode', 'assume-prod', + 'assume-async', 'action=s', 'warning=s', 'critical=s', 'include=s@', 'exclude=s@', + 'alldb', + 'includedb=s@', + 'excludedb=s@', 'includeuser=s@', 'excludeuser=s@', @@ -968,11 +1706,13 @@ package check_postgres; 'dbuser|u|dbuser1|u1=s@', 'dbpass|dbpass1=s@', 'dbservice|dbservice1=s@', + 'role=s', 'PGBINDIR=s', 'PSQL=s', 'tempdir=s', + 'audit-file-dir=s', 'get_method=s', 'language=s', 'mrtg=s', ## used by MRTG checks only @@ -988,6 +1728,11 @@ package check_postgres; 'filter=s@', ## used by same_schema only 'suffix=s', ## used by same_schema only 'replace', ## used by same_schema only + 'skipsequencevals', ## used by same_schema only + 'lsfunc=s', ## used by wal_files and archive_ready + 'object=s@', ## used by same_schema for object types to include + 'skipobject=s@', ## used by same_schema for object types to exclude + 'skipcycled', ## used by sequence only ); die $USAGE if ! keys %opt and ! @ARGV; @@ -998,25 +1743,25 @@ package check_postgres; while (my $arg = pop @ARGV) { ## These must be of the form x=y - if ($arg =~ /^\-?\-?(\w+)\s*=\s*(.+)/o) { + if ($arg =~ /^\-?\-?(\w+)\s*=\s*(.+)/) { my ($name,$value) = (lc $1, $2); - if ($name =~ /^(?:db)?port(\d+)$/o or $name =~ /^p(\d+)$/o) { - $opt{"port$1"} = $value; + if ($name =~ /^(?:db)?port(\d+)$/ or $name =~ /^p(\d+)$/) { + push @{ $opt{port} } => $value; } - elsif ($name =~ /^(?:db)?host(\d+)$/o or $name =~ /^H(\d+)$/o) { - $opt{"host$1"} = $value; + elsif ($name =~ /^(?:db)?host(\d+)$/ or $name =~ /^H(\d+)$/) { + push @{ $opt{host} } => $value; } - elsif ($name =~ /^db(?:name)?(\d+)$/o) { - $opt{"dbname$1"} = $value; + elsif ($name =~ /^db(?:name)?(\d+)$/) { + push @{ $opt{dbname} } => $value; } - elsif ($name =~ /^dbuser(\d+)$/o or $name =~ /^u(\d+)/o) { - $opt{"dbuser$1"} = $value; + elsif ($name =~ /^dbuser(\d+)$/ or $name =~ /^u(\d+)/) { + push @{ $opt{dbuser} } => $value; } - elsif ($name =~ /^dbpass(\d+)$/o) { - $opt{"dbpass$1"} = $value; + elsif ($name =~ /^dbpass(\d+)$/) { + push @{ $opt{dbpass} } => $value; } - elsif ($name =~ /^dbservice(\d+)$/o) { - $opt{"dbservice$1"} = $value; + elsif ($name =~ /^dbservice(\d+)$/) { + push @{ $opt{dbservice} } => $value; } else { push @badargs => $arg; @@ -1048,13 +1793,13 @@ package check_postgres; our $VERBOSE = $opt{verbose} || 0; $VERBOSE = 5 if $opt{vv}; -our $OUTPUT = lc($opt{output} || ''); +our $OUTPUT = lc($opt{output} // ''); ## Allow the optimization of the get_methods list by an argument if ($opt{get_method}) { my $found = 0; for my $meth (@get_methods) { - if ($meth =~ /^$opt{get_method}/io) { + if ($meth =~ /^$opt{get_method}/i) { @get_methods = ($meth); $found = 1; last; @@ -1085,7 +1830,7 @@ package check_postgres; if (!$OUTPUT) { my $dir = getcwd; - if ($dir =~ /(nagios|mrtg|simple|cacti)/io) { + if ($dir =~ /(nagios|mrtg|simple|cacti)/i) { $OUTPUT = lc $1; } elsif ($opt{simple}) { @@ -1102,7 +1847,7 @@ package check_postgres; if ($OUTPUT =~ /\b(kb|mb|gb|tb|eb)\b/) { $opt{transform} = uc $1; } -if ($OUTPUT =~ /(nagios|mrtg|simple|cacti)/io) { +if ($OUTPUT =~ /(nagios|mrtg|simple|cacti)/i) { $OUTPUT = lc $1; } ## Check for a valid output setting @@ -1147,21 +1892,25 @@ package check_postgres; fsm_relations => [1, 'Checks percentage of relations used in free space map.'], hitratio => [0, 'Report if the hit ratio of a database is too low.'], hot_standby_delay => [1, 'Check the replication delay in hot standby setup'], - index_size => [0, 'Checks the size of indexes only.'], - table_size => [0, 'Checks the size of tables only.'], relation_size => [0, 'Checks the size of tables and indexes.'], + index_size => [0, 'Checks the size of indexes.'], + table_size => [0, 'Checks the size of tables (including TOAST).'], + indexes_size => [0, 'Checks the size of indexes on tables.'], + total_relation_size => [0, 'Checks the size of tables (including indexes and TOAST).'], last_analyze => [0, 'Check the maximum time in seconds since any one table has been analyzed.'], last_vacuum => [0, 'Check the maximum time in seconds since any one table has been vacuumed.'], last_autoanalyze => [0, 'Check the maximum time in seconds since any one table has been autoanalyzed.'], last_autovacuum => [0, 'Check the maximum time in seconds since any one table has been autovacuumed.'], listener => [0, 'Checks for specific listeners.'], locks => [0, 'Checks the number of locks.'], + lockwait => [0, 'Checks for blocking locks.'], logfile => [1, 'Checks that the logfile is being written to correctly.'], new_version_bc => [0, 'Checks if a newer version of Bucardo is available.'], new_version_box => [0, 'Checks if a newer version of boxinfo is available.'], new_version_cp => [0, 'Checks if a newer version of check_postgres.pl is available.'], new_version_pg => [0, 'Checks if a newer version of Postgres is available.'], new_version_tnm => [0, 'Checks if a newer version of tail_n_mail is available.'], + partman_premake => [1, 'Checks if premake partitions are present.'], pgb_pool_cl_active => [1, 'Check the number of active clients in each pgbouncer pool.'], pgb_pool_cl_waiting => [1, 'Check the number of waiting clients in each pgbouncer pool.'], pgb_pool_sv_active => [1, 'Check the number of active server connections in each pgbouncer pool.'], @@ -1172,11 +1921,13 @@ package check_postgres; pgb_pool_maxwait => [1, 'Check the current maximum wait time for client connections in pgbouncer pools.'], pgbouncer_backends => [0, 'Check how many clients are connected to pgbouncer compared to max_client_conn.'], pgbouncer_checksum => [0, 'Check that no pgbouncer settings have changed since the last check.'], + pgbouncer_maxwait => [0, 'Check how long the first (oldest) client in queue has been waiting.'], pgagent_jobs => [0, 'Check for no failed pgAgent jobs within a specified period of time.'], prepared_txns => [1, 'Checks number and age of prepared transactions.'], query_runtime => [0, 'Check how long a specific query takes to run.'], query_time => [1, 'Checks the maximum running time of current queries.'], replicate_row => [0, 'Verify a simple update gets replicated to another server.'], + replication_slots => [1, 'Check the replication delay for replication slots'], same_schema => [0, 'Verify that two databases have the exact same tables, columns, etc.'], sequence => [0, 'Checks remaining calls left in sequences.'], settings_checksum => [0, 'Check that no settings have changed since the last check.'], @@ -1222,12 +1973,16 @@ package check_postgres; -c value, --critical=value the critical threshold, range depends on the action --include=name(s) items to specifically include (e.g. tables), depends on the action --exclude=name(s) items to specifically exclude (e.g. tables), depends on the action + --alldb list all postgres databases and run action over them + --excludedb=name regex filter for the alldb option to select only certain databases + --includedb=name regex filter for the alldb option to select only certain databases --includeuser=include objects owned by certain users --excludeuser=exclude objects owned by certain users Other options: --assume-standby-mode assume that server in continious WAL recovery mode --assume-prod assume that server in production mode + --assume-async assume that any replication is asynchronous --PGBINDIR=PATH path of the postgresql binaries; avoid using if possible --PSQL=FILE (deprecated) location of the psql executable; avoid using if possible -v, --verbose verbosity level; can be used more than once to increase the level @@ -1244,7 +1999,7 @@ package check_postgres; $ME --man -Or visit: http://bucardo.org/check_postgres/ +Or visit: https://bucardo.org/check_postgres/ }; @@ -1351,12 +2106,12 @@ sub msg_en { } else { my $psql = (defined $PGBINDIR) ? "$PGBINDIR/psql" : 'psql'; - chomp($PSQL = qx{which "$psql"}); + chomp($PSQL = qx{command -v "$psql"}); $PSQL or ndie msg('opt-psql-nofind'); } -x $PSQL or ndie msg('opt-psql-noexec', $PSQL); $res = qx{$PSQL --version}; -$res =~ /psql\D+(\d+\.\d+)/ or ndie msg('opt-psql-nover'); +$res =~ /psql\D+(\d+(?:\.\d+)?)/ or ndie msg('opt-psql-nover'); our $psql_version = $1; $VERBOSE >= 2 and warn qq{psql=$PSQL version=$psql_version\n}; @@ -1364,6 +2119,44 @@ sub msg_en { $opt{defaultdb} = $psql_version >= 8.0 ? 'postgres' : 'template1'; $opt{defaultdb} = 'pgbouncer' if $action =~ /^pgb/; +## If alldb is set then run a psql command to find out all the databases +if (defined $opt{alldb}){ + + my $pg_port = $opt{defaultport}; + if ($opt{port}[0]){ + $pg_port = $opt{port}[0]; + } + my $psql_output = join(",", map /^([\w|-]+?)\|/, qx{$PSQL -A -l -t -p $pg_port }); + my $pg_db; + # optionally exclude or include each db + my @psql_output_array = split(/,/, $psql_output); + for $pg_db (@psql_output_array) { + if (defined $opt{includedb}){ + if ($pg_db =~ /$opt{includedb}[0]/) { + # do nothing + } else { + # strip the database from the listing + $psql_output =~ s/($pg_db),//; + } + } + if (defined $opt{excludedb}){ + if ($pg_db =~ /$opt{excludedb}[0]/) { + # strip the database from the listing + $psql_output =~ s/($pg_db),//; + } else { + # do nothing + } + } + } + # strip out some dbs we're not interested in + $psql_output =~ s/(template0,)//; + $psql_output =~ s/(root,)//; + # pg8.4 + $psql_output =~ s/(,:)//g; + $opt{dbname}[0] = $psql_output; +} + + ## Check the current database mode our $STANDBY = 0; our $MASTER = 0; @@ -1415,7 +2208,7 @@ sub add_response { } $same_schema_header .= sprintf "\nDB %s: %s%s%s%s%s", $number, - defined $row->{dbservice} ? qq{dbservice=$row->{dbservice} } : '', + (defined $row->{dbservice} and length $row->{dbservice}) ? qq{dbservice=$row->{dbservice} } : '', defined $row->{port} ? qq{port=$row->{port} } : '', defined $row->{host} ? qq{host=$row->{host} } : '', defined $row->{dbname} ? qq{dbname=$row->{dbname} } : '', @@ -1455,7 +2248,7 @@ sub add_response { $dbport; $header =~ s/\s+$//; $header =~ s/^ //; - my $perf = ($opt{showtime} and $db->{totaltime} and $action ne 'bloat') ? "time=$db->{totaltime}s" : ''; + my $perf = ($opt{showtime} and $db->{totaltime} and $action ne 'bloat' and $action !~ /^pgb_pool_/ ) ? "time=$db->{totaltime}s" : ''; if ($db->{perf}) { $db->{perf} =~ s/^ +//; if (length $same_schema_header) { @@ -1663,11 +2456,11 @@ sub finishup { ## Are we showing DEBUG_INFO? my $showdebug = 0; if ($DEBUGOUTPUT) { - $showdebug = 1 if $DEBUGOUTPUT =~ /a/io - or ($DEBUGOUTPUT =~ /c/io and $type eq 'c') - or ($DEBUGOUTPUT =~ /w/io and $type eq 'w') - or ($DEBUGOUTPUT =~ /o/io and $type eq 'o') - or ($DEBUGOUTPUT =~ /u/io and $type eq 'u'); + $showdebug = 1 if $DEBUGOUTPUT =~ /a/i + or ($DEBUGOUTPUT =~ /c/i and $type eq 'c') + or ($DEBUGOUTPUT =~ /w/i and $type eq 'w') + or ($DEBUGOUTPUT =~ /o/i and $type eq 'o') + or ($DEBUGOUTPUT =~ /u/i and $type eq 'u'); } for (sort keys %$info) { printf '%s %s%s ', @@ -1752,6 +2545,7 @@ sub finishup { fsm_pages => 'VERSION: 8.2 MAX: 8.3', fsm_relations => 'VERSION: 8.2 MAX: 8.3', hot_standby_delay => 'VERSION: 9.0', + replication_slots => 'VERSION: 9.4', listener => 'MAX: 8.4', ); if ($opt{test}) { @@ -1773,11 +2567,11 @@ sub finishup { my $limit = $testaction{lc $ac}; next if ! defined $limit; - if ($limit =~ /VERSION: ((\d+)\.(\d+))/) { + if ($limit =~ /VERSION: ((\d+)\.?(\d+))/) { my ($rver,$rmaj,$rmin) = ($1,$2,$3); for my $db (@{$info->{db}}) { next unless exists $db->{ok}; - if ($set{server_version} !~ /((\d+)\.(\d+))/) { + if ($set{server_version} !~ /((\d+)\.?(\d+))/) { print msgn('testmode-nover', $db->{pname}); next; } @@ -1793,7 +2587,7 @@ sub finishup { my ($rver,$rmaj,$rmin) = ($1,$2,$3); for my $db (@{$info->{db}}) { next unless exists $db->{ok}; - if ($set{server_version} !~ /((\d+)\.(\d+))/) { + if ($set{server_version} !~ /((\d+)\.?(\d+))/) { print msgn('testmode-nover', $db->{pname}); next; } @@ -1836,10 +2630,10 @@ sub finishup { my $safename; if (1 == keys %userlist) { ($safename = each %userlist) =~ s/'/''/g; - $USERWHERECLAUSE = " AND usename = '$safename'"; + $USERWHERECLAUSE = " AND rolname = '$safename'"; } else { - $USERWHERECLAUSE = ' AND usename IN ('; + $USERWHERECLAUSE = ' AND rolname IN ('; for my $user (sort keys %userlist) { ($safename = $user) =~ s/'/''/g; $USERWHERECLAUSE .= "'$safename',"; @@ -1858,10 +2652,10 @@ sub finishup { my $safename; if (1 == keys %userlist) { ($safename = each %userlist) =~ s/'/''/g; - $USERWHERECLAUSE = " AND usename <> '$safename'"; + $USERWHERECLAUSE = " AND rolname <> '$safename'"; } else { - $USERWHERECLAUSE = ' AND usename NOT IN ('; + $USERWHERECLAUSE = ' AND rolname NOT IN ('; for my $user (sort keys %userlist) { ($safename = $user) =~ s/'/''/g; $USERWHERECLAUSE .= "'$safename',"; @@ -1892,10 +2686,12 @@ sub finishup { ## Check local disk_space - local means it must be run from the same box! check_disk_space() if $action eq 'disk_space'; -## Check the size of relations, or more specifically, tables and indexes -check_index_size() if $action eq 'index_size'; -check_table_size() if $action eq 'table_size'; -check_relation_size() if $action eq 'relation_size'; +## Check the size of relations (tables, toast tables, indexes) +check_relation_size('relation', 'rtim') if $action eq 'relation_size'; +check_relation_size('relation', 'i') if $action eq 'index_size'; +check_relation_size('table', 'rm') if $action eq 'table_size'; +check_relation_size('indexes', 'rtm') if $action eq 'indexes_size'; +check_relation_size('total_relation', 'rm') if $action eq 'total_relation_size'; ## Check how long since the last full analyze check_last_analyze() if $action eq 'last_analyze'; @@ -1915,6 +2711,9 @@ sub finishup { ## Check number and type of locks check_locks() if $action eq 'locks'; +## Check lock wait +check_lockwait() if $action eq 'lockwait'; + ## Logfile is being written to check_logfile() if $action eq 'logfile'; @@ -1945,6 +2744,9 @@ sub finishup { ## Check the replication delay in hot standby setup check_hot_standby_delay() if $action eq 'hot_standby_delay'; +## Check the delay on replication slots. warning and critical are sizes +check_replication_slots() if $action eq 'replication_slots'; + ## Check the maximum transaction age of all connections check_txn_time() if $action eq 'txn_time'; @@ -1990,6 +2792,9 @@ sub finishup { ## Make sure Slony is behaving check_slony_status() if $action eq 'slony_status'; +## Make sure Partman premake is working +check_partman_premake() if $action eq 'partman_premake'; + ## Verify that the pgbouncer settings are what we think they should be check_pgbouncer_checksum() if $action eq 'pgbouncer_checksum'; @@ -2017,6 +2822,9 @@ sub finishup { ## Check the current maximum wait time for client connections in pgbouncer pools check_pgb_pool('maxwait') if $action eq 'pgb_pool_maxwait'; +## Check how long the first (oldest) client in queue has been waiting. +check_pgbouncer_maxwait() if $action eq 'pgbouncer_maxwait'; + ## Check how many clients are connected to pgbouncer compared to max_client_conn. check_pgbouncer_backends() if $action eq 'pgbouncer_backends'; @@ -2121,15 +2929,15 @@ sub pretty_time { ## Just seconds (< 2:00) if ($sec < 120 or $tweak =~ /s/) { - return sprintf "$sec %s", $sec==1 ? msg('time-second') : msg('time-seconds'); + return sprintf "$sec %s", 1==$sec ? msg('time-second') : msg('time-seconds'); } ## Minutes and seconds (< 60:00) if ($sec < 60*60 or $tweak =~ /m/) { my $min = int $sec / 60; $sec %= 60; - my $ret = sprintf "$min %s", $min==1 ? msg('time-minute') : msg('time-minutes'); - $sec and $tweak !~ /S/ and $ret .= sprintf " $sec %s", $sec==1 ? msg('time-second') : msg('time-seconds'); + my $ret = sprintf "$min %s", 1==$min ? msg('time-minute') : msg('time-minutes'); + $sec and $tweak !~ /S/ and $ret .= sprintf " $sec %s", 1==$sec ? msg('time-second') : msg('time-seconds'); return $ret; } @@ -2139,9 +2947,9 @@ sub pretty_time { $sec -= ($hour*60*60); my $min = int $sec / 60; $sec -= ($min*60); - my $ret = sprintf "$hour %s", $hour==1 ? msg('time-hour') : msg('time-hours'); - $min and $tweak !~ /M/ and $ret .= sprintf " $min %s", $min==1 ? msg('time-minute') : msg('time-minutes'); - $sec and $tweak !~ /[SM]/ and $ret .= sprintf " $sec %s", $sec==1 ? msg('time-second') : msg('time-seconds'); + my $ret = sprintf "$hour %s", 1==$hour ? msg('time-hour') : msg('time-hours'); + $min and $tweak !~ /M/ and $ret .= sprintf " $min %s", 1==$min ? msg('time-minute') : msg('time-minutes'); + $sec and $tweak !~ /[SM]/ and $ret .= sprintf " $sec %s", 1==$sec ? msg('time-second') : msg('time-seconds'); return $ret; } @@ -2153,10 +2961,10 @@ sub pretty_time { $sec -= ($our*60*60); my $min = int $sec / 60; $sec -= ($min*60); - my $ret = sprintf "$day %s", $day==1 ? msg('time-day') : msg('time-days'); - $our and $tweak !~ /H/ and $ret .= sprintf " $our %s", $our==1 ? msg('time-hour') : msg('time-hours'); - $min and $tweak !~ /[HM]/ and $ret .= sprintf " $min %s", $min==1 ? msg('time-minute') : msg('time-minutes'); - $sec and $tweak !~ /[HMS]/ and $ret .= sprintf " $sec %s", $sec==1 ? msg('time-second') : msg('time-seconds'); + my $ret = sprintf "$day %s", 1==$day ? msg('time-day') : msg('time-days'); + $our and $tweak !~ /H/ and $ret .= sprintf " $our %s", 1==$our ? msg('time-hour') : msg('time-hours'); + $min and $tweak !~ /[HM]/ and $ret .= sprintf " $min %s", 1==$min ? msg('time-minute') : msg('time-minutes'); + $sec and $tweak !~ /[HMS]/ and $ret .= sprintf " $sec %s", 1==$sec ? msg('time-second') : msg('time-seconds'); return $ret; } @@ -2169,11 +2977,11 @@ sub pretty_time { $sec -= ($our*60*60); my $min = int $sec / 60; $sec -= ($min*60); - my $ret = sprintf "$week %s", $week==1 ? msg('time-week') : msg('time-weeks'); - $day and $tweak !~ /D/ and $ret .= sprintf " $day %s", $day==1 ? msg('time-day') : msg('time-days'); - $our and $tweak !~ /[DH]/ and $ret .= sprintf " $our %s", $our==1 ? msg('time-hour') : msg('time-hours'); - $min and $tweak !~ /[DHM]/ and $ret .= sprintf " $min %s", $min==1 ? msg('time-minute') : msg('time-minutes'); - $sec and $tweak !~ /[DHMS]/ and $ret .= sprintf " $sec %s", $sec==1 ? msg('time-second') : msg('time-seconds'); + my $ret = sprintf "$week %s", 1==$week ? msg('time-week') : msg('time-weeks'); + $day and $tweak !~ /D/ and $ret .= sprintf " $day %s", 1==$day ? msg('time-day') : msg('time-days'); + $our and $tweak !~ /[DH]/ and $ret .= sprintf " $our %s", 1==$our ? msg('time-hour') : msg('time-hours'); + $min and $tweak !~ /[DHM]/ and $ret .= sprintf " $min %s", 1==$min ? msg('time-minute') : msg('time-minutes'); + $sec and $tweak !~ /[DHMS]/ and $ret .= sprintf " $sec %s", 1==$sec ? msg('time-second') : msg('time-seconds'); return $ret; } ## end of pretty_time @@ -2229,9 +3037,11 @@ sub run_command { $tempdir = tempdir(@tempdirargs); ($tempfh,$tempfile) = tempfile('check_postgres_psql.XXXXXXX', SUFFIX => '.tmp', DIR => $tempdir); + binmode($tempfh, ':utf8'); ## Create another one to catch any errors ($errfh,$errorfile) = tempfile('check_postgres_psql_stderr.XXXXXXX', SUFFIX => '.tmp', DIR => $tempdir); + binmode($errfh, ':utf8'); ## Mild cleanup of the query $string =~ s/^\s*(.+?)\s*$/$1/s; @@ -2254,6 +3064,10 @@ sub run_command { if ($arg->{dbnumber} and $arg->{dbnumber} != $num) { next; } + ## Likewise if we have specified "target" database info and this is not our choice + if ($arg->{target} and $arg->{target} != $db) { + next; + } ## Just to keep things clean: truncate $tempfh, 0; @@ -2262,7 +3076,7 @@ sub run_command { ## Store this target in the global target list push @{$info->{db}}, $db; - my @args = ('-q', '-t'); + my @args = ('-X', '-q', '-t'); if (defined $db->{dbservice} and length $db->{dbservice}) { ## XX Check for simple names $db->{pname} = "service=$db->{dbservice}"; $ENV{PGSERVICE} = $db->{dbservice}; @@ -2276,12 +3090,14 @@ sub run_command { return $db->{pname}; } - defined $db->{dbname} and push @args, '-d', $db->{dbname}; - defined $db->{dbuser} and push @args, '-U', $db->{dbuser}; - defined $db->{port} and push @args => '-p', $db->{port}; - if ($db->{host} ne '<none>') { - push @args => '-h', $db->{host}; - $host{$db->{host}}++; ## For the overall count + if ($db->{pname} !~ /service=/) { + defined $db->{dbname} and push @args, '-d', $db->{dbname}; + defined $db->{dbuser} and push @args, '-U', $db->{dbuser}; + defined $db->{port} and push @args => '-p', $db->{port}; + if ($db->{host} ne '<none>') { + push @args => '-h', $db->{host}; + $host{$db->{host}}++; ## For the overall count + } } if (defined $db->{dbpass} and length $db->{dbpass}) { @@ -2308,7 +3124,7 @@ sub run_command { else { $string = $arg->{oldstring} || $arg->{string}; for my $row (@{$arg->{version}}) { - if ($row !~ s/^([<>]?)(\d+\.\d+)\s+//) { + if ($row !~ s/^([<>]?)(\d+\.?\d+|1\d+)\s+//) { ndie msg('die-badversion', $row); } my ($mod,$ver) = ($1||'',$2); @@ -2329,9 +3145,17 @@ sub run_command { } } - local $SIG{ALRM} = sub { die 'Timed out' }; + local $SIG{ALRM} = sub { die "Timed out\n" }; alarm 0; + if ($opt{role}) { + if ($psql_version < 9.6) { + ndie msg('no-role') + } + else { + push @args, '-c', "SET ROLE $opt{role}"; + } + } push @args, '-c', $string; $VERBOSE >= 3 and warn Dumper \@args; @@ -2347,6 +3171,10 @@ sub run_command { alarm 0; open STDERR, '>&', $oldstderr or ndie msg('runcommand-noerr'); close $oldstderr or ndie msg('file-noclose', 'STDERR copy', $!); + if ($err and $action eq 'connection') { + $info->{fatal} = $err; + return $info; + } if ($err) { if ($err =~ /Timed out/) { ndie msg('runcommand-timeout', $timeout); @@ -2371,8 +3199,8 @@ sub run_command { } ## If we are just trying to connect, failed attempts are critical - if ($action eq 'connection' and $db->{error} =~ /FATAL|could not connect/) { - $info->{fatal} = 1; + if ($action eq 'connection' and $db->{error}) { + $info->{fatal} = $db->{error}; return $info; } @@ -2413,10 +3241,13 @@ sub run_command { $db->{ok} = 1; ## Unfortunately, psql outputs "(No rows)" even with -t and -x - $db->{slurp} = '' if ! defined $db->{slurp} or index($db->{slurp},'(')==0; + $db->{slurp} = '' if ! defined $db->{slurp} or 0 == index($db->{slurp},'('); + + ## Remove carriage returns (i.e. on Win32) + $db->{slurp} =~ s/\r//g; ## Allow an empty query (no matching rows) if requested - if ($arg->{emptyok} and $db->{slurp} =~ /^\s*$/o) { + if ($arg->{emptyok} and $db->{slurp} =~ /^\s*$/) { $arg->{emptyok2} = 1; } ## If we just want a version, grab it and redo @@ -2424,7 +3255,7 @@ sub run_command { if ($db->{error}) { ndie $db->{error}; } - if ($db->{slurp} !~ /(\d+\.\d+)/) { + if ($db->{slurp} !~ /([789]\.\d+|1\d+)/) { ndie msg('die-badversion', $db->{slurp}); } $db->{version} = $1; @@ -2455,15 +3286,15 @@ sub run_command { my $lastval; for my $line (split /\n/ => $db->{slurp}) { - if (index($line,'-')==0) { + if (0 == index($line,'-')) { $lnum++; next; } - if ($line =~ /^([\?\w]+)\s+\| (.*)/) { + if ($line =~ /^ ?([\?\w]+)\s+\| (.*?)\s*$/) { $stuff[$lnum]{$1} = $2; $lastval = $1; } - elsif ($line =~ /^QUERY PLAN\s+\| (.*)/) { + elsif ($line =~ /^ ?QUERY PLAN\s+\| (.*)/) { $stuff[$lnum]{queryplan} = $1; $lastval = 'queryplan'; } @@ -2488,7 +3319,7 @@ sub run_command { warn "OS: $^O\n"; warn "Action: $action\n"; warn "Calling line: $cline\n"; - warn "Output: $line\n"; + warn "Output: >>$line<<\n"; $args =~ s/ -c (.+)/ -c "$1"/s; warn "Command: $PSQL $args\n"; ## Next to last thing is to see if we can grab the PG version @@ -2500,7 +3331,7 @@ sub run_command { warn "Postgres version: $v\n"; } ## This is a serious parsing fail, so it can be helpful to have the whole enchilada: - warn "Full output: $db->{slurp}\n\n"; + warn 'Full output: ' . (Dumper $db->{slurp}) . "\n\n"; exit 1; } } @@ -2575,13 +3406,6 @@ sub setup_target_databases { dbservice => [''], }; - ## Don't set any default values if a service is being used - if (defined $opt{dbservice} and defined $opt{dbservice}->[0] and length $opt{dbservice}->[0]) { - $conn->{dbname} = []; - $conn->{port} = []; - $conn->{dbuser} = []; - } - ## If we were passed in a target, use that and move on if (exists $arg->{target}) { ## Make a copy, in case we are passed in a ref @@ -2607,6 +3431,17 @@ sub setup_target_databases { my %group; my $found_new_var = 0; + ## Do we have any using multiple values for non-passwords? + my $got_multiple = 0; + for my $v (keys %$conn) { + next if $v eq 'dbpass' or ! defined $opt{$v}[0]; + my $num = $opt{$v}->@*; + if ($num > 1 or $opt{$v}[0] =~ /,/) { + $got_multiple = 1; + last; + } + } + for my $v (keys %$conn) { ## For each connection var such as port, host... my $vname = $v; @@ -2619,7 +3454,7 @@ sub setup_target_databases { $new =~ s/\s+//g unless $vname eq 'dbservice' or $vname eq 'host'; ## Set this as the new default for this connection var moving forward - $conn->{$vname} = [split /,/ => $new]; + $conn->{$vname} = $got_multiple ? [split /,/ => $new, -1] : [$new]; ## Make a note that we found something new this round $found_new_var = 1; @@ -2675,8 +3510,7 @@ sub verify_version { my $limit = $testaction{lc $action} || ''; my $versiononly = shift || 0; - - return if ! $limit and ! $versiononly; + return if ! $limit and ! $versiononly and !defined wantarray; ## We almost always need the version, so just grab it for any limitation $SQL = q{SELECT setting FROM pg_settings WHERE name = 'server_version'}; @@ -2689,16 +3523,27 @@ sub verify_version { ndie $info->{db}[0]{error}; } - if (!defined $info->{db}[0] or $info->{db}[0]{slurp}[0]{setting} !~ /((\d+)\.(\d+))/) { + if (!defined $info->{db}[0]) { + ndie msg('die-badversion', $SQL); + } + + my ($sver,$smaj,$smin); + if ( + $info->{db}[0]{slurp}[0]{setting} !~ /^(([2-9])\.(\d+))/ && + $info->{db}[0]{slurp}[0]{setting} !~ /^((1\d+)())/ + ){ + ndie msg('die-badversion', $SQL); } - my ($sver,$smaj,$smin) = ($1,$2,$3); + else { + ($sver,$smaj,$smin) = ($1,$2,$3||0); + } if ($versiononly) { return $sver; } - if ($limit =~ /VERSION: ((\d+)\.(\d+))/) { + if ($limit =~ /VERSION: ((\d+)(?:\.(\d+))?)/) { my ($rver,$rmaj,$rmin) = ($1,$2,$3); if ($smaj < $rmaj or ($smaj==$rmaj and $smin < $rmin)) { ndie msg('die-action-version', $action, $rver, $sver); @@ -2725,7 +3570,7 @@ sub verify_version { } $db->{slurp} = $oldslurp; - return; + return $sver; } ## end of verify_version @@ -2781,7 +3626,7 @@ sub skip_item { if (defined $opt{exclude}) { $stat = 1; for (@{$opt{exclude}}) { - for my $ex (split /\s*,\s*/o => $_) { + for my $ex (split /\s*,\s*/ => $_) { if ($ex =~ s/\.$//) { if ($ex =~ s/^~//) { ($stat += 2 and last) if $schema =~ /$ex/; @@ -2802,7 +3647,7 @@ sub skip_item { if (defined $opt{include}) { $stat += 4; for (@{$opt{include}}) { - for my $in (split /\s*,\s*/o => $_) { + for my $in (split /\s*,\s*/ => $_) { if ($in =~ s/\.$//) { if ($in =~ s/^~//) { ($stat += 8 and last) if $schema =~ /$in/; @@ -2863,7 +3708,7 @@ sub validate_range { } if (length $critical) { if ($critical !~ $timesecre) { - ndie msg('range-seconds', 'critical') + ndie msg('range-seconds', 'critical'); } $critical = $1; if (!$arg->{any_warning} and length $warning and $warning > $critical) { @@ -2883,10 +3728,10 @@ sub validate_range { } elsif ('version' eq $type) { my $msg = msg('range-version'); - if (length $warning and $warning !~ /^\d+\.\d+\.?[\d\w]*$/) { + if (length $warning and $warning !~ /^\d+\.?\d+\.?[\d\w]*$/) { ndie msg('range-badversion', 'warning', $msg); } - if (length $critical and $critical !~ /^\d+\.\d+\.?[\d\w]*$/) { + if (length $critical and $critical !~ /^\d+\.?\d+\.?[\d\w]*$/) { ndie msg('range-badversion', 'critical', $msg); } if (! length $critical and ! length $warning) { @@ -3201,17 +4046,39 @@ sub open_controldata { } ## Run pg_controldata - ## We still catch deprecated option my $pgc; if (defined $ENV{PGCONTROLDATA} and length $ENV{PGCONTROLDATA}) { - # ndie msg('depr-pgcontroldata'); $pgc = "$ENV{PGCONTROLDATA}"; } + elsif (defined $ENV{PGBINDIR} and length $ENV{PGBINDIR}) { + $pgc = "$PGBINDIR/pg_controldata"; + } else { - $pgc = (defined $PGBINDIR) ? "$PGBINDIR/pg_controldata" : 'pg_controldata'; - chomp($pgc = qx{which "$pgc"}); + my $pgctry = 'pg_controldata'; + my $result = qx{$pgctry --version 2>/dev/null}; + if ($result =~ /\d/) { + $pgc = $pgctry; + } + else { + ## Need to refactor this someday + my $basedir = '/usr/lib/postgresql/'; + if (opendir my $dh, $basedir) { + for my $subdir (sort { $b <=> $a } grep { /^\d+[\d\.]+$/ } readdir $dh) { + $pgctry = catfile($basedir, $subdir, 'bin', 'pg_controldata'); + next if ! -e $pgctry; + $result = qx{$pgctry --version 2>/dev/null}; + if ($result =~ /\d/) { + $pgc = $pgctry; + last; + } + } + closedir $dh; + } + if (! defined $pgc) { + ndie msg('checkpoint-nobin'); + } + } } - -x $pgc or ndie msg('opt-psql-noexec', $pgc); $COM = qq{$pgc "$dir"}; eval { @@ -3442,7 +4309,7 @@ sub check_backends { $nwarn = $limit-$w2; } elsif ($w3) { - $nwarn = (int $w2*$limit/100) + $nwarn = (int $w2*$limit/100); } if (! skip_item($r->{datname})) { @@ -3526,8 +4393,8 @@ sub check_bloat { ## Can also specify percentages ## Don't bother with tables or indexes unless they have at least this many bloated pages - my $MINPAGES = 0; - my $MINIPAGES = 10; + my $MINPAGES = 10; + my $MINIPAGES = 15; my $LIMIT = 10; if ($opt{perflimit}) { @@ -3547,12 +4414,12 @@ sub check_bloat { ROUND(CASE WHEN otta=0 OR sml.relpages=0 OR sml.relpages=otta THEN 0.0 ELSE sml.relpages/otta::numeric END,1) AS tbloat, CASE WHEN relpages < otta THEN 0 ELSE relpages::bigint - otta END AS wastedpages, CASE WHEN relpages < otta THEN 0 ELSE bs*(sml.relpages-otta)::bigint END AS wastedbytes, - CASE WHEN relpages < otta THEN '0 bytes'::text ELSE (bs*(relpages-otta))::bigint || ' bytes' END AS wastedsize, + CASE WHEN relpages < otta THEN '0 bytes'::text ELSE (bs*(relpages-otta))::bigint::text || ' bytes' END AS wastedsize, iname, ituples::bigint AS itups, ipages::bigint AS ipages, iotta, ROUND(CASE WHEN iotta=0 OR ipages=0 OR ipages=iotta THEN 0.0 ELSE ipages/iotta::numeric END,1) AS ibloat, CASE WHEN ipages < iotta THEN 0 ELSE ipages::bigint - iotta END AS wastedipages, CASE WHEN ipages < iotta THEN 0 ELSE bs*(ipages-iotta) END AS wastedibytes, - CASE WHEN ipages < iotta THEN '0 bytes' ELSE (bs*(ipages-iotta))::bigint || ' bytes' END AS wastedisize, + CASE WHEN ipages < iotta THEN '0 bytes' ELSE (bs*(ipages-iotta))::bigint::text || ' bytes' END AS wastedisize, CASE WHEN relpages < otta THEN CASE WHEN ipages < iotta THEN 0 ELSE bs*(ipages-iotta::bigint) END ELSE CASE WHEN ipages < iotta THEN bs*(relpages-otta::bigint) @@ -3588,9 +4455,9 @@ sub check_bloat { FROM pg_stats s2 WHERE null_frac<>0 AND s2.schemaname = ns.nspname AND s2.tablename = tbl.relname ) AS nullhdr - FROM pg_attribute att + FROM pg_attribute att JOIN pg_class tbl ON att.attrelid = tbl.oid - JOIN pg_namespace ns ON ns.oid = tbl.relnamespace + JOIN pg_namespace ns ON ns.oid = tbl.relnamespace LEFT JOIN pg_stats s ON s.schemaname=ns.nspname AND s.tablename = tbl.relname AND s.inherited=false @@ -3623,14 +4490,10 @@ sub check_bloat { ## Alternate versions for old versions my $SQL2 = $SQL; - if ($psql_version <= 8.4) { - $SQL2 =~ s/AND s.inherited=false//; - } + $SQL2 =~ s/AND s.inherited=false//; # 8.4 and earlier my $SQL3 = $SQL2; - if ($psql_version <= 7.4) { - $SQL3 =~ s/SELECT current_setting.+?AS bs/(SELECT 8192) AS bs/; - } + $SQL3 =~ s/SELECT current_setting.+?AS bs/(SELECT 8192) AS bs/; # 7.4 and earlier my $info = run_command($SQL, { version => [ "<8.0 $SQL3", "<9.0 $SQL2" ] } ); @@ -3645,7 +4508,7 @@ sub check_bloat { $db = $info->{db}[0]; - if ($db->{slurp} !~ /\w+/o) { + if ($db->{slurp} !~ /\w+/) { add_ok msg('bloat-nomin') unless $MRTG; return; } @@ -3676,7 +4539,7 @@ sub check_bloat { qw/ iname irows ipages iotta ibloat wastedipgaes wastedibytes wastedisize/}; ## Made it past the exclusions - $max = -2 if $max == -1; + $max = -2 if -1 == $max; ## Do the table first if we haven't seen it if (! $seenit{"$dbname.$schema.$table"}++) { @@ -3747,10 +4610,10 @@ sub check_bloat { $db->{perf} = ''; } - if ($max == -1) { + if (-1 == $max) { add_unknown msg('no-match-rel'); } - elsif ($max != -1) { + elsif (-1 != $max) { add_ok $maxmsg; } @@ -3832,7 +4695,7 @@ sub check_checkpoint { ndie msg('checkpoint-noparse', $last); } my $diff = time - $dt; - my $msg = $diff==1 ? msg('checkpoint-ok') : msg('checkpoint-ok2', $diff); + my $msg = 1==$diff ? msg('checkpoint-ok') : msg('checkpoint-ok2', $diff); $db->{perf} = sprintf '%s=%s;%s;%s', perfname(msg('age')), $diff, $warning, $critical; @@ -3951,10 +4814,10 @@ sub check_commitratio { SELECT round(100.*sd.xact_commit/(sd.xact_commit+sd.xact_rollback), 2) AS dcommitratio, d.datname, - u.usename + r.rolname AS rolname FROM pg_stat_database sd JOIN pg_database d ON (d.oid=sd.datid) -JOIN pg_user u ON (u.usesysid=d.datdba) +JOIN pg_roles r ON (r.oid=d.datdba) WHERE sd.xact_commit+sd.xact_rollback<>0 $USERWHERECLAUSE }; @@ -4034,17 +4897,21 @@ sub check_connection { } my $info = run_command('SELECT version() AS v'); + if ($info->{fatal}) { + add_critical $info->{fatal}; + return; + } for $db (@{$info->{db}}) { my $err = $db->{error} || ''; - if ($err =~ /FATAL|could not connect/) { + if ($err) { $MRTG and do_mrtg({one => 0}); add_critical $db->{error}; return; } - my $ver = ($db->{slurp}[0]{v} =~ /(\d+\.\d+\S+)/o) ? $1 : ''; + my $ver = ($db->{slurp}[0]{v} =~ /((?:\b1\d\S+)|(?:\d+\.\d+\S+))/) ? $1 : ''; $MRTG and do_mrtg({one => $ver ? 1 : 0}); @@ -4166,9 +5033,9 @@ sub check_database_size { SELECT pg_database_size(d.oid) AS dsize, pg_size_pretty(pg_database_size(d.oid)) AS pdsize, datname, - usename + r.rolname AS rolname FROM pg_database d -LEFT JOIN pg_user u ON (u.usesysid=d.datdba)$USERWHERECLAUSE +LEFT JOIN pg_roles r ON (r.oid=d.datdba)$USERWHERECLAUSE }; if ($opt{perflimit}) { $SQL .= " ORDER BY 1 DESC LIMIT $opt{perflimit}"; @@ -4430,7 +5297,7 @@ sub check_disk_space { ## Check log_directory: relative or absolute if (length $logdir) { - if ($logdir =~ /^\w/) { ## relative, check only if symlinked + if ($logdir =~ /^[\w\.]/) { ## relative, check only if symlinked $logdir = "$datadir/$logdir"; if (-l $logdir) { my $linkdir = readlink($logdir); @@ -4555,9 +5422,9 @@ sub check_fsm_pages { (my $c = $critical) =~ s/\D//; my $SQL = q{ SELECT pages, maxx, ROUND(100*(pages/maxx)) AS percent -FROM +FROM (SELECT (sumrequests+numrels)*chunkpages AS pages - FROM (SELECT SUM(CASE WHEN avgrequest IS NULL + FROM (SELECT SUM(CASE WHEN avgrequest IS NULL THEN interestingpages/32 ELSE interestingpages/16 END) AS sumrequests, COUNT(relfilenode) AS numrels, 16 AS chunkpages FROM pg_freespacemap_relations) AS foo) AS foo2, (SELECT setting::NUMERIC AS maxx FROM pg_settings WHERE name = 'max_fsm_pages') AS foo3 @@ -4617,7 +5484,7 @@ sub check_fsm_relations { my $SQL = q{ SELECT maxx, cur, ROUND(100*(cur/maxx)) AS percent -FROM (SELECT +FROM (SELECT (SELECT COUNT(*) FROM pg_freespacemap_relations) AS cur, (SELECT setting::NUMERIC FROM pg_settings WHERE name='max_fsm_relations') AS maxx) x }; @@ -4675,10 +5542,10 @@ sub check_hitratio { SELECT round(100.*sd.blks_hit/(sd.blks_read+sd.blks_hit), 2) AS dhitratio, d.datname, - u.usename + r.rolname AS rolname FROM pg_stat_database sd JOIN pg_database d ON (d.oid=sd.datid) -JOIN pg_user u ON (u.usesysid=d.datdba) +JOIN pg_roles r ON (r.oid=d.datdba) WHERE sd.blks_read+sd.blks_hit<>0 $USERWHERECLAUSE }; @@ -4758,8 +5625,10 @@ sub check_hot_standby_delay { ## --warning=5min ## --warning='1048576 and 2min' --critical='16777216 and 10min' + my $version = verify_version(); + my ($warning, $wtime, $critical, $ctime) = validate_integer_for_time({default_to_int => 1}); - if ($psql_version < 9.1 and (length $wtime or length $ctime)) { + if ($version < 9.1 and (length $wtime or length $ctime)) { add_unknown msg('hs-time-version'); return; } @@ -4798,8 +5667,13 @@ sub check_hot_standby_delay { my ($moffset, $s_rec_offset, $s_rep_offset, $time_delta); ## On slave - $SQL = q{SELECT pg_last_xlog_receive_location() AS receive, pg_last_xlog_replay_location() AS replay}; - if ($psql_version >= 9.1) { + if ($version >= 10) { + $SQL = q{SELECT pg_last_wal_receive_lsn() AS receive, pg_last_wal_replay_lsn() AS replay}; + } + else { + $SQL = q{SELECT pg_last_xlog_receive_location() AS receive, pg_last_xlog_replay_location() AS replay}; + } + if ($version >= 9.1) { $SQL .= q{, COALESCE(ROUND(EXTRACT(epoch FROM now() - pg_last_xact_replay_timestamp())),0) AS seconds}; } my $info = run_command($SQL, { dbnumber => $slave, regex => qr/\// }); @@ -4809,12 +5683,12 @@ sub check_hot_standby_delay { my $replay = $db->{slurp}[0]{replay}; $time_delta = $db->{slurp}[0]{seconds}; - if (defined $receive) { + if (defined $receive and length $receive) { my ($a, $b) = split(/\//, $receive); $s_rec_offset = (hex('ff000000') * hex($a)) + hex($b); } - if (defined $replay) { + if (defined $replay and length $replay) { my ($a, $b) = split(/\//, $replay); $s_rep_offset = (hex('ff000000') * hex($a)) + hex($b); } @@ -4828,7 +5702,12 @@ sub check_hot_standby_delay { } ## On master - $SQL = q{SELECT pg_current_xlog_location() AS location}; + if ($version >= 10) { + $SQL = q{SELECT pg_current_wal_lsn() AS location}; + } + else { + $SQL = q{SELECT pg_current_xlog_location() AS location}; + } $info = run_command($SQL, { dbnumber => $master }); for $db (@{$info->{db}}) { my $location = $db->{slurp}[0]{location}; @@ -4846,34 +5725,39 @@ sub check_hot_standby_delay { ## Compute deltas $db = $saved_db; - my $rec_delta = $moffset - $s_rec_offset; - my $rep_delta = $moffset - $s_rep_offset; + my ($rec_delta, $rep_delta); + $rec_delta = $moffset - $s_rec_offset if defined $s_rec_offset; + $rep_delta = $moffset - $s_rep_offset if defined $s_rep_offset; # Make sure it's always positive or zero - $rec_delta = 0 if $rec_delta < 0; - $rep_delta = 0 if $rep_delta < 0; + $rec_delta = 0 if defined $rec_delta and $rec_delta < 0; + $rep_delta = 0 if defined $rep_delta and $rep_delta < 0; if (defined $time_delta and $time_delta < 0) { add_unknown msg('hs-future-replica'); return; } - $MRTG and do_mrtg($psql_version >= 9.1 ? + $MRTG and do_mrtg($version >= 9.1 ? {one => $rep_delta, two => $rec_delta, three => $time_delta} : {one => $rep_delta, two => $rec_delta}); - $db->{perf} = sprintf ' %s=%s;%s;%s ', - perfname(msg('hs-replay-delay')), $rep_delta, $warning, $critical; - $db->{perf} .= sprintf ' %s=%s;%s;%s', - perfname(msg('hs-receive-delay')), $rec_delta, $warning, $critical; - if ($psql_version >= 9.1) { + if (defined $rep_delta) { + $db->{perf} = sprintf ' %s=%s;%s;%s', + perfname(msg('hs-replay-delay')), $rep_delta, $warning, $critical; + } + if (defined $rec_delta) { + $db->{perf} .= sprintf ' %s=%s;%s;%s', + perfname(msg('hs-receive-delay')), $rec_delta, $warning, $critical; + } + if ($version >= 9.1) { $db->{perf} .= sprintf ' %s=%s;%s;%s', perfname(msg('hs-time-delay')), $time_delta, $wtime, $ctime; } ## Do the check on replay delay in case SR has disconnected because it way too far behind my $msg = qq{$rep_delta}; - if ($psql_version >= 9.1) { - $msg .= qq{ and $time_delta seconds} + if ($version >= 9.1) { + $msg .= qq{ and $time_delta seconds}; } if ((length $critical or length $ctime) and (!length $critical or length $critical and $rep_delta > $critical) and (!length $ctime or length $ctime and $time_delta > $ctime)) { add_critical $msg; @@ -4889,6 +5773,98 @@ sub check_hot_standby_delay { } ## end of check_hot_standby_delay +sub check_replication_slots { + + ## Check the delay on one or more replication slots + ## Supports: Nagios, MRTG + ## mrtg reports the largest two delays + ## By default, checks all replication slots + ## Can check specific one(s) with include + ## Can ignore some with exclude + ## Warning and critical are bytes + ## Valid units: b, k, m, g, t, e + ## All above may be written as plural or with a trailing 'b' + + my ($warning, $critical) = validate_range({type => 'size'}); + + $SQL = q{ + WITH slots AS (SELECT slot_name, + slot_type, + coalesce(restart_lsn, '0/0'::pg_lsn) AS slot_lsn, + coalesce( + pg_xlog_location_diff( + case when pg_is_in_recovery() then pg_last_xlog_receive_location() else pg_current_xlog_location() end, + restart_lsn), + 0) AS delta, + active + FROM pg_replication_slots) + SELECT *, pg_size_pretty(delta) AS delta_pretty FROM slots; + }; + + if ($opt{perflimit}) { + $SQL .= " ORDER BY 1 DESC LIMIT $opt{perflimit}"; + } + my $SQL10 = $SQL; + $SQL10 =~ s/xlog_location/wal_lsn/g; + $SQL10 =~ s/xlog_receive_location/wal_receive_lsn/g; + + my $info = run_command($SQL, { regex => qr{\d+}, emptyok => 1, version => [">9.6 $SQL10"] } ); + my $found = 0; + + for $db (@{$info->{db}}) { + my $max = -1; + $found = 1; + my %s; + + for my $r (@{$db->{slurp}}) { + if (skip_item($r->{slot_name})) { + $max = -2 if -1 == $max; + next; + } + if ($r->{delta} >= $max) { + $max = $r->{delta}; + } + $s{$r->{slot_name}} = [$r->{delta},$r->{delta_pretty},$r->{slot_type},$r->{slot_lsn},$r->{active}]; + } + if ($MRTG) { + do_mrtg({one => $max, msg => "SLOT: $db->{slot_name}"}); + } + if ($max < 0) { + $stats{$db->{dbname}} = 0; + add_ok msg('no-match-slotok') if -1 == $max; + add_unknown msg('no-match-slot') if -2 == $max; + next; + } + + my $msg = ''; + for (sort {$s{$b}[0] <=> $s{$a}[0] or $a cmp $b } keys %s) { + $msg .= "$_: $s{$_}[1] ($s{$_}[2] $s{$_}[3] " . ($s{$_}[4] eq 't' ? 'active' : 'inactive') .') '; + $db->{perf} .= sprintf ' %s=%s;%s;%s', + perfname($_), $s{$_}[0], $warning, $critical; + } + if (length $critical and $max >= $critical) { + add_critical $msg; + } + elsif (length $warning and $max >= $warning) { + add_warning $msg; + } + else { + add_ok $msg; + } + } + + ## If no results, probably a version problem + if (!$found and keys %unknown) { + (my $first) = values %unknown; + if ($first->[0][0] =~ /pg_replication_slots/) { + ndie msg('repslot-version'); + } + } + + return; + +} ## end of check_replication_slot_delay + sub check_last_analyze { my $auto = shift || ''; @@ -4948,7 +5924,7 @@ sub check_last_vacuum_analyze { } if ($USERWHERECLAUSE) { - $SQL =~ s/ WHERE/, pg_user u WHERE u.usesysid=c.relowner$USERWHERECLAUSE AND/; + $SQL =~ s/ WHERE/, pg_roles r WHERE r.oid=c.relowner$USERWHERECLAUSE AND/; } my $info = run_command($SQL, { regex => qr{\w}, emptyok => 1 } ); @@ -4999,7 +5975,7 @@ sub check_last_vacuum_analyze { do_mrtg({one => $mintime, msg => $maxrel}); return; } - if ($maxtime == -2) { + if (-2 == $maxtime) { add_unknown ( $found ? $type eq 'vacuum' ? msg('vac-nomatch-v') : msg('vac-nomatch-a') @@ -5134,7 +6110,7 @@ sub check_locks { } if ($critical) { for my $l (keys %{$critical}) { - #$dblock{$k}{$l} = 0 if ! exists $dblock{$k}{$l}; + $dblock{$k}{$l} = 0 if ! exists $dblock{$k}{$l}; } } for my $m (keys %{$dblock{$k}}){ @@ -5206,6 +6182,63 @@ sub check_locks { } ## end of check_locks +sub check_lockwait { + + ## Check lock wait + ## By default, checks all databases + ## Can check specific databases with include + ## Can ignore databases with exclude + ## Warning and critical is time + ## Example: --warning='1 min' --critical='2 min' + + my ($warning, $critical) = validate_range + ({ + type => 'time', + default_warning => '1 min', + default_critical => '2 min', + }); + + $SQL = qq{SELECT a.datname AS datname, + bl.pid AS blocked_pid, + a.usename AS blocked_user, + ka.pid AS blocking_pid, + ka.usename AS blocking_user, + round(extract (epoch from current_timestamp - a.query_start)) AS waited_sec, + a.query AS blocked_statement + FROM pg_catalog.pg_locks bl + JOIN pg_catalog.pg_stat_activity a ON a.pid = bl.pid + JOIN pg_catalog.pg_stat_activity ka ON (ka.pid = ANY(pg_blocking_pids(bl.pid))) + WHERE NOT bl.granted + }; + my $info = run_command($SQL, { regex => qr{\w}, emptyok => 1 }); + my $n = 0; + for $db (@{$info->{db}}) { + ROW: for my $r (@{$db->{slurp}}) { + my ($dbname,$blocked_pid,$blocked_user,$blocking_pid,$blocking_user,$waited_sec,$blocked_statement) + = ($r->{datname},$r->{blocked_pid}, $r->{blocked_user}, $r->{blocking_pid}, + $r->{blocking_user},$r->{waited_sec},$r->{blocked_statement}); + + ## May be forcibly skipping this database via arguments + next ROW if skip_item($dbname); + + my $msg = msg 'lockwait-msg',$dbname,$blocking_user,$blocking_pid,$blocked_user,$blocked_pid,pretty_time($waited_sec),$blocked_statement; + if (length $critical and $waited_sec >= $critical) { + add_critical $msg; + } + elsif (length $warning and $waited_sec >= $warning) { + add_warning $msg; + } + else { + add_ok $msg; + } + $n++; + } + } + add_ok 'No blocking locks' if ($n==0); + do_mrtg( {one => $n} ) if $MRTG; + return; + +} ## end of check_lockwait sub check_logfile { @@ -5341,7 +6374,7 @@ sub check_logfile { } close $logfh or ndie msg('file-noclose', $logfile, $!); - if ($found == 1) { + if (1 == $found) { $MRTG and do_mrtg({one => 1}); add_ok msg('logfile-ok', $logfile); } @@ -5360,8 +6393,10 @@ sub find_new_version { my $url = shift or die; ## The format is X.Y.Z [optional message] - my $versionre = qr{((\d+)\.(\d+)\.(\d+))\s*(.*)}; + my $versionre = qr{((\d+)\.(\d+)(?:\.(\d+))?)(?:\s+(.*))?}; + my ($cversion,$cmajor,$cminor,$crevision,$cmessage) = ('','','','',''); + my $found = 0; ## Try to fetch the current version from the web @@ -5373,16 +6408,21 @@ sub find_new_version { ## Postgres is slightly different if ($program eq 'Postgres') { $cmajor = {}; - while ($info =~ /<title>(\d+)\.(\d+)\.(\d+)/g) { + while ($info =~ /<title>(\d+)\.(\d+)(?:\.(\d+))?/g) { $found = 1; - $cmajor->{"$1.$2"} = $3; + if (defined $3) { + $cmajor->{"$1.$2"} = $3; + } + else { + $cmajor->{$1} = $2; + } } } elsif ($info =~ $versionre) { $found = 1; ($cversion,$cmajor,$cminor,$crevision,$cmessage) = ($1, int $2, int $3, int $4, $5); + $info =~ s/\s+$//s; if ($VERBOSE >= 1) { - $info =~ s/\s+$//s; warn "Remote version string: $info\n"; warn "Remote version: $cversion\n"; } @@ -5419,16 +6459,22 @@ sub find_new_version { add_unknown msg('new-ver-nolver', $program); return; } - my ($lversion,$lmajor,$lminor,$lrevision) = ($1, int $2, int $3, int $4); + my ($lversion,$lmajor,$lminor,$lrevision) = ('',0,0,0); + if ($2 >= 10 && $program eq 'Postgres') { + ($lversion,$lmajor,$lrevision) = ($1, int $2, int $3); + } else { + ($lversion,$lmajor,$lminor,$lrevision) = ($1, int $2, int $3, int $4); + } + + $output =~ s/\s+$//s; if ($VERBOSE >= 1) { - $output =~ s/\s+$//s; warn "Local version string: $output\n"; warn "Local version: $lversion\n"; } ## Postgres is a special case if ($program eq 'Postgres') { - my $lver = "$lmajor.$lminor"; + my $lver = $lmajor >= 10 ? $lmajor : "$lmajor.$lminor"; if (! exists $cmajor->{$lver}) { add_unknown msg('new-ver-nocver', $program); return; @@ -5436,7 +6482,11 @@ sub find_new_version { $crevision = $cmajor->{$lver}; $cmajor = $lmajor; $cminor = $lminor; - $cversion = "$cmajor.$cminor.$crevision"; + if ($lmajor >= 10) { + $cversion = "$cmajor.$crevision"; + } else { + $cversion = "$cmajor.$cminor.$crevision"; + } } ## Most common case: everything matches @@ -5468,7 +6518,7 @@ sub check_new_version_bc { ## Check if a newer version of Bucardo is available - my $url = 'http://bucardo.org/bucardo/latest_version.txt'; + my $url = 'https://bucardo.org/bucardo/latest_version.txt'; find_new_version('Bucardo', 'bucardo_ctl', $url); return; @@ -5480,7 +6530,7 @@ sub check_new_version_box { ## Check if a newer version of boxinfo is available - my $url = 'http://bucardo.org/boxinfo/latest_version.txt'; + my $url = 'https://bucardo.org/boxinfo/latest_version.txt'; find_new_version('boxinfo', 'boxinfo.pl', $url); return; @@ -5492,7 +6542,7 @@ sub check_new_version_cp { ## Check if a new version of check_postgres.pl is available - my $url = 'http://bucardo.org/check_postgres/latest_version.txt'; + my $url = 'https://bucardo.org/check_postgres/latest_version.txt'; find_new_version('check_postgres', $VERSION, $url); return; @@ -5504,13 +6554,13 @@ sub check_new_version_pg { ## Check if a new version of Postgres is available - my $url = 'http://www.postgresql.org/versions.rss'; + my $url = 'https://www.postgresql.org/versions.rss'; ## Grab the local version my $info = run_command('SELECT version() AS version'); my $lversion = $info->{db}[0]{slurp}[0]{version}; ## Make sure it is parseable and check for development versions - if ($lversion !~ /\d+\.\d+\.\d+/) { + if ($lversion !~ /1\d+\.\d+|\d+\.\d+\.\d+/) { if ($lversion =~ /(\d+\.\d+\S+)/) { add_ok msg('new-ver-dev', 'Postgres', $1); return; @@ -5530,7 +6580,7 @@ sub check_new_version_tnm { ## Check if a new version of tail_n_mail is available - my $url = 'http://bucardo.org/tail_n_mail/latest_version.txt'; + my $url = 'https://bucardo.org/tail_n_mail/latest_version.txt'; find_new_version('tail_n_mail', 'tail_n_mail', $url); return; @@ -5570,6 +6620,7 @@ sub check_pgagent_jobs { JOIN pgagent.pga_jobsteplog slog ON jlog.jlgid = slog.jsljlgid AND step.jstid = slog.jsljstid WHERE ((slog.jslresult = -1 AND step.jstkind='s') OR (slog.jslresult <> 0 AND step.jstkind='b')) AND EXTRACT('epoch' FROM NOW() - (jlog.jlgstart + jlog.jlgduration)) < $seconds + ORDER BY jlog.jlgstart DESC }; my $info = run_command($SQL); @@ -5605,23 +6656,188 @@ sub check_pgagent_jobs { return; } -sub check_pgbouncer_checksum { - - ## Verify the checksum of all pgbouncer settings - ## Supports: Nagios, MRTG - ## Not that the connection will be done on the pgbouncer database - ## One of warning or critical must be given (but not both) - ## It should run one time to find out the expected checksum - ## You can use --critical="0" to find out the checksum - ## You can include or exclude settings as well - ## Example: - ## check_postgres_pgbouncer_checksum --critical="4e7ba68eb88915d3d1a36b2009da4acd" +sub check_partman_premake { - my ($warning, $critical) = validate_range({type => 'checksum', onlyone => 1}); + ## Checks if all premade partitions are present + ## Monthly and daily interval only + ## Supports: Nagios - eval { - require Digest::MD5; - }; + my $msg = msg('partman-premake-ok'); + my $found = 0; + my ($warning, $critical) = validate_range + ({ + type => 'integer', # in days + default_warning => '1', + default_critical => '3', + }); + + # check missing Config for range partitioned tables + + my $SQL = q{ +SELECT + current_database() AS database, + c.relnamespace::regnamespace || '.' || c.relname AS parent_table +FROM + pg_class c + JOIN pg_partitioned_table t ON t.partrelid = c.oid +WHERE + c.relkind = 'p' + AND t.partstrat = 'r' + AND NOT EXISTS ( + SELECT + 1 + FROM + partman.part_config + WHERE + parent_table = c.relnamespace::regnamespace || '.' || c.relname); +}; + + my $info = run_command($SQL, {regex => qr[\w+], emptyok => 1 } ); + my (@crit,@warn,@ok); + + for $db (@{$info->{db}}) { + my ($maxage,$maxdb) = (0,''); ## used by MRTG only + ROW: for my $r (@{$db->{slurp}}) { + my ($dbname,$parent_table) = ($r->{database},$r->{parent_table}); + $found = 1 if ! $found; + next ROW if skip_item($dbname); + $found = 2; + + $msg = "$dbname=$parent_table " . msg('partman-conf-mis'); + push @crit => $msg; + }; + }; + + # check Config Errors + + $SQL = q{ +SELECT + current_database() as database, + parent_table +FROM ( + SELECT + parent_table, + retention, + partition_interval, + EXTRACT(EPOCH FROM retention::interval) / EXTRACT(EPOCH FROM partition_interval::interval) AS configured_partitions + FROM + partman.part_config) p +WHERE + configured_partitions < 1; +}; + + $info = run_command($SQL, {regex => qr[\w+], emptyok => 1 } ); + + for $db (@{$info->{db}}) { + my ($maxage,$maxdb) = (0,''); ## used by MRTG only + ROW: for my $r (@{$db->{slurp}}) { + my ($dbname,$parent_table) = ($r->{database},$r->{parent_table}); + $found = 1 if ! $found; + next ROW if skip_item($dbname); + $found = 2; + + $msg = "$dbname=$parent_table " . msg('partman-conf-tbl'); + push @warn => $msg; + }; + }; + + + $SQL = q{ +SELECT + current_database() as database, + a.parent_table, + b.date - a.date::date AS missing_days +FROM +( +SELECT parent_table, date +FROM ( SELECT + i.inhparent::regclass as parent_table, + substring(pg_catalog.pg_get_expr(c.relpartbound, i.inhrelid)::text FROM '%TO __#"_{10}#"%' FOR '#') as date, + rank() OVER (PARTITION BY i.inhparent ORDER BY pg_catalog.pg_get_expr(c.relpartbound, i.inhrelid) DESC) + FROM pg_inherits i + JOIN pg_class c ON c.oid = i.inhrelid +WHERE c.relkind = 'r' + AND pg_catalog.pg_get_expr(c.relpartbound, i.inhrelid) != 'DEFAULT') p +WHERE + p.rank = 1 +) a +JOIN +( +SELECT + parent_table, + (now() + premake * partition_interval::interval)::date +FROM + partman.part_config +) b +ON + a.parent_table::text = b.parent_table::text +WHERE + b.date - a.date::date > 0 +ORDER BY 3, 2 DESC +}; + + $info = run_command($SQL, {regex => qr[\w+], emptyok => 1 } ); + + for $db (@{$info->{db}}) { + my ($maxage,$maxdb) = (0,''); ## used by MRTG only + ROW: for my $r (@{$db->{slurp}}) { + my ($dbname,$parent_table,$missing_days) = ($r->{database},$r->{parent_table},$r->{missing_days}); + $found = 1 if ! $found; + next ROW if skip_item($dbname); + $found = 2; + + $msg = "$dbname=$parent_table ($missing_days)"; + print "$msg"; + $db->{perf} .= sprintf ' %s=%sd;%s;%s', + perfname($dbname), $missing_days, $warning, $critical; + if (length $critical and $missing_days >= $critical) { + push @crit => $msg; + } + elsif (length $warning and $missing_days >= $warning) { + push @warn => $msg; + } + else { + push @ok => $msg; + } + } + if (0 == $found) { + add_ok msg('partman-premake-ok'); + } + elsif (1 == $found) { + add_unknown msg('no-match-db'); + } + elsif (@crit) { + add_critical join ' ' => @crit; + } + elsif (@warn) { + add_warning join ' ' => @warn; + } + else { + add_ok join ' ' => @ok; + } + } + + return; + +} ## end of check_partman_premake + +sub check_pgbouncer_checksum { + + ## Verify the checksum of all pgbouncer settings + ## Supports: Nagios, MRTG + ## Not that the connection will be done on the pgbouncer database + ## One of warning or critical must be given (but not both) + ## It should run one time to find out the expected checksum + ## You can use --critical="0" to find out the checksum + ## You can include or exclude settings as well + ## Example: + ## check_postgres_pgbouncer_checksum --critical="4e7ba68eb88915d3d1a36b2009da4acd" + + my ($warning, $critical) = validate_range({type => 'checksum', onlyone => 1}); + + eval { + require Digest::MD5; + }; if ($@) { ndie msg('checksum-nomd'); } @@ -5666,6 +6882,107 @@ sub check_pgbouncer_checksum { } ## end of check_pgbouncer_checksum +sub check_pgbouncer_maxwait { + + ## Check how long the first (oldest) client in queue has waited, in + ## seconds. + ## Supports: Nagios, MRTG + ## Warning and critical are time limits - defaults to seconds + ## Valid units: s[econd], m[inute], h[our], d[ay] + ## All above may be written as plural as well (e.g. "2 hours") + ## Can also ignore databases with exclude and limit with include + + my $arg = shift || {}; + + my ($warning, $critical) = validate_range + ({ + type => 'time', + }); + + ## Grab information from the pg_stat_activity table + ## Since we clobber old info on a qtime "tie", use an ORDER BY + $SQL = qq{SHOW POOLS}; + + my $info = run_command($SQL, { regex => qr{\d+}, emptyok => 1 } ); + + ## Default values for information gathered + my ($maxwait, $database, $user, $cl_active, $cl_waiting) = + (0,'?','?',0,0); + + for $db (@{$info->{db}}) { + + ## Parse the psql output and gather stats from the winning row + ## Read in and parse the psql output + my $skipped = 0; + ROW: for my $r (@{$db->{slurp}}) { + + ## Apply --exclude and --include arguments to the database name + if (skip_item($r->{database})) { + $skipped++; + next ROW; + } + + ## Assign stats if we have a new winner + if ($r->{maxwait} > $maxwait) { + $database = $r->{database}; + $user = $r->{user}; + $cl_active = $r->{cl_active}; + $cl_waiting = $r->{cl_waiting}; + $maxwait = $r->{maxwait}; + } + } + + ## We don't really care why things matches as far as the final output + ## But it's nice to report what we can + if ($database eq '?') { + $MRTG and do_mrtg({one => 0, msg => 'No rows'}); + $db->{perf} = "0;$warning;$critical"; + + if ($skipped) { + add_ok msg('pgb-maxwait-skipped', $skipped); + } + else { + add_ok msg('pgb-maxwait-nomatch', $maxwait); + } + return; + } + + ## Details on who the offender was + my $whodunit = sprintf q{%s:%s %s:%s cl_active:%s cl_waiting:%s}, + msg('database'), + $database, + msg('username'), + $user, + $cl_active, + $cl_waiting; + + $MRTG and do_mrtg({one => $maxwait, msg => "$whodunit"}); + + $db->{perf} .= sprintf q{'%s'=%s;%s;%s}, + $whodunit, + $maxwait, + $warning, + $critical; + + my $m = msg('pgb-maxwait-msg', $maxwait); + my $msg = sprintf '%s (%s)', $m, $whodunit; + + if (length $critical and $maxwait >= $critical) { + add_critical $msg; + } + elsif (length $warning and $maxwait >= $warning) { + add_warning $msg; + } + else { + add_ok $msg; + } + } + + return; + + +} ## end of check_pgbouncer_maxwait + sub check_pgbouncer_backends { ## Check the number of connections to pgbouncer compared to @@ -5752,7 +7069,7 @@ sub check_pgbouncer_backends { $nwarn = $limit-$w2; } elsif ($w3) { - $nwarn = (int $w2*$limit/100) + $nwarn = (int $w2*$limit/100); } if (! skip_item($r->{database})) { @@ -5843,6 +7160,7 @@ sub check_pgb_pool { $statsmsg{$i->{database}} = msg('pgbouncer-pool', $i->{database}, $stat, $i->{$stat}); next; } + $db->{perf} = sprintf ' %s=%s;%s;%s', $i->{database}, $i->{$stat}, $warning, $critical; if ($critical and $i->{$stat} >= $critical) { add_critical $msg; @@ -5948,7 +7266,7 @@ sub check_query_runtime { ## Valid units: s[econd], m[inute], h[our], d[ay] ## Does a "EXPLAIN ANALYZE SELECT COUNT(1) FROM xyz" ## where xyz is given by the option --queryname - ## This could also be a table or a function, or course, but must be a + ## This could also be a table or a function, or course, but must be a ## single word. If a function, it must be empty (with "()") ## Examples: ## --warning="100s" --critical="120s" --queryname="speedtest1" @@ -6019,7 +7337,7 @@ sub check_query_time { sub check_relation_size { - my $relkind = shift || 'relation'; + my ($sizefct, $relkinds) = @_; ## Check the size of one or more relations ## Supports: Nagios, MRTG @@ -6034,25 +7352,31 @@ sub check_relation_size { my ($warning, $critical) = validate_range({type => 'size'}); + ## no critic $SQL = sprintf q{ -SELECT pg_relation_size(c.oid) AS rsize, - pg_size_pretty(pg_relation_size(c.oid)) AS psize, +SELECT pg_%1$s_size(c.oid) AS rsize, + pg_size_pretty(pg_%1$s_size(c.oid)) AS psize, relkind, relname, nspname -FROM pg_class c, pg_namespace n WHERE (relkind = %s) AND n.oid = c.relnamespace +FROM pg_class c JOIN pg_namespace n ON (c.relnamespace = n.oid) +WHERE relkind IN (%2$s) }, - $relkind eq 'table' ? q{'r'} - : $relkind eq 'index' ? q{'i'} - : q{'r' OR relkind = 'i'}; + $sizefct, ## no critic + join (',', map { "'$_'" } split (//, $relkinds)); ## no critic + ## use critic if ($opt{perflimit}) { $SQL .= " ORDER BY 1 DESC LIMIT $opt{perflimit}"; } if ($USERWHERECLAUSE) { - $SQL =~ s/ WHERE/, pg_user u WHERE u.usesysid=c.relowner$USERWHERECLAUSE AND/; + $SQL =~ s/WHERE/JOIN pg_roles r ON (c.relowner = r.oid) WHERE/; + $SQL .= $USERWHERECLAUSE; } - my $info = run_command($SQL, {emptyok => 1}); + my $SQL8 = $SQL; + $SQL8 =~ s/pg_table_size/pg_relation_size/g; # 8.4 and earlier + + my $info = run_command($SQL, {emptyok => 1, version => [ "<9.0 $SQL8" ] }); my $found = 0; for $db (@{$info->{db}}) { @@ -6074,7 +7398,7 @@ sub check_relation_size { my $nicename = $kind eq 'r' ? "$schema.$name" : $name; $db->{perf} .= sprintf '%s%s=%sB;%s;%s', - $VERBOSE==1 ? "\n" : ' ', + 1 == $VERBOSE ? "\n" : ' ', perfname($nicename), $size, $warning, $critical; ($max=$size, $pmax=$psize, $kmax=$kind, $nmax=$name, $smax=$schema) if $size > $max; } @@ -6093,19 +7417,22 @@ sub check_relation_size { } my $msg; - if ($relkind eq 'relation') { - if ($kmax eq 'r') { + if ($action eq 'relation_size') { + if ($kmax =~ /[rt]/) { $msg = msg('relsize-msg-relt', "$smax.$nmax", $pmax); } else { - $msg = msg('relsize-msg-reli', $nmax, $pmax); + $msg = msg('relsize-msg-reli', "$smax.$nmax", $pmax); } } - elsif ($relkind eq 'table') { + elsif ($action =~ /table|total_relation/) { $msg = msg('relsize-msg-tab', "$smax.$nmax", $pmax); } + elsif ($action eq 'indexes_size') { + $msg = msg('relsize-msg-indexes', "$smax.$nmax", $pmax); + } else { - $msg = msg('relsize-msg-ind', $nmax, $pmax); + $msg = msg('relsize-msg-ind', "$smax.$nmax", $pmax); } if (length $critical and $max >= $critical) { add_critical $msg; @@ -6123,14 +7450,6 @@ sub check_relation_size { } ## end of check_relation_size -sub check_table_size { - return check_relation_size('table'); -} -sub check_index_size { - return check_relation_size('index'); -} - - sub check_replicate_row { ## Make an update on one server, make sure it propogates to others @@ -6153,7 +7472,7 @@ sub check_replicate_row { my ($table,$pk,$id,$col,$val1,$val2) = (@repinfo); ## Quote everything, just to be safe (e.g. columns named 'desc') - $table = qq{"$table"}; + $table =~ s/([^\.]+)/\"$1\"/g; $pk = qq{"$pk"}; $col = qq{"$col"}; @@ -6175,11 +7494,11 @@ sub check_replicate_row { if (!defined $sourcedb) { ndie msg('rep-norow', "$table.$col"); } - my $value1 = $info1->{db}[0]{slurp}[0]{c} || ''; + my $value1 = (defined($info1->{db}[0]{slurp}[0]{c})?$info1->{db}[0]{slurp}[0]{c}:''); my $numslaves = @{$info1->{db}} - 1; for my $d ( @{$info1->{db}}[1 .. $numslaves] ) { - my $value2 = $d->{slurp}[0]{c} || ''; + my $value2 = (defined($d->{slurp}[0]{c})?$d->{slurp}[0]{c}:''); if ($value1 ne $value2) { ndie msg('rep-notsame'); } @@ -6286,12 +7605,12 @@ sub check_same_schema { for my $phrase (split /[\s,]+/ => $item) { ## Can be plain (e.g. nouser) or regex based exclusion, e.g. nouser=bob - next if $phrase !~ /(\w+)=?\s*(.*)/o; + next if $phrase !~ /(\w+)=?\s*(.*)/; my ($name,$regex) = (lc $1,$2||''); ## Names are standardized with regards to plurals and casing - $name =~ s/([aeiou])s$/$1/o; - $name =~ s/s$//o; + $name =~ s/([aeiou])s$/$1/; + $name =~ s/s$//; if (! length $regex) { $filter{"$name"} = 1; @@ -6339,7 +7658,7 @@ sub check_same_schema { my $foo = $info->{db}[0]; my $version = $foo->{slurp}[0]{version}; - $version =~ /\D+(\d+\.\d+)(\S+)/i or die qq{Invalid version: $version\n}; + $version =~ /\D+(\d+\.?\d+)(\S+)/i or die qq{Invalid version: $version\n}; my ($full,$major,$revision) = ("$1$2",$1,$2); $revision =~ s/^\.//; $dbver{$num} = { @@ -6360,23 +7679,75 @@ sub check_same_schema { ## We also indicate which columns should be ignored when comparing, ## as well as which columns are of a 'list' nature my @catalog_items = ( - [user => 'usesysid', 'useconfig' ], - [language => 'laninline,lanplcallfoid,lanvalidator', '' ], - [operator => '', '' ], - [type => '', '' ], - [schema => '', '' ], - [function => 'source_checksum,prolang,prorettype', '' ], - [table => 'reltype,relfrozenxid,relminmxid,relpages, - reltuples,relnatts,relallvisible', '' ], - [view => 'reltype', '' ], - [sequence => 'reltype,log_cnt,relnatts,is_called', '' ], - [index => 'relpages,reltuples,indpred,indclass, - indexprs,indcheckxmin', '' ], - [trigger => '', '' ], - [constraint => 'conbin', '' ], - [column => 'atttypid,attnum,attbyval', '' ], + [user => 'usesysid', 'useconfig' ], + [language => 'laninline,lanplcallfoid,lanvalidator', '' ], + [cast => '', '' ], + [domain => '', '' ], + [operator => '', '' ], + [aggregate => '', '' ], + [comment => '', '' ], + [extension => '', '' ], + [operator => 'oprleft,oprright,oprresult,oprnegate,oprcom', '' ], + [type => 'typarray', '' ], + [schema => '', '' ], + [function => 'source_checksum,prolang,prorettype, + proargtypes,proallargtypes,provariadic, + proargdefaults', '' ], + [table => 'reltype,relfrozenxid,relminmxid,relpages, + reltuples,relnatts,relallvisible,relhaspkey', '' ], + [view => 'reltype', '' ], + [sequence => 'reltype,log_cnt,relnatts,is_called', '' ], + [index => 'relpages,reltuples,indpred,indclass,indexprs, + indcheckxmin,reltablespace,indkey', '' ], + [trigger => 'tgqual,tgconstraint,tgattr', '' ], + [constraint => 'conbin,conindid,conkey,confkey,confmatchtype', '' ], + [column => 'atttypid,attnum,attbyval,attndims', '' ], + [foreign_server => '', '' ], + [foreign_data_wrapper => '', '' ], + [foreign_table => '', '' ], ); + ## TODO: + ## operator class, conversion, tablespace, collation + + ## Allow the above list to be adjusted by exclusion: + if (exists $opt{skipobject}) { + my (%skiplist, %badlist); + for my $item (map { split /,\s*/ => lc $_ } @{ $opt{skipobject} }) { + if (grep { $_->[0] eq $item } @catalog_items) { + $skiplist{$item} = 1; + } + else { + $badlist{$item} = 1; + } + } + if (keys %badlist) { + ndie msg('ss-badobject', join $COMMA => sort keys %badlist); + } + + ## Shrink the original list to remove excluded items + @catalog_items = grep { ! exists $skiplist{$_->[0]} } @catalog_items; + } + + ## Allow the list to be adjusted by inclusion: + if (exists $opt{object}) { + my (%goodlist, %badlist); + for my $item (map { split /,\s*/ => lc $_ } @{ $opt{object} }) { + if (grep { $_->[0] eq $item } @catalog_items) { + $goodlist{$item} = 1; + } + else { + $badlist{$item} = 1; + } + } + if (keys %badlist) { + ndie msg('ss-badobject', join $COMMA => sort keys %badlist); + } + + ## Shrink the original list to only have the requested items + @catalog_items = grep { exists $goodlist{$_->[0]} } @catalog_items; + } + ## Where we store all the information, per-database my %thing; @@ -6436,13 +7807,9 @@ sub check_same_schema { $dbinfo->{$name} = find_catalog_info($name, $x, $dbver{$x}); } - ## TODO: - ## operator class, cast, aggregate, conversion, domain, tablespace, foreign tables - ## foreign server, wrapper, collation, extensions, roles? - ## Map the oid back to the user, for ease later on for my $row (values %{ $dbinfo->{user} }) { - $dbinfo->{useroid}{$row->{usesysid}} = $row->{usename}; + $dbinfo->{useroid}{$row->{usesysid}} = $row->{rolname}; } $thing{$x} = $dbinfo; @@ -6497,7 +7864,7 @@ sub check_same_schema { ## Set the total time $db->{totaltime} = sprintf '%.2f', tv_interval($start); - ## Before we outpu any results, rewrite the audit file if needed + ## Before we output any results, rewrite the audit file if needed ## We do this if we are reading from a saved file, ## and the "replace" argument is set if ($samedb and $opt{replace}) { @@ -6563,6 +7930,9 @@ sub check_same_schema { ## Show the list of the item, and a CSV of which databases have it and which don't my $isthere = join ', ' => sort { $a<=>$b } keys %{ $e->{$name}{isthere} }; my $nothere = join ', ' => sort { $a<=>$b } keys %{ $e->{$name}{nothere} }; + ## Extra information (e.g. tablename) may be stuffed into the hash value + (my $extra) = values %{ $e->{$name}{isthere} }; + $name .= " ($extra)" if defined $extra and length $extra > 1; $msg .= sprintf "%s\n %-*s %s\n %-*s %s\n", msg('ss-noexist', $pitem, $name), $maxsize, $msg_exists, @@ -6616,8 +7986,8 @@ sub check_same_schema { next if $one eq '' and $two eq '-'; } - ## If we are doing a historical comparison, skip some items - if ($samedb) { + ## If we are doing a historical comparison or checking asynchronous replicas, skip some items + if ($samedb or $opt{'assume-async'}) { if ($item eq 'sequence' and $col eq 'last_value') { next; @@ -6691,7 +8061,7 @@ sub check_same_schema { $msg .= $_ for @msg; } else { - ## No message means it was all filtered out, so we decrment the master count + ## No message means it was all filtered out, so we decrement the master count $opt{failcount}--; } } @@ -6737,6 +8107,13 @@ sub audit_filename { $filename .= ".$opt{suffix}"; } + ## Need to store in a separate directory? + my $adir = $opt{'audit-file-dir'}; + if (defined $adir) { + -d $adir or die qq{Cannot write to directory "$adir": $!\n}; + $filename = catfile($adir, $filename); + } + return $filename; } ## end of audit_filename @@ -6751,9 +8128,10 @@ sub write_audit_file { ## Create a connection information string my $row = $targetdb[0]; - my $conninfo = sprintf '%s%s%s%s', + my $conninfo = sprintf '%s%s%s%s%s', + defined $row->{dbservice} ? qq{service=$row->{dbservice} } : '', defined $row->{port} ? qq{port=$row->{port} } : '', - defined $row->{host} ? qq{host=$row->{host} } : '', + (defined $row->{host} and $row->{host} ne '<none>') ? qq{host=$row->{host} } : '', defined $row->{dbname} ? qq{dbname=$row->{dbname} } : '', defined $row->{dbuser} ? qq{user=$row->{dbuser} } : ''; @@ -6763,10 +8141,10 @@ sub write_audit_file { print {$fh} "## PG version: $arg->{pgversion}\n"; printf {$fh} "## Created: %s\n", scalar localtime(); print {$fh} "## Connection: $conninfo\n"; - print {$fh} "## Database name: $row->{dbname}\n"; - print {$fh} "## Host: $row->{host}\n"; - print {$fh} "## Port: $row->{port}\n"; - print {$fh} "## User: $row->{dbuser}\n"; + print {$fh} "## Database name: $row->{dbname}\n" if defined $row->{dbname}; + print {$fh} "## Host: $row->{host}\n" if defined $row->{host} and $row->{host} ne '<none>'; + print {$fh} "## Port: $row->{port}\n" if defined $row->{port}; + print {$fh} "## User: $row->{dbuser}\n" if defined $row->{dbuser}; if ($arg->{same_schema}) { print {$fh} "## Start of same_schema information:\n"; { @@ -6868,25 +8246,33 @@ sub schema_item_exists { for my $name (sort keys %{ $itemhash->{$db1}{$item_class} }) { ## Can exclude by 'filter' based regex - next if grep { $name eq $_ } @$exclude_regex; + next if grep { $name =~ $_ } @$exclude_regex; if (! exists $itemhash->{$db2}{$item_class}{$name}) { - ## Special exception for columns: do not add if the table is non-existent - if ($item_class eq 'column') { - (my $tablename = $name) =~ s/(.+)\..+/$1/; - next if ! exists $itemhash->{$db2}{table}{$tablename}; + ## Skip if the schema does not match (and we have at least one schema, indicating lack of 'noschema') + if ($item_class ne 'schema') { + my $it = $itemhash->{$db1}{$item_class}{$name}; + next if exists $it->{schemaname} and keys %{ $itemhash->{$db1}{schema} } and ! exists $itemhash->{$db2}{schema}{ $it->{schemaname} }; } - ## Special exception for triggers: do not add if the table is non-existent - if ($item_class eq 'trigger') { + ## Skip if this item has a table, but the table does not match (or if 'notables' is set) + if ($item_class ne 'table') { my $it = $itemhash->{$db1}{$item_class}{$name}; - my $tablename = "$it->{tschema}.$it->{tname}"; - next if ! exists $itemhash->{$db2}{table}{$tablename}; + next if exists $it->{tablename} + and exists $itemhash->{$db1}{table}{ $it->{tablename} } + and ! exists $itemhash->{$db2}{table}{ $it->{tablename} }; + next if exists $it->{tablename} and $opt{filtered}{notable}; } - $nomatch{$name}{isthere}{$db1} = 1; - $nomatch{$name}{nothere}{$db2} = 1; + my $one = '1'; + + if ($item_class eq 'index') { + $one = $itemhash->{$db1}{$item_class}{$name}{tablename}; + } + + $nomatch{$name}{isthere}{$db1} = $one; + $nomatch{$name}{nothere}{$db2} = $one; } } } @@ -6944,7 +8330,7 @@ sub schema_item_differences { for my $name (sort keys %{ $itemhash->{$db1}{$item_class} }) { ## Can exclude by 'filter' based regex - next if grep { $name eq $_ } @$exclude_regex; + next if grep { $name =~ $_ } @$exclude_regex; ## This case has already been handled: next if ! exists $itemhash->{$db2}{$item_class}{$name}; @@ -6972,6 +8358,9 @@ sub schema_item_differences { ## Skip certain known numeric fields that have text versions: next if $col =~ /.(?:namespace|owner|filenode|oid|relid)$/; + ## We may want to skip the "data" of the sequences + next if $opt{skipsequencevals} and $item_class eq 'sequence'; + ## If not a list, just report on the exact match here and move on: if (! exists $lists->{$col} and $col !~ /.acl$/) { $nomatch{$name}{coldiff}{$col}{$db1} = $one->{$col}; @@ -7070,6 +8459,16 @@ sub find_catalog_info { ## The version information my $dbver = shift or die; + ## Cannot check extensions if we are before 9.4 + if ('extension' eq $type and $dbver->{major} < 9.4) { + return {}; + } + + ## Foreign tables arrived in 9.1 + if ($type =~ /foreign/ and $dbver->{major} < 9.1) { + return {}; + } + ## The SQL we use my $SQL = $ci->{SQL} or die "No SQL found for type '$type'\n"; @@ -7079,6 +8478,60 @@ sub find_catalog_info { $SQL = $ci->{SQL2}; } } + if ($type eq 'trigger' and $dbver->{major} <= 8.4) { + $SQL =~ s/t.tgconstrindid = 0 AND //; + } + if ($type eq 'sequence' and $dbver->{major} >= 10) { + $SQL = $ci->{SQL10}; + delete $ci->{innerSQL}; + } + + if ($type eq 'comment' and $dbver->{major} < 11) { + $SQL =~ s/prokind='a'/proisagg/g; + $SQL =~ s/\Qprokind IN ('f','p')/NOT proisagg/g; + } + + if ($dbver->{major} < 9.1) { + if ($SQL =~ /FORMAT/) { + #warn $SQL; + $SQL =~ s{FORMAT\('(.+?)', (.+)\) AS name}{ + my ($format, @args) = ($1, split /, / => $2); + $format =~ s/ (\w+) / || ' $1 ' || /g; + $format =~ s/([,.\(\)])/ || '$1' || /g; + $format =~ s/%s/shift @args/ge; + $format =~ s/\|\|\s*$//; + "$format As name"; + }ge; + + #warn $SQL; + } + } + + + ## The regnamespace shortcut arrived with version 9.5 + if ($dbver->{major} < 9.5) { + $SQL =~ s{(\w+)::regnamespace}{(SELECT nspname FROM pg_namespace WHERE oid=$1)}g; + } + + ## The string_agg function appeared in 9.0. For now, we return nothing. + return {} if $dbver->{major} < 9.0 and $SQL =~ /string_agg/; + + ## Many new tables showed up in version 9.4 + if ($dbver->{major} < 9.4) { + $SQL =~ s{UNION ALL.+pg_collation}{}; + $SQL =~ s{UNION ALL.+pg_extension}{}; + $SQL =~ s{UNION ALL.+pg_event_trigger}{}; + } + + ## Tables appearing in version 9.5 + if ($dbver->{major} < 9.5) { + $SQL =~ s{UNION ALL.+pg_policy}{}; + } + + ## Tables appearing in version 9.1 + if ($dbver->{major} < 9.1) { + $SQL =~ s{UNION ALL.+pg_extension}{}; + } if (exists $ci->{exclude}) { if ('temp_schemas' eq $ci->{exclude}) { @@ -7146,8 +8599,8 @@ sub find_catalog_info { ## For columns, reduce the attnum to a simpler canonical form without holes if ($type eq 'column') { - if ($row->{tname} ne $last_table) { - $last_table = $row->{tname}; + if ($row->{tablename} ne $last_table) { + $last_table = $row->{tablename}; $colnum = 1; } $row->{column_number} = $colnum++; @@ -7160,7 +8613,7 @@ sub find_catalog_info { if (exists $ci->{innerSQL}) { if ($type eq 'sequence') { - ## If this is a sequence, we want to grab them all at once to reduce + ## If this is a sequence, we want to grab them all at once to reduce ## the amount of round-trips we do with 'SELECT * FROM seqname' if (! exists $opt{seqinfoss}{$dbnum}) { $SQL = q{SELECT quote_ident(nspname)||'.'||quote_ident(relname) AS sname } @@ -7208,6 +8661,12 @@ sub check_sequence { ## Warning and critical are percentages ## Can exclude and include sequences + my $skipcycled = $opt{'skipcycled'} || 0; + my $percsql = 'ROUND(used/slots*100)'; + if($skipcycled) { + $percsql = 'CASE WHEN cycle THEN 0 ELSE ' . $percsql . ' END'; + } + my ($warning, $critical) = validate_range ({ type => 'percent', @@ -7262,9 +8721,18 @@ sub check_sequence { WHERE nspname !~ '^pg_temp.*' ORDER BY nspname, seqname, typname }; + my $SQL10 = qq{ +SELECT seqname, last_value, slots, used, $percsql AS percent, + CASE WHEN slots < used THEN 0 ELSE slots - used END AS numleft +FROM ( + SELECT quote_ident(schemaname)||'.'||quote_ident(sequencename) AS seqname, COALESCE(last_value,min_value) AS last_value, + cycle, + CEIL((max_value-min_value::NUMERIC+1)/increment_by::NUMERIC) AS slots, + CEIL((COALESCE(last_value,min_value)-min_value::NUMERIC+1)/increment_by::NUMERIC) AS used +FROM pg_sequences) foo}; ## use critic - my $info = run_command($SQL, {regex => qr{\w}, emptyok => 1} ); + my $info = run_command($SQL, {regex => qr{\w}, emptyok => 1, version => ['>9.6 SELECT 1']} ); # actual SQL10 is executed below my $MAXINT2 = 32767; my $MAXINT4 = 2147483647; @@ -7278,42 +8746,54 @@ sub check_sequence { my %seqinfo; my %seqperf; my $multidb = @{$info->{db}} > 1 ? "$db->{dbname}." : ''; - for my $r (@{$db->{slurp}}) { + my @seq_sql; + for my $r (@{$db->{slurp}}) { # for each sequence, create SQL command to inspect it + next if ($db->{version} >= 10); # TODO: skip loop entirely my ($schema, $seq, $seqname, $typename) = @$r{qw/ nspname seqname safename typname /}; next if skip_item($seq); my $maxValue = $typename eq 'int2' ? $MAXINT2 : $typename eq 'int4' ? $MAXINT4 : $MAXINT8; - $SQL = qq{ -SELECT last_value, slots, used, ROUND(used/slots*100) AS percent, + my $seqname_l = $seqname; + $seqname_l =~ s/'/''/g; # SQL literal quoting (name is already identifier-quoted) + push @seq_sql, qq{ +SELECT '$seqname_l' AS seqname, last_value, slots, used, $percsql AS percent, CASE WHEN slots < used THEN 0 ELSE slots - used END AS numleft FROM ( SELECT last_value, - CEIL((LEAST(max_value, $maxValue)-min_value::numeric+1)/increment_by::NUMERIC) AS slots, - CEIL((last_value-min_value::numeric+1)/increment_by::NUMERIC) AS used + is_cycled AS cycle, + CEIL((LEAST(max_value, $maxValue)-min_value::NUMERIC+1)/increment_by::NUMERIC) AS slots, + CEIL((last_value-min_value::NUMERIC+1)/increment_by::NUMERIC) AS used FROM $seqname) foo }; - - my $seqinfo = run_command($SQL, { target => $db }); - my $r2 = $seqinfo->{db}[0]{slurp}[0]; - my ($last, $slots, $used, $percent, $left) = @$r2{qw/ last_value slots used percent numleft / }; - if (! defined $last) { - ndie msg('seq-die', $seqname); - } - my $msg = msg('seq-msg', $seqname, $percent, $left); - my $nicename = perfname("$multidb$seqname"); - $seqperf{$percent}{$seqname} = [$left, " $nicename=$percent%;$w%;$c%"]; - if ($percent >= $maxp) { - $maxp = $percent; - if (! exists $opt{perflimit} or $limit++ < $opt{perflimit}) { - push @{$seqinfo{$percent}} => $MRTG ? [$seqname,$percent,$slots,$used,$left] : $msg; + } + if ($db->{version} >= 10) { + @seq_sql = ($SQL10); # inject PG10 query here (TODO: pull this out of loops) + } + # Use UNION ALL to query multiple sequences at once, however if there are too many sequences this can exceed + # maximum argument length; so split into chunks of 200 sequences or less and iterate over them. + while (my @seq_sql_chunk = splice @seq_sql, 0, 200) { + my $seqinfo = run_command(join("\nUNION ALL\n", @seq_sql_chunk), { target => $db }); # execute all SQL commands at once + for my $r2 (@{$seqinfo->{db}[0]{slurp}}) { # now look at all results + my ($seqname, $last, $slots, $used, $percent, $left) = @$r2{qw/ seqname last_value slots used percent numleft / }; + if (! defined $last) { + ndie msg('seq-die', $seqname); } - } - next if $MRTG; + my $msg = msg('seq-msg', $seqname, $percent, $left); + my $nicename = perfname("$multidb$seqname"); + $seqperf{$percent}{$seqname} = [$left, " $nicename=$percent%;$w%;$c%"]; + if ($percent >= $maxp) { + $maxp = $percent; + if (! exists $opt{perflimit} or $limit++ < $opt{perflimit}) { + push @{$seqinfo{$percent}} => $MRTG ? [$seqname,$percent,$slots,$used,$left] : $msg; + } + } + next if $MRTG; - if (length $critical and $percent >= $c) { - push @crit => $msg; - } - elsif (length $warning and $percent >= $w) { - push @warn => $msg; + if (length $critical and $percent >= $c) { + push @crit => $msg; + } + elsif (length $warning and $percent >= $w) { + push @warn => $msg; + } } } if ($MRTG) { @@ -7587,13 +9067,15 @@ sub check_txn_idle { ## We don't GROUP BY because we want details on every connection ## Someday we may even break things down by database - my ($SQL2, $SQL3); + my ($SQL2, $SQL3, $SQL4); if ($type ne 'qtime') { $SQL = q{SELECT datname, datid, procpid AS pid, usename, client_addr, xact_start, current_query AS current_query, '' AS state, }. q{CASE WHEN client_port < 0 THEN 0 ELSE client_port END AS client_port, }. qq{COALESCE(ROUND(EXTRACT(epoch FROM now()-$start)),0) AS seconds }. qq{FROM pg_stat_activity WHERE ($clause)$USERWHERECLAUSE }. q{ORDER BY xact_start, query_start, procpid DESC}; + # Handle usename /rolname differences + $SQL =~ s/rolname/usename/; ## Craft an alternate version for old servers that do not have the xact_start column: ($SQL2 = $SQL) =~ s/xact_start/query_start AS xact_start/; $SQL2 =~ s/BY xact_start,/BY/; @@ -7604,6 +9086,9 @@ sub check_txn_idle { qq{COALESCE(ROUND(EXTRACT(epoch FROM now()-$start)),0) AS seconds }. qq{FROM pg_stat_activity WHERE ($clause)$USERWHERECLAUSE }. q{ORDER BY query_start, procpid DESC}; + # Handle usename /rolname differences + $SQL =~ s/rolname/usename/; + $SQL2 =~ s/rolname/usename/; } ## Craft an alternate version for new servers which do not have procpid and current_query is split @@ -7612,8 +9097,12 @@ sub check_txn_idle { $SQL3 =~ s/current_query NOT LIKE '<IDLE>%'/(state NOT LIKE 'idle%' OR state IS NULL)/; # query_time $SQL3 =~ s/current_query/query/g; $SQL3 =~ s/'' AS state/state AS state/; + $SQL3 =~ s/query_start/state_change/g; - my $info = run_command($SQL, { emptyok => 1 , version => [ "<8.3 $SQL2", ">9.1 $SQL3" ] } ); + ## For Pg 10 and above, consider only client backends + ($SQL4 = $SQL3) =~ s/ WHERE / WHERE backend_type = 'client backend' AND /; + + my $info = run_command($SQL, { emptyok => 1 , version => [ "<8.3 $SQL2", ">9.6 $SQL4", ">9.1 $SQL3" ] } ); ## Extract the first entry $db = $info->{db}[0]; @@ -7641,23 +9130,23 @@ sub check_txn_idle { next if skip_item($r->{datname}); ## We do a lot of filtering based on the current_query or state in 9.2+ - my $cq = $r->{query} || $r->{current_query}; - my $st = $r->{state} || ''; + my $cq = defined($r->{query}) ? $r->{query} : $r->{current_query}; + my $st = defined($r->{state}) ? $r->{state} : ''; ## Return unknown if we cannot see because we are a non-superuser - if ($cq =~ /insufficient/o) { + if ($cq =~ /insufficient/) { add_unknown msg('psa-nosuper'); return; } ## Return unknown if stats_command_string / track_activities is off - if ($cq =~ /disabled/o or $cq =~ /<command string not enabled>/) { + if ($st =~ /disabled/ or $cq =~ /<command string not enabled>/) { add_unknown msg('psa-disabled'); return; } ## Detect other cases where pg_stat_activity is not fully populated - if ($type ne 'qtime' and length $r->{xact_start} and $r->{xact_start} !~ /\d/o) { + if ($type ne 'qtime' and length $r->{xact_start} and $r->{xact_start} !~ /\d/) { add_unknown msg('psa-noexact'); return; } @@ -7683,6 +9172,7 @@ sub check_txn_idle { ## Extract the seconds to avoid typing out the hash each time my $max = $maxr->{seconds}; + $max = 0 if $max eq '-0'; ## See if we have a minimum number of matches my $base_count = $wcount || $ccount; @@ -7701,7 +9191,7 @@ sub check_txn_idle { $maxr->{client_addr} eq '' ? '' : (sprintf ' %s:%s', msg('address'), $maxr->{client_addr}), ($maxr->{client_port} eq '' or $maxr->{client_port} < 1) ? '' : (sprintf ' %s:%s', msg('port'), $maxr->{client_port}), - msg('query'), $maxr->{query} || $maxr->{current_query}; + msg('query'), $maxr->{query} // $maxr->{current_query}; } ## For MRTG, we can simply exit right now @@ -7767,7 +9257,7 @@ sub check_txn_idle { sub check_txn_time { - ## This is the same as check_txn_idle, but we want where the + ## This is the same as check_txn_idle, but we want where the ## transaction start time is not null check_txn_idle('txntime', @@ -7787,7 +9277,7 @@ sub check_txn_wraparound { ## Supports: Nagios, MRTG ## Warning and critical are the number of transactions performed ## Thus, anything *over* that number will trip the alert - ## See: http://www.postgresql.org/docs/current/static/routine-vacuuming.html#VACUUM-FOR-WRAPAROUND + ## See: https://www.postgresql.org/docs/current/static/routine-vacuuming.html#VACUUM-FOR-WRAPAROUND ## It makes no sense to run this more than once on the same cluster my ($warning, $critical) = validate_range @@ -7849,10 +9339,10 @@ sub check_version { ## or the major, minor, and revision (e.g. 8.2.4 or even 8.3beta4) if ($MRTG) { - if (!exists $opt{mrtg} or $opt{mrtg} !~ /^\d+\.\d+/) { + if (!exists $opt{mrtg} or $opt{mrtg} !~ /^\d+\.?\d+/) { ndie msg('version-badmrtg'); } - if ($opt{mrtg} =~ /^\d+\.\d+$/) { + if ($opt{mrtg} =~ /^\d+\.?\d+$/) { $opt{critical} = $opt{mrtg}; } else { @@ -7862,13 +9352,13 @@ sub check_version { my ($warning, $critical) = validate_range({type => 'version', forcemrtg => 1}); - my ($warnfull, $critfull) = (($warning =~ /^\d+\.\d+$/ ? 0 : 1),($critical =~ /^\d+\.\d+$/ ? 0 : 1)); + my ($warnfull, $critfull) = (($warning =~ /^(?:1\d+|[789]\.\d+)$/ ? 0 : 1),($critical =~ /^(?:1\d+|[789]\.\d+)$/ ? 0 : 1)); my $info = run_command('SELECT version() AS version'); for $db (@{$info->{db}}) { my $row = $db->{slurp}[0]; - if ($row->{version} !~ /((\d+\.\d+)(\w+|\.\d+))/o) { + if ($row->{version} !~ /((\d+\.?\d+)(\w+|\.\d+))/) { add_unknown msg('invalid-query', $row->{version}); next; } @@ -7928,10 +9418,15 @@ sub check_wal_files { my ($warning, $critical) = validate_range($arg); + my $lsfunc = $opt{lsfunc} || 'pg_ls_dir'; + my $lsargs = $opt{lsfunc} ? q{} : "'pg_xlog$subdir'"; + ## Figure out where the pg_xlog directory is - $SQL = qq{SELECT count(*) AS count FROM pg_ls_dir('pg_xlog$subdir') WHERE pg_ls_dir ~ E'^[0-9A-F]{24}$extrabit\$'}; ## no critic (RequireInterpolationOfMetachars) + $SQL = qq{SELECT count(*) AS count FROM $lsfunc($lsargs) WHERE $lsfunc ~ E'^[0-9A-F]{24}$extrabit\$'}; ## no critic (RequireInterpolationOfMetachars) + my $SQL10 = $opt{lsfunc} ? $SQL : + qq{SELECT count(*) AS count FROM pg_ls_waldir() WHERE name ~ E'^[0-9A-F]{24}$extrabit\$'}; ## no critic (RequireInterpolationOfMetachars) - my $info = run_command($SQL, {regex => qr[\d] }); + my $info = run_command($SQL, {regex => qr[\d], version => [">9.6 $SQL10"] }); my $found = 0; for $db (@{$info->{db}}) { @@ -7959,8 +9454,6 @@ sub check_wal_files { } ## end of check_wal_files - - =pod =encoding utf8 @@ -7969,7 +9462,7 @@ =head1 NAME B<check_postgres.pl> - a Postgres monitoring script for Nagios, MRTG, Cacti, and others -This documents describes check_postgres.pl version 2.21.0 +This documents describes check_postgres.pl version 2.26.0 =head1 SYNOPSIS @@ -7991,7 +9484,7 @@ =head1 SYNOPSIS ## There are many other actions and options, please keep reading. The latest news and documentation can always be found at: - http://bucardo.org/check_postgres/ + https://bucardo.org/check_postgres/ =head1 DESCRIPTION @@ -8106,7 +9599,7 @@ =head1 DATABASE CONNECTION OPTIONS =item B<--dbservice=NAME> The name of a service inside of the pg_service.conf file. Before version 9.0 of Postgres, this is -a global file, usually found in /etc/pg_service.conf. If you are using version 9.0 or higher of +a global file, usually found in F</etc/pg_service.conf>. If you are using version 9.0 or higher of Postgres, you can use the file ".pg_service.conf" in the home directory of the user running the script, e.g. nagios. @@ -8114,7 +9607,14 @@ =head1 DATABASE CONNECTION OPTIONS when using this option such as --dbservice="maindatabase sslmode=require" The documentation for this file can be found at -http://www.postgresql.org/docs/current/static/libpq-pgservice.html +L<https://www.postgresql.org/docs/current/static/libpq-pgservice.html> + +=item B<--role=ROLE> + +Provides the role to switch to after connecting to the database but before running the given check. +This provides the ability to have superuser privileges assigned to a role without LOGIN access for +the purposes of audit and other security considerations. Requires a local `psql` version 9.6 or +higher. =back @@ -8187,6 +9687,14 @@ =head1 OTHER OPTIONS postgres@db$./check_postgres.pl --action=checkpoint --datadir /var/lib/postgresql/8.3/main/ --assume-prod POSTGRES_CHECKPOINT OK: Last checkpoint was 72 seconds ago | age=72;;300 mode=MASTER +=item B<--assume-async> + +If specified, indicates that any replication between servers is asynchronous. +The option is only relevant for (C<symlink: check_postgres_same_schema>). + +Example: + postgres@db$./check_postgres.pl --action=same_schema --assume-async --dbhost=star,line + =item B<-h> or B<--help> Displays a help screen with a summary of all actions and options. @@ -8293,7 +9801,7 @@ =head1 OTHER OPTIONS =head1 ACTIONS -The script runs one or more actions. This can either be done with the --action +The action to be run is selected using the --action flag, or by using a symlink to the main file that contains the name of the action inside of it. For example, to run the action "timesync", you may either issue: @@ -8304,12 +9812,12 @@ =head1 ACTIONS check_postgres_timesync All the symlinks are created for you in the current directory -if use the option --symlinks +if use the option --symlinks: perl check_postgres.pl --symlinks If the file name already exists, it will not be overwritten. If the file exists -and is a symlink, you can force it to overwrite by using "--action=build_symlinks_force" +and is a symlink, you can force it to overwrite by using "--action=build_symlinks_force". Most actions take a I<--warning> and a I<--critical> option, indicating at what point we change from OK to WARNING, and what point we go to CRITICAL. Note that @@ -8321,8 +9829,8 @@ =head1 ACTIONS =head2 B<archive_ready> (C<symlink: check_postgres_archive_ready>) Checks how many WAL files with extension F<.ready> -exist in the F<pg_xlog/archive_status> directory, which is found -off of your B<data_directory>. This action must be run as a superuser, in order to access the +exist in the F<pg_xlog/archive_status> directory (PostgreSQL 10 and later: F<pg_wal/archive_status>), which is found +off of your B<data_directory>. If the I<--lsfunc> option is not used then this action must be run as a superuser, in order to access the contents of the F<pg_xlog/archive_status> directory. The minimum version to use this action is Postgres 8.1. The I<--warning> and I<--critical> options are simply the number of F<.ready> files in the F<pg_xlog/archive_status> directory. @@ -8332,9 +9840,26 @@ =head2 B<archive_ready> If the archive command fail, number of WAL in your F<pg_xlog> directory will grow until exhausting all the disk space and force PostgreSQL to stop immediately. -Example 1: Check that the number of ready WAL files is 10 or less on host "pluto" +To avoid running as a database superuser, a wrapper function around +C<pg_ls_dir()> should be defined as a superuser with SECURITY DEFINER, +and the I<--lsfunc> option used. This example function, if defined by +a superuser, will allow the script to connect as a normal user +I<nagios> with I<--lsfunc=ls_archive_status_dir> + + BEGIN; + CREATE FUNCTION ls_archive_status_dir() + RETURNS SETOF TEXT + AS $$ SELECT pg_ls_dir('pg_xlog/archive_status') $$ + LANGUAGE SQL + SECURITY DEFINER; + REVOKE ALL ON FUNCTION ls_archive_status_dir() FROM PUBLIC; + GRANT EXECUTE ON FUNCTION ls_archive_status_dir() to nagios; + COMMIT; - check_postgres_archive_ready --host=pluto --critical=10 +Example 1: Check that the number of ready WAL files is 10 or less on host "pluto", +using a wrapper function C<ls_archive_status_dir> to avoid the need for superuser permissions + + check_postgres_archive_ready --host=pluto --critical=10 --lsfunc=ls_archive_status_dir For MRTG output, reports the number of ready WAL files on line 1. @@ -8370,8 +9895,9 @@ =head2 B<backends> You can also filter the databases by use of the I<--include> and I<--exclude> options. See the L</"BASIC FILTERING"> section for more details. -To view only non-idle processes, you can use the I<--noidle> argument. Note that the -user you are connecting as must be a superuser for this to work properly. +To view only non-idle processes, you can use the I<--noidle> argument. Note that the user you are +running as (either connecting directly or switching via I<--role>) must be a superuser for this to +work properly. Example 1: Give a warning when the number of connections on host quirm reaches 120, and a critical if it reaches 150. @@ -8740,7 +10266,8 @@ =head2 B<disk_space> (C<symlink: check_postgres_disk_space>) Checks on the available physical disk space used by Postgres. This action requires that you have the executable "/bin/df" available to report on disk sizes, and it -also needs to be run as a superuser, so it can examine the B<data_directory> +also needs to be run as a superuser (either connecting directly or switching via I<--role>), +so it can examine the B<data_directory> setting inside of Postgres. The I<--warning> and I<--critical> options are given in either sizes or percentages or both. If using sizes, the standard unit types are allowed: bytes, kilobytes, gigabytes, megabytes, gigabytes, terabytes, or @@ -8754,7 +10281,7 @@ =head2 B<disk_space> B<log directory> - The disk that the log files are on. -B<WAL file directory> - The disk that the write-ahead logs are on (e.g. symlinked pg_xlog) +B<WAL file directory> - The disk that the write-ahead logs are on (e.g. symlinked pg_xlog or pg_wal) B<tablespaces> - Each tablespace that is on a separate disk. @@ -8767,7 +10294,7 @@ =head2 B<disk_space> check_postgres_disk_space --port=5432 --warning='90%' --critical='90%' -Example 2: Check that all file systems starting with /dev/sda are smaller than 10 GB and 11 GB (warning and critical) +Example 2: Check that all file systems starting with F</dev/sda> are smaller than 10 GB and 11 GB (warning and critical) check_postgres_disk_space --port=5432 --warning='10 GB' --critical='11 GB' --include="~^/dev/sda" @@ -8852,7 +10379,7 @@ =head2 B<hot_standby_delay> to it. The slave server must be in hot_standby (e.g. read only) mode, therefore the minimum version to use this action is Postgres 9.0. The I<--warning> and I<--critical> options are the delta between the xlog locations. Since these values are byte offsets in the WAL they should match the expected transaction volume -of your application to prevent false postives or negatives. +of your application to prevent false positives or negatives. The first "--dbname", "--host", and "--port", etc. options are considered the master; the second belongs to the slave. @@ -8864,7 +10391,7 @@ =head2 B<hot_standby_delay> form 'I<bytes> and I<time>', both conditions must be true for the threshold to be met. You must provide information on how to reach the databases by providing a comma separated list to the ---dbost and --dbport parameters, such as "--dbport=5432,5543". If not given, the action fails. +--dbhost and --dbport parameters, such as "--dbport=5432,5543". If not given, the action fails. Example 1: Warn a database with a local replica on port 5433 is behind on any xlog replay at all @@ -8878,16 +10405,35 @@ =head2 B<hot_standby_delay> check_hot_standby_delay --dbhost=master,replica1 --warning='1048576 and 2 min' --critical='16777216 and 10 min' +=head2 B<relation_size> + =head2 B<index_size> =head2 B<table_size> -=head2 B<relation_size> +=head2 B<indexes_size> + +=head2 B<total_relation_size> -(symlinks: C<check_postgres_index_size>, C<check_postgres_table_size>, and C<check_postgres_relation_size>) -The actions B<table_size> and B<index_size> are simply variations of the -B<relation_size> action, which checks for a relation that has grown too big. -Relations (in other words, tables and indexes) can be filtered with the +(symlinks: C<check_postgres_relation_size>, C<check_postgres_index_size>, +C<check_postgres_table_size>, C<check_postgres_indexes_size>, and +C<check_postgres_total_relation_size>) + +The actions B<relation_size> and B<index_size> check for a relation (table, +index, materialized view), respectively an index that has grown too big, using +the B<pg_relation_size()> function. + +The action B<table_size> checks tables and materialized views using +B<pg_table_size()>, i.e. including relation forks and TOAST table. + +The action B<indexes_size> checks tables and materialized views for +the size of the attached indexes using B<pg_indexes_size()>. + +The action B<total_relation_size> checks relations using +B<pg_total_relation_size()>, i.e. including relation forks, indexes and TOAST +table. + +Relations can be filtered with the I<--include> and I<--exclude> options. See the L</"BASIC FILTERING"> section for more details. Relations can also be filtered by the user that owns them, by using the I<--includeuser> and I<--excludeuser> options. @@ -8961,7 +10507,7 @@ =head2 B<last_autovacuum> Example 2: Same as above, but skip tables belonging to the users 'eve' or 'mallory' - check_postgres_last_vacuum --host=wormwood --warning='3d' --critical='7d' --excludeusers=eve,mallory + check_postgres_last_vacuum --host=wormwood --warning='3d' --critical='7d' --excludeuser=eve,mallory For MRTG output, returns (on the first line) the LEAST amount of time in seconds since a table was last vacuumed or analyzed. The fourth line returns the name of the database and name of the table. @@ -9009,6 +10555,22 @@ =head2 B<locks> For MRTG output, returns the number of locks on the first line, and the name of the database on the fourth line. +=head2 B<lockwait> + +(C<symlink: check_postgres_lockwait>) Check if there are blocking blocks and for how long. There is no +need to run this more than once per database cluster. Databases can be filtered +with the I<--include> and I<--exclude> options. See the L</"BASIC FILTERING"> section +for more details. + +The I<--warning> and I<--critical> options is time, +which represent the time for which the lock has been blocking. + +Example 1: Warn if a lock has been blocking for more than a minute, critcal if for more than 2 minutes + + check_postgres_lockwait --host=garrett --warning='1 min' --critical='2 min' + +For MRTG output, returns the number of blocked sessions. + =head2 B<logfile> (C<symlink: check_postgres_logfile>) Ensures that the logfile is in the expected location and is being logged to. @@ -9042,7 +10604,7 @@ =head2 B<new_version_bc> program is available. The current version is obtained by running C<bucardo_ctl --version>. If a major upgrade is available, a warning is returned. If a revision upgrade is available, a critical is returned. (Bucardo is a master to slave, and master to master -replication system for Postgres: see http://bucardo.org for more information). +replication system for Postgres: see L<https://bucardo.org/> for more information). See also the information on the C<--get_method> option. =head2 B<new_version_box> @@ -9052,7 +10614,7 @@ =head2 B<new_version_box> If a major upgrade is available, a warning is returned. If a revision upgrade is available, a critical is returned. (boxinfo is a program for grabbing important information from a server and putting it into a HTML format: see -http://bucardo.org/wiki/boxinfo for more information). See also the information on +L<https://bucardo.org/Boxinfo/> for more information). See also the information on the C<--get_method> option. =head2 B<new_version_cp> @@ -9081,9 +10643,15 @@ =head2 B<new_version_tnm> C<tail_n_mail --version>. If a major upgrade is available, a warning is returned. If a revision upgrade is available, a critical is returned. (tail_n_mail is a log monitoring tool that can send mail when interesting events appear in your Postgres logs. -See: http://bucardo.org/wiki/Tail_n_mail for more information). +See: L<https://bucardo.org/tail_n_mail/> for more information). See also the information on the C<--get_method> option. +=head2 B<partman_premake> + +(C<symlink: check_postgres_partman_premake>) Checks if all partitions that +B<pg_parman>'s maintenance routine should have created are actually present. +Monthly and daily intervals are supported. + =head2 B<pgb_pool_cl_active> =head2 B<pgb_pool_cl_waiting> @@ -9134,9 +10702,9 @@ =head2 B<pgbouncer_backends> of the I<--include> and I<--exclude> options. See the L</"BASIC FILTERING"> section for more details. -To view only non-idle processes, you can use the I<--noidle> argument. Note -that the user you are connecting as must be a superuser for this to work -properly. +To view only non-idle processes, you can use the I<--noidle> argument. Note that the user you are +running as (either connecting directly or switching via I<--role>) must be a superuser for this to +work properly. Example 1: Give a warning when the number of connections on host quirm reaches 120, and a critical if it reaches 150. @@ -9183,6 +10751,30 @@ =head2 B<pgbouncer_checksum> checksum must be provided as the C<--mrtg> argument. The fourth line always gives the current checksum. +=head2 B<pgbouncer_maxwait> + +(C<symlink: check_postgres_pgbouncer_maxwait>) Checks how long the first +(oldest) client in the queue has been waiting, in seconds. If this starts +increasing, then the current pool of servers does not handle requests quick +enough. Reason may be either overloaded server or just too small of a +pool_size setting in pbouncer config file. Databases can be filtered by use +of the I<--include> and I<--exclude> options. See the L</"BASIC FILTERING"> +section for more details. The values or the I<--warning> and I<--critical> +options are units of time, and must be provided (no default). Valid units are +'seconds', 'minutes', 'hours', or 'days'. Each may be written singular or +abbreviated to just the first letter. If no units are given, the units are +assumed to be seconds. + +This action requires Postgres 8.3 or better. + +Example 1: Give a critical if any transaction has been open for more than 10 +minutes: + + check_postgres_pgbouncer_maxwait -p 6432 -u pgbouncer --critical='10 minutes' + +For MRTG output, returns the maximum time in seconds a transaction has been +open on the first line. The fourth line gives the name of the database. + =head2 B<pgagent_jobs> (C<symlink: check_postgres_pgagent_jobs>) Checks that all the pgAgent jobs @@ -9258,7 +10850,7 @@ =head2 B<query_time> See the L</"USER NAME FILTERING"> section for more details. The values for the I<--warning> and I<--critical> options are amounts of -time, and default to '2 minutes' and '5 minutes' respectively. Valid units +time, and at least one must be provided (no defaults). Valid units are 'seconds', 'minutes', 'hours', or 'days'. Each may be written singular or abbreviated to just the first letter. If no units are given, the unit is assumed to be seconds. @@ -9323,13 +10915,23 @@ =head2 B<replicate_row> The maximum time is set to 4 minutes 30 seconds: if no replication has taken place in that long a time, an error is thrown. +=head2 B<replication_slots> + +(C<symlink: check_postgres_replication_slots>) Check the quantity of WAL retained for any replication +slots in the target database cluster. This is handy for monitoring environments where all WAL archiving +and replication is taking place over replication slots. + +Warning and critical are total bytes retained for the slot. E.g: + + check_postgres_replication_slots --port=5432 --host=yellow -warning=32M -critical=64M + +Specific named slots can be monitored using --include/--exclude + =head2 B<same_schema> (C<symlink: check_postgres_same_schema>) Verifies that two or more databases are identical as far as their -schema (but not the data within). This is particularly handy for making sure your slaves have not -been modified or corrupted in any way when using master to slave replication. Unlike most other -actions, this has no warning or critical criteria - the databases are either in sync, or are not. -If they are different, a detailed list of the differences is presented. +schema (but not the data within). Unlike most other actions, this has no warning or critical criteria - +the databases are either in sync, or are not. If they are different, a detailed list of the differences is presented. You may want to exclude or filter out certain differences. The way to do this is to add strings to the C<--filter> option. To exclude a type of object, use "noname", where 'name' is the type of @@ -9380,6 +10982,12 @@ =head2 B<same_schema> To replace the old stored file with the new version, use the --replace argument. +If you need to write the stored file to a specific directory, use +the --audit-file-dir argument. + +To avoid false positives on value based checks caused by replication lag on +asynchronous replicas, use the I<--assume-async> option. + To enable snapshots at various points in time, you can use the "--suffix" argument to make the filenames unique to each run. See the examples below. @@ -9408,6 +11016,10 @@ =head2 B<same_schema> check_postgres_same_schema --dbname=cylon --suffix=daily --replace +Example 7: Verify that two databases on hosts star and line are the same, excluding value data (i.e. sequence last_val): + + check_postgres_same_schema --dbhost=star,line --assume-async + =head2 B<sequence> (C<symlink: check_postgres_sequence>) Checks how much room is left on all sequences in the database. @@ -9415,7 +11027,8 @@ =head2 B<sequence> The I<--warning> and I<--critical> options should be expressed as percentages. The default values are B<85%> for the warning and B<95%> for the critical. You may use --include and --exclude to control which sequences are to be checked. Note that this check does account for unusual B<minvalue> -and B<increment by> values, but does not care if the sequence is set to cycle or not. +and B<increment by> values. By default it does not care if the sequence is set to cycle or not, +and by passing I<--skipcycled> sequenced set to cycle are reported with 0% usage. The output for Nagios gives the name of the sequence, the percentage used, and the number of 'calls' left, indicating how many more times nextval can be called on that sequence before running into @@ -9502,13 +11115,18 @@ =head2 B<txn_idle> section below for more details. The I<--warning> and I<--critical> options are given as units of time, signed -integers, or integers for units of time, and both must be provided (there are +integers, or integers for units of time, and at least one must be provided (there are no defaults). Valid units are 'seconds', 'minutes', 'hours', or 'days'. Each may be written singular or abbreviated to just the first letter. If no units are given and the numbers are unsigned, the units are assumed to be seconds. This action requires Postgres 8.3 or better. +As of PostgreSQL 10, you can just GRANT I<pg_read_all_stats> to an unprivileged user account. In +all earlier versions, superuser privileges are required to see the queries of all users in the +system; UNKNOWN is returned if queries cannot be checked. To only include queries by the connecting +user, use I<--includeuser>. + Example 1: Give a warning if any connection has been idle in transaction for more than 15 seconds: check_postgres_txn_idle --port=5432 --warning='15 seconds' @@ -9536,7 +11154,7 @@ =head2 B<txn_time> See the L</"USER NAME FILTERING"> section for more details. The values or the I<--warning> and I<--critical> options are units of time, and -must be provided (no default). Valid units are 'seconds', 'minutes', 'hours', +at least one must be provided (no default). Valid units are 'seconds', 'minutes', 'hours', or 'days'. Each may be written singular or abbreviated to just the first letter. If no units are given, the units are assumed to be seconds. @@ -9560,7 +11178,7 @@ =head2 B<txn_wraparound> If either option is not given, the default values of 1.3 and 1.4 billion are used. There is no need to run this command more than once per database cluster. For a more detailed discussion of what this number represents and what to do about it, please visit the page -L<http://www.postgresql.org/docs/current/static/routine-vacuuming.html#VACUUM-FOR-WRAPAROUND> +L<https://www.postgresql.org/docs/current/static/routine-vacuuming.html#VACUUM-FOR-WRAPAROUND> The warning and critical values can have underscores in the number for legibility, as Perl does. @@ -9595,9 +11213,9 @@ =head2 B<version> =head2 B<wal_files> -(C<symlink: check_postgres_wal_files>) Checks how many WAL files exist in the F<pg_xlog> directory, which is found +(C<symlink: check_postgres_wal_files>) Checks how many WAL files exist in the F<pg_xlog> directory (PostgreSQL 10 and later" F<pg_wal>), which is found off of your B<data_directory>, sometimes as a symlink to another physical disk for -performance reasons. This action must be run as a superuser, in order to access the +performance reasons. If the I<--lsfunc> option is not used then this action must be run as a superuser, in order to access the contents of the F<pg_xlog> directory. The minimum version to use this action is Postgres 8.1. The I<--warning> and I<--critical> options are simply the number of files in the F<pg_xlog> directory. What number to set this to will vary, but a general @@ -9609,9 +11227,26 @@ =head2 B<wal_files> create too many files. Ultimately, this will cause the disk they are on to run out of space, at which point Postgres will shut down. -Example 1: Check that the number of WAL files is 20 or less on host "pluto" +To avoid connecting as a database superuser, a wrapper function around +C<pg_ls_dir()> should be defined as a superuser with SECURITY DEFINER, +and the I<--lsfunc> option used. This example function, if defined by +a superuser, will allow the script to connect as a normal user +I<nagios> with I<--lsfunc=ls_xlog_dir> + + BEGIN; + CREATE FUNCTION ls_xlog_dir() + RETURNS SETOF TEXT + AS $$ SELECT pg_ls_dir('pg_xlog') $$ + LANGUAGE SQL + SECURITY DEFINER; + REVOKE ALL ON FUNCTION ls_xlog_dir() FROM PUBLIC; + GRANT EXECUTE ON FUNCTION ls_xlog_dir() to nagios; + COMMIT; + +Example 1: Check that the number of ready WAL files is 10 or less on host "pluto", +using a wrapper function C<ls_xlog_dir> to avoid the need for superuser permissions - check_postgres_wal_files --host=pluto --critical=20 + check_postgres_archive_ready --host=pluto --critical=10 --lsfunc=ls_xlog_dir For MRTG output, reports the number of WAL files on line 1. @@ -9678,6 +11313,10 @@ =head1 BASIC FILTERING --exclude='pg_catalog.' +Exclude all items in the 'pg_temp_nnn' per-session temporary schemas: + + --exclude=~^pg_temp_. + Exclude all items containing the letters 'ace', but allow the item 'faceoff': --exclude=~ace --include=faceoff @@ -9745,7 +11384,7 @@ =head1 FILES In addition to command-line configurations, you can put any options inside of a file. The file F<.check_postgresrc> in the current directory will be used if found. If not found, then the file -F<~/.check_postgresrc> will be used. Finally, the file /etc/check_postgresrc will be used if available. +F<~/.check_postgresrc> will be used. Finally, the file F</etc/check_postgresrc> will be used if available. The format of the file is option = value, one per line. Any line starting with a '#' will be skipped. Any values loaded from a check_postgresrc file will be overwritten by command-line options. All check_postgresrc files can be ignored by supplying a C<--no-checkpostgresrc> argument. @@ -9794,24 +11433,20 @@ =head1 DEVELOPMENT Development happens using the git system. You can clone the latest version by doing: - git clone git://bucardo.org/check_postgres.git + https://github.com/bucardo/check_postgres + git clone https://github.com/bucardo/check_postgres.git =head1 MAILING LIST Three mailing lists are available. For discussions about the program, bug reports, feature requests, and commit notices, send email to check_postgres@bucardo.org -https://mail.endcrypt.com/mailman/listinfo/check_postgres +L<https://bucardo.org/mailman/listinfo/check_postgres> A low-volume list for announcement of new versions and important notices is the 'check_postgres-announce' list: -https://mail.endcrypt.com/mailman/listinfo/check_postgres-announce - -Source code changes (via git-commit) are sent to the -'check_postgres-commit' list: - -https://mail.endcrypt.com/mailman/listinfo/check_postgres-commit +L<https://bucardo.org/mailman/listinfo/check_postgres-announce> =head1 HISTORY @@ -9819,23 +11454,176 @@ =head1 HISTORY =over 4 -=item B<Version 2.21.1> +=item B<Version 2.26.1> not yet released + + Add new action "lockwait" showing details of blocked queries + (Github user miraclesvenni) + [Github issue #154] + + Raise minimum version of Perl to 5.10.0 + + Allow commas in passwords via --dbpass for one-connection queries (Greg Sabino Mullane) [Github issue #133] + + Fix undefined variable error (Greg Sabino Mullane) [Github issue #141] + + +=item B<Version 2.26.0> Released April 3, 2023 + + Add new action "pgbouncer_maxwait" (Ruslan Kabalin) [Github pull #59] + + For the bloat check, add option to populate all known databases, + as well as includsion and exclusion regexes. (Giles Westwood) [Github pull #86] + + Add Partman premake check (Jens Wilke) [Github pull #196] + + Add --role flag to explicitly set the role of the user after connecting (David Christensen) + + Fix check_replication_slots on recently promoted servers (Christoph Berg) + + Allow the check_disk_space action to handle relative log_directory paths (jacksonfoz) [Github pull #174] + + Fix MINPAGES and MINIPAGES in the "check_bloat" action (Christoph Moench-Tegeder) [Github pull #82] + + Replace 'which' with 'command -v' (Christoph Berg) + + Fix check_replication_slots on recently promoted servers (Christoph Berg) + + Fix undefined variable warning (Michael van Bracht) [Github pull #158] + + In the tests, force log_destination to stderr (Christoph Moench-Tegeder) [Github pull #185] + + Add to docs how to exclude all items in the 'pg_temp_nnn' per-session temporary schemas (Michael Banck) + + Various fixes for the CI system (Emre Hasegeli) [Github pull #181] + + Various improvements to the tests (Christoph Berg, Emre Hasegeli) + + +=item B<Version 2.25.0> Released February 3, 2020 + + Allow same_schema objects to be included or excluded with --object and --skipobject + (Greg Sabino Mullane) + + Fix to allow mixing service names and other connection parameters for same_schema + (Greg Sabino Mullane) + + +=item B<Version 2.24.0> Released May 30, 2018 + + Support new_version_pg for PG10 + (Michael Pirogov) + + Option to skip CYCLE sequences in action sequence + (Christoph Moench-Tegeder) + + Output per-database perfdata for pgbouncer pool checks + (George Hansper) + + German message translations + (Holger Jacobs) + + Consider only client backends in query_time and friends + (David Christensen) + +=item B<Version 2.23.0> Released October 31, 2017 + + Support PostgreSQL 10. + (David Christensen, Christoph Berg) + + Change table_size to use pg_table_size() on 9.0+, i.e. include the TOAST + table size in the numbers reported. Add new actions indexes_size and + total_relation_size, using the respective pg_indexes_size() and + pg_total_relation_size() functions. All size checks will now also check + materialized views where applicable. + (Christoph Berg) + + Connection errors are now always critical, not unknown. + (Christoph Berg) + + New action replication_slots checking if logical or physical replication + slots have accumulated too much data + (Glyn Astill) + + Multiple same_schema improvements + (Glyn Astill) + + Add Spanish message translations + (Luis Vazquez) + + Allow a wrapper function to run wal_files and archive_ready actions as + non-superuser + (Joshua Elsasser) + + Add some defensive casting to the bloat query + (Greg Sabino Mullane) + + Invoke psql with option -X + (Peter Eisentraut) + + Update postgresql.org URLs to use https. + (Magnus Hagander) + + check_txn_idle: Don't fail when query contains 'disabled' word + (Marco Nenciarini) + + check_txn_idle: Use state_change instead of query_start. + (Sebastian Webber) + + check_hot_standby_delay: Correct extra space in perfdata + (Adrien Nayrat) + + Remove \r from psql output as it can confuse some regexes + (Greg Sabino Mullane) + + Sort failed jobs in check_pgagent_jobs for stable output. + (Christoph Berg) + +=item B<Version 2.22.0> June 30, 2015 + + Add xact timestamp support to hot_standby_delay. + Allow the hot_standby_delay check to accept xlog byte position or + timestamp lag intervals as thresholds, or even both at the same time. + (Josh Williams) + + Query all sequences per DB in parallel for action=sequence. + (Christoph Berg) + + Fix bloat check to use correct SQL depending on the server version. + (Adrian Vondendriesch) + + Show actual long-running query in query_time output + (Peter Eisentraut) Add explicit ORDER BY to the slony_status check to get the most lagged server. (Jeff Frost) - Declare POD encoding to be utf8. (Christoph Berg) + Improved multi-slave support in replicate_row. + (Andrew Yochum) + + Change the way tables are quoted in replicate_row. + (Glyn Astill) + + Don't swallow space before the -c flag when reporting errors + (Jeff Janes) + + Fix and extend hot_standby_delay documentation + (Michael Renner) + + Declare POD encoding to be utf8. + (Christoph Berg) =item B<Version 2.21.0> September 24, 2013 Fix issue with SQL steps in check_pgagent_jobs for sql steps which perform deletes (Rob Emery via github pull) - Install man page in section 1. (Peter Eisentraut, bug 53, github issue 26) + Install man page in section 1. + (Peter Eisentraut, bug 53, github issue 26) Order lock types in check_locks output to make the ordering predictable; setting SKIP_NETWORK_TESTS will skip the new_version tests; other minor test - suite fixes. (Christoph Berg) + suite fixes. + (Christoph Berg) Fix same_schema check on 9.3 by ignoring relminmxid differences in pg_class (Christoph Berg) @@ -10511,7 +12299,7 @@ =head1 BUGS AND LIMITATIONS =head1 AUTHOR -Greg Sabino Mullane <greg@endpoint.com> +Greg Sabino Mullane <greg@turnstep.com> =head1 NAGIOS EXAMPLES @@ -10545,7 +12333,7 @@ =head1 NAGIOS EXAMPLES =head1 LICENSE AND COPYRIGHT -Copyright (c) 2007-2013 Greg Sabino Mullane <greg@endpoint.com>. +Copyright 2007 - 2025 Greg Sabino Mullane <greg@turnstep.com>. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: diff --git a/check_postgres.pl.asc b/check_postgres.pl.asc index 8569f01d..de691862 100644 --- a/check_postgres.pl.asc +++ b/check_postgres.pl.asc @@ -1,6 +1,6 @@ -----BEGIN PGP SIGNATURE----- -iEYEABEDAAYFAlJB46IACgkQvJuQZxSWSshP4gCgjJSVQ2NhZR6V6JsOgwyR6f5D -lQcAoJPJfp7P485pAIfLWDPuGo1Lk48G -=yH+B +iF0EABECAB0WIQQlKd9quPeUB+lERbS8m5BnFJZKyAUCZCsXMgAKCRC8m5BnFJZK +yPU8AJ4xhTwuzxoO+0Kwl55rbBa2GYWmXACfWgFZLE2wUEFSOSVgzgtg3bjdefc= +=43vZ -----END PGP SIGNATURE----- diff --git a/check_postgres.pl.html b/check_postgres.pl.html index 9ba2625b..6abc9842 100644 --- a/check_postgres.pl.html +++ b/check_postgres.pl.html @@ -2,2727 +2,2717 @@ <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> <html xmlns="http://www.w3.org/1999/xhtml"> <head> -<title>check_postgres.pl - a Postgres monitoring script for Nagios, MRTG, Cacti, and others +check_postgres.pl - - - - -
-

- -
- - -

-

-
-

NAME

-

check_postgres.pl - a Postgres monitoring script for Nagios, MRTG, Cacti, and others

-

This documents describes check_postgres.pl version 2.21.0

-

-

-
-

SYNOPSIS

-
-  ## Create all symlinks
-  check_postgres.pl --symlinks
-
-  ## Check connection to Postgres database 'pluto':
-  check_postgres.pl --action=connection --db=pluto
-
-  ## Same things, but using the symlink
-  check_postgres_connection --db=pluto
-
-  ## Warn if > 100 locks, critical if > 200, or > 20 exclusive
-  check_postgres_locks --warning=100 --critical="total=200:exclusive=20"
-
-  ## Show the current number of idle connections on port 6543:
-  check_postgres_txn_idle --port=6543 --output=simple
-
-  ## There are many other actions and options, please keep reading.
-
-  The latest news and documentation can always be found at:
-  http://bucardo.org/check_postgres/
-

-

-
-

DESCRIPTION

-

check_postgres.pl is a Perl script that runs many different tests against -one or more Postgres databases. It uses the psql program to gather the -information, and outputs the results in one of three formats: Nagios, MRTG, -or simple.

-

-

-

Output Modes

-

The output can be changed by use of the --output option. The default output -is nagios, although this can be changed at the top of the script if you wish. The -current option choices are nagios, mrtg, and simple. To avoid having to -enter the output argument each time, the type of output is automatically set -if no --output argument is given, and if the current directory has one of the -output options in its name. For example, creating a directory named mrtg and -populating it with symlinks via the --symlinks argument would ensure that -any actions run from that directory will always default to an output of "mrtg" -As a shortcut for --output=simple, you can enter --simple, which also overrides -the directory naming trick.

-

-

-

Nagios output

-

The default output format is for Nagios, which is a single line of information, along -with four specific exit codes:

-
    -
  1. (OK) - -
    2. -
  2. (WARNING) - -
    3. -
  3. (CRITICAL) - -
    4. -
  4. (UNKNOWN) - -
    5. -
-

The output line is one of the words above, a colon, and then a short description of what -was measured. Additional statistics information, as well as the total time the command -took, can be output as well: see the documentation on the arguments ---showperf, ---perflimit, and ---showtime.

-

-

-

MRTG output

-

The MRTG output is four lines, with the first line always giving a single number of importance. -When possible, this number represents an actual value such as a number of bytes, but it -may also be a 1 or a 0 for actions that only return "true" or "false", such as check_postgres_version. -The second line is an additional stat and is only used for some actions. The third line indicates -an "uptime" and is not used. The fourth line is a description and usually indicates the name of -the database the stat from the first line was pulled from, but may be different depending on the -action.

-

Some actions accept an optional --mrtg argument to further control the output.

-

See the documentation on each action for details on the exact MRTG output for each one.

-

-

-

Simple output

-

The simple output is simply a truncated version of the MRTG one, and simply returns the first number -and nothing else. This is very useful when you just want to check the state of something, regardless -of any threshold. You can transform the numeric output by appending KB, MB, GB, TB, or EB to the output -argument, for example:

-
-  --output=simple,MB
-

-

-

Cacti output

-

The Cacti output consists of one or more items on the same line, with a simple name, a colon, and -then a number. At the moment, the only action with explicit Cacti output is 'dbstats', and using -the --output option is not needed in this case, as Cacti is the only output for this action. For many -other actions, using --simple is enough to make Cacti happy.

-

-

-
-

DATABASE CONNECTION OPTIONS

-

All actions accept a common set of database options.

+

NAME

+ +

check_postgres.pl - a Postgres monitoring script for Nagios, MRTG, Cacti, and others

+ +

This documents describes check_postgres.pl version 2.26.0

+ +

SYNOPSIS

+ +
## Create all symlinks
+check_postgres.pl --symlinks
+
+## Check connection to Postgres database 'pluto':
+check_postgres.pl --action=connection --db=pluto
+
+## Same things, but using the symlink
+check_postgres_connection --db=pluto
+
+## Warn if > 100 locks, critical if > 200, or > 20 exclusive
+check_postgres_locks --warning=100 --critical="total=200:exclusive=20"
+
+## Show the current number of idle connections on port 6543:
+check_postgres_txn_idle --port=6543 --output=simple
+
+## There are many other actions and options, please keep reading.
+
+The latest news and documentation can always be found at:
+https://bucardo.org/check_postgres/
+ +

DESCRIPTION

+ +

check_postgres.pl is a Perl script that runs many different tests against one or more Postgres databases. It uses the psql program to gather the information, and outputs the results in one of three formats: Nagios, MRTG, or simple.

+ +

Output Modes

+ +

The output can be changed by use of the --output option. The default output is nagios, although this can be changed at the top of the script if you wish. The current option choices are nagios, mrtg, and simple. To avoid having to enter the output argument each time, the type of output is automatically set if no --output argument is given, and if the current directory has one of the output options in its name. For example, creating a directory named mrtg and populating it with symlinks via the --symlinks argument would ensure that any actions run from that directory will always default to an output of "mrtg" As a shortcut for --output=simple, you can enter --simple, which also overrides the directory naming trick.

+ +

Nagios output

+ +

The default output format is for Nagios, which is a single line of information, along with four specific exit codes:

+
-
-H NAME or --host=NAME
+
0 (OK)
+
+ +
+
1 (WARNING)
-

Connect to the host indicated by NAME. Can be a comma-separated list of names. Multiple host arguments -are allowed. If no host is given, defaults to the PGHOST environment variable or no host at all -(which indicates using a local Unix socket). You may also use "--dbhost".

+
-
-p PORT or --port=PORT
+
2 (CRITICAL)
+
+
+
3 (UNKNOWN)
-

Connects using the specified PORT number. Can be a comma-separated list of port numbers, and multiple -port arguments are allowed. If no port number is given, defaults to the PGPORT environment variable. If -that is not set, it defaults to 5432. You may also use "--dbport"

+
-
-db NAME or --dbname=NAME
+
+ +

The output line is one of the words above, a colon, and then a short description of what was measured. Additional statistics information, as well as the total time the command took, can be output as well: see the documentation on the arguments --showperf, --perflimit, and --showtime.

+ +

MRTG output

+ +

The MRTG output is four lines, with the first line always giving a single number of importance. When possible, this number represents an actual value such as a number of bytes, but it may also be a 1 or a 0 for actions that only return "true" or "false", such as check_postgres_version. The second line is an additional stat and is only used for some actions. The third line indicates an "uptime" and is not used. The fourth line is a description and usually indicates the name of the database the stat from the first line was pulled from, but may be different depending on the action.

+ +

Some actions accept an optional --mrtg argument to further control the output.

+ +

See the documentation on each action for details on the exact MRTG output for each one.

+ +

Simple output

+ +

The simple output is simply a truncated version of the MRTG one, and simply returns the first number and nothing else. This is very useful when you just want to check the state of something, regardless of any threshold. You can transform the numeric output by appending KB, MB, GB, TB, or EB to the output argument, for example:

+ +
--output=simple,MB
+ +

Cacti output

+ +

The Cacti output consists of one or more items on the same line, with a simple name, a colon, and then a number. At the moment, the only action with explicit Cacti output is 'dbstats', and using the --output option is not needed in this case, as Cacti is the only output for this action. For many other actions, using --simple is enough to make Cacti happy.

+ +

DATABASE CONNECTION OPTIONS

+ +

All actions accept a common set of database options.

+ +
+
-H NAME or --host=NAME
-

Specifies which database to connect to. Can be a comma-separated list of names, and multiple dbname -arguments are allowed. If no dbname option is provided, defaults to the PGDATABASE environment variable. -If that is not set, it defaults to 'postgres' if psql is version 8 or greater, and 'template1' otherwise.

+ +

Connect to the host indicated by NAME. Can be a comma-separated list of names. Multiple host arguments are allowed. If no host is given, defaults to the PGHOST environment variable or no host at all (which indicates using a local Unix socket). You may also use "--dbhost".

+
-
-u USERNAME or --dbuser=USERNAME
+
-p PORT or --port=PORT
+
+

Connects using the specified PORT number. Can be a comma-separated list of port numbers, and multiple port arguments are allowed. If no port number is given, defaults to the PGPORT environment variable. If that is not set, it defaults to 5432. You may also use "--dbport"

+ +
+
-db NAME or --dbname=NAME
-

The name of the database user to connect as. Can be a comma-separated list of usernames, and multiple -dbuser arguments are allowed. If this is not provided, it defaults to the PGUSER environment variable, otherwise -it defaults to 'postgres'.

+ +

Specifies which database to connect to. Can be a comma-separated list of names, and multiple dbname arguments are allowed. If no dbname option is provided, defaults to the PGDATABASE environment variable. If that is not set, it defaults to 'postgres' if psql is version 8 or greater, and 'template1' otherwise.

+
-
--dbpass=PASSWORD
+
-u USERNAME or --dbuser=USERNAME
+
+ +

The name of the database user to connect as. Can be a comma-separated list of usernames, and multiple dbuser arguments are allowed. If this is not provided, it defaults to the PGUSER environment variable, otherwise it defaults to 'postgres'.

+
+
--dbpass=PASSWORD
-

Provides the password to connect to the database with. Use of this option is highly discouraged. -Instead, one should use a .pgpass or pg_service.conf file.

+ +

Provides the password to connect to the database with. Use of this option is highly discouraged. Instead, one should use a .pgpass or pg_service.conf file.

+
-
--dbservice=NAME
+
--dbservice=NAME
+
+

The name of a service inside of the pg_service.conf file. Before version 9.0 of Postgres, this is a global file, usually found in /etc/pg_service.conf. If you are using version 9.0 or higher of Postgres, you can use the file ".pg_service.conf" in the home directory of the user running the script, e.g. nagios.

+ +

This file contains a simple list of connection options. You can also pass additional information when using this option such as --dbservice="maindatabase sslmode=require"

+ +

The documentation for this file can be found at https://www.postgresql.org/docs/current/static/libpq-pgservice.html

+ +
+
--role=ROLE
-

The name of a service inside of the pg_service.conf file. Before version 9.0 of Postgres, this is -a global file, usually found in /etc/pg_service.conf. If you are using version 9.0 or higher of -Postgres, you can use the file ".pg_service.conf" in the home directory of the user running -the script, e.g. nagios.

-

This file contains a simple list of connection options. You can also pass additional information -when using this option such as --dbservice="maindatabase sslmode=require"

-

The documentation for this file can be found at -http://www.postgresql.org/docs/current/static/libpq-pgservice.html

+ +

Provides the role to switch to after connecting to the database but before running the given check. This provides the ability to have superuser privileges assigned to a role without LOGIN access for the purposes of audit and other security considerations. Requires a local `psql` version 9.6 or higher.

+
-

The database connection options can be grouped: --host=a,b --host=c --port=1234 --port=3344 -would connect to a-1234, b-1234, and c-3344. Note that once set, an option -carries over until it is changed again.

+ +

The database connection options can be grouped: --host=a,b --host=c --port=1234 --port=3344 would connect to a-1234, b-1234, and c-3344. Note that once set, an option carries over until it is changed again.

+

Examples:

-
-  --host=a,b --port=5433 --db=c
-  Connects twice to port 5433, using database c, to hosts a and b: a-5433-c b-5433-c
-
-  --host=a,b --port=5433 --db=c,d
-  Connects four times: a-5433-c a-5433-d b-5433-c b-5433-d
-
-  --host=a,b --host=foo --port=1234 --port=5433 --db=e,f
-  Connects six times: a-1234-e a-1234-f b-1234-e b-1234-f foo-5433-e foo-5433-f
-
-  --host=a,b --host=x --port=5432,5433 --dbuser=alice --dbuser=bob -db=baz
-  Connects three times: a-5432-alice-baz b-5433-alice-baz x-5433-bob-baz
-
-  --dbservice="foo" --port=5433
-  Connects using the named service 'foo' in the pg_service.conf file, but overrides the port
-

-

-
-

OTHER OPTIONS

+ +
--host=a,b --port=5433 --db=c
+Connects twice to port 5433, using database c, to hosts a and b: a-5433-c b-5433-c
+
+--host=a,b --port=5433 --db=c,d
+Connects four times: a-5433-c a-5433-d b-5433-c b-5433-d
+
+--host=a,b --host=foo --port=1234 --port=5433 --db=e,f
+Connects six times: a-1234-e a-1234-f b-1234-e b-1234-f foo-5433-e foo-5433-f
+
+--host=a,b --host=x --port=5432,5433 --dbuser=alice --dbuser=bob -db=baz
+Connects three times: a-5432-alice-baz b-5433-alice-baz x-5433-bob-baz
+
+--dbservice="foo" --port=5433
+Connects using the named service 'foo' in the pg_service.conf file, but overrides the port
+ +

OTHER OPTIONS

+

Other options include:

+
-
--action=NAME
+
--action=NAME
-

States what action we are running. Required unless using a symlinked file, -in which case the name of the file is used to figure out the action.

-
-
--warning=VAL or -w VAL
-
-

Sets the threshold at which a warning alert is fired. The valid options for this -option depends on the action used.

-
-
--critical=VAL or -c VAL
+

States what action we are running. Required unless using a symlinked file, in which case the name of the file is used to figure out the action.

-
-

Sets the threshold at which a critical alert is fired. The valid options for this -option depends on the action used.

-
-t VAL or --timeout=VAL
+
--warning=VAL or -w VAL
+
+ +

Sets the threshold at which a warning alert is fired. The valid options for this option depends on the action used.

+
+
--critical=VAL or -c VAL
-

Sets the timeout in seconds after which the script will abort whatever it is doing -and return an UNKNOWN status. The timeout is per Postgres cluster, not for the entire -script. The default value is 10; the units are always in seconds.

+ +

Sets the threshold at which a critical alert is fired. The valid options for this option depends on the action used.

+
-
--assume-standby-mode
+
-t VAL or --timeout=VAL
+
+

Sets the timeout in seconds after which the script will abort whatever it is doing and return an UNKNOWN status. The timeout is per Postgres cluster, not for the entire script. The default value is 10; the units are always in seconds.

+ +
+
--assume-standby-mode
-

If specified, first the check if server in standby mode will be performed -(--datadir is required), if so, all checks that require SQL queries will be -ignored and "Server in standby mode" with OK status will be returned instead.

+ +

If specified, first the check if server in standby mode will be performed (--datadir is required), if so, all checks that require SQL queries will be ignored and "Server in standby mode" with OK status will be returned instead.

+

Example:

-
-    postgres@db$./check_postgres.pl --action=version --warning=8.1 --datadir /var/lib/postgresql/8.3/main/ --assume-standby-mode
-    POSTGRES_VERSION OK:  Server in standby mode | time=0.00
-
-
--assume-prod
+
postgres@db$./check_postgres.pl --action=version --warning=8.1 --datadir /var/lib/postgresql/8.3/main/ --assume-standby-mode
+POSTGRES_VERSION OK:  Server in standby mode | time=0.00
+ + +
--assume-prod
-

If specified, check if server in production mode is performed (--datadir is required). -The option is only relevant for (symlink: check_postgres_checkpoint).

+ +

If specified, check if server in production mode is performed (--datadir is required). The option is only relevant for (symlink: check_postgres_checkpoint).

+

Example:

-
-    postgres@db$./check_postgres.pl --action=checkpoint --datadir /var/lib/postgresql/8.3/main/ --assume-prod
-    POSTGRES_CHECKPOINT OK: Last checkpoint was 72 seconds ago | age=72;;300 mode=MASTER
+ +
postgres@db$./check_postgres.pl --action=checkpoint --datadir /var/lib/postgresql/8.3/main/ --assume-prod
+POSTGRES_CHECKPOINT OK: Last checkpoint was 72 seconds ago | age=72;;300 mode=MASTER
+
-
-h or --help
+
--assume-async
+
+ +

If specified, indicates that any replication between servers is asynchronous. The option is only relevant for (symlink: check_postgres_same_schema).

+ +

Example: postgres@db$./check_postgres.pl --action=same_schema --assume-async --dbhost=star,line

+
+
-h or --help
+

Displays a help screen with a summary of all actions and options.

-
-
--man
+ +
--man
+

Displays the entire manual.

-
-
-V or --version
+ +
-V or --version
+

Shows the current version.

-
-
-v or --verbose
-
-

Set the verbosity level. Can call more than once to boost the level. Setting it to three -or higher (in other words, issuing -v -v -v) turns on debugging information for this -program which is sent to stderr.

-
--showperf=VAL
- +
-v or --verbose
-

Determines if we output additional performance data in standard Nagios format -(at end of string, after a pipe symbol, using name=value). -VAL should be 0 or 1. The default is 1. Only takes effect if using Nagios output mode.

-
-
--perflimit=i
-
-

Sets a limit as to how many items of interest are reported back when using the -showperf option. This only has an effect for actions that return a large -number of items, such as table_size. The default is 0, or no limit. Be -careful when using this with the --include or --exclude options, as -those restrictions are done after the query has been run, and thus your -limit may not include the items you want. Only takes effect if using Nagios output mode.

-
-
--showtime=VAL
+

Set the verbosity level. Can call more than once to boost the level. Setting it to three or higher (in other words, issuing -v -v -v) turns on debugging information for this program which is sent to stderr.

-
-

Determines if the time taken to run each query is shown in the output. VAL -should be 0 or 1. The default is 1. No effect unless showperf is on. -Only takes effect if using Nagios output mode.

-
--test
- +
--showperf=VAL
-

Enables test mode. See the TEST MODE section below.

+ +

Determines if we output additional performance data in standard Nagios format (at end of string, after a pipe symbol, using name=value). VAL should be 0 or 1. The default is 1. Only takes effect if using Nagios output mode.

+
-
--PGBINDIR=PATH
+
--perflimit=i
+
+ +

Sets a limit as to how many items of interest are reported back when using the showperf option. This only has an effect for actions that return a large number of items, such as table_size. The default is 0, or no limit. Be careful when using this with the --include or --exclude options, as those restrictions are done after the query has been run, and thus your limit may not include the items you want. Only takes effect if using Nagios output mode.

+
+
--showtime=VAL
-

Tells the script where to find the psql binaries. Useful if you have more than -one version of the PostgreSQL executables on your system, or if there are not -in your path. Note that this option is in all uppercase. By default, this option -is not allowed. To enable it, you must change the $NO_PSQL_OPTION near the -top of the script to 0. Avoid using this option if you can, and instead use -environment variable c<PGBINDIR> or hard-coded $PGBINDIR variable, also near -the top of the script, to set the path to the PostgreSQL to use.

+ +

Determines if the time taken to run each query is shown in the output. VAL should be 0 or 1. The default is 1. No effect unless showperf is on. Only takes effect if using Nagios output mode.

+
-
--PSQL=PATH
+
--test
+
+

Enables test mode. See the "TEST MODE" section below.

+ +
+
--PGBINDIR=PATH
-

(deprecated, this option may be removed in a future release!) -Tells the script where to find the psql program. Useful if you have more than -one version of the psql executable on your system, or if there is no psql program -in your path. Note that this option is in all uppercase. By default, this option -is not allowed. To enable it, you must change the $NO_PSQL_OPTION near the -top of the script to 0. Avoid using this option if you can, and instead hard-code -your psql location into the $PSQL variable, also near the top of the script.

+ +

Tells the script where to find the psql binaries. Useful if you have more than one version of the PostgreSQL executables on your system, or if there are not in your path. Note that this option is in all uppercase. By default, this option is not allowed. To enable it, you must change the $NO_PSQL_OPTION near the top of the script to 0. Avoid using this option if you can, and instead use environment variable c<PGBINDIR> or hard-coded $PGBINDIR variable, also near the top of the script, to set the path to the PostgreSQL to use.

+
-
--symlinks
+
--PSQL=PATH
+
+ +

(deprecated, this option may be removed in a future release!) Tells the script where to find the psql program. Useful if you have more than one version of the psql executable on your system, or if there is no psql program in your path. Note that this option is in all uppercase. By default, this option is not allowed. To enable it, you must change the $NO_PSQL_OPTION near the top of the script to 0. Avoid using this option if you can, and instead hard-code your psql location into the $PSQL variable, also near the top of the script.

+
+
+

Creates symlinks to the main program for each action.

-
-
--output=VAL
-
-

Determines the format of the output, for use in various programs. The -default is 'nagios'. Available options are 'nagios', 'mrtg', 'simple' -and 'cacti'.

-
--mrtg=VAL
+
--output=VAL
+
+ +

Determines the format of the output, for use in various programs. The default is 'nagios'. Available options are 'nagios', 'mrtg', 'simple' and 'cacti'.

+
+
--mrtg=VAL
+

Used only for the MRTG or simple output, for a few specific actions.

-
-
--debugoutput=VAL
-
-

Outputs the exact string returned by psql, for use in debugging. The value is one or more letters, -which determine if the output is displayed or not, where 'a' = all, 'c' = critical, 'w' = warning, -'o' = ok, and 'u' = unknown. Letters can be combined.

-
--get_method=VAL
- +
--debugoutput=VAL
-

Allows specification of the method used to fetch information for the new_version_cp, -new_version_pg, new_version_bc, new_version_box, and new_version_tnm checks. -The following programs are tried, in order, to grab the information from the web: -GET, wget, fetch, curl, lynx, links. To force the use of just one (and thus remove the -overhead of trying all the others until one of those works), enter one of the names as -the argument to get_method. For example, a BSD box might enter the following line in -their .check_postgresrc file:

-
-  get_method=fetch
+ +

Outputs the exact string returned by psql, for use in debugging. The value is one or more letters, which determine if the output is displayed or not, where 'a' = all, 'c' = critical, 'w' = warning, 'o' = ok, and 'u' = unknown. Letters can be combined.

+
-
--language=VAL
+
--get_method=VAL
+
+

Allows specification of the method used to fetch information for the new_version_cp, new_version_pg, new_version_bc, new_version_box, and new_version_tnm checks. The following programs are tried, in order, to grab the information from the web: GET, wget, fetch, curl, lynx, links. To force the use of just one (and thus remove the overhead of trying all the others until one of those works), enter one of the names as the argument to get_method. For example, a BSD box might enter the following line in their .check_postgresrc file:

+ +
get_method=fetch
+ +
+
--language=VAL
-

Set the language to use for all output messages. Normally, this is detected by examining -the environment variables LC_ALL, LC_MESSAGES, and LANG, but setting this option -will override any such detection.

+ +

Set the language to use for all output messages. Normally, this is detected by examining the environment variables LC_ALL, LC_MESSAGES, and LANG, but setting this option will override any such detection.

+
-

-

-
-

ACTIONS

-

The script runs one or more actions. This can either be done with the --action -flag, or by using a symlink to the main file that contains the name of the action -inside of it. For example, to run the action "timesync", you may either issue:

-
-  check_postgres.pl --action=timesync
+ +

ACTIONS

+ +

The action to be run is selected using the --action flag, or by using a symlink to the main file that contains the name of the action inside of it. For example, to run the action "timesync", you may either issue:

+ +
check_postgres.pl --action=timesync
+

or use a program named:

-
-  check_postgres_timesync
-

All the symlinks are created for you in the current directory -if use the option --symlinks

-
-  perl check_postgres.pl --symlinks
-

If the file name already exists, it will not be overwritten. If the file exists -and is a symlink, you can force it to overwrite by using "--action=build_symlinks_force"

-

Most actions take a --warning and a --critical option, indicating at what -point we change from OK to WARNING, and what point we go to CRITICAL. Note that -because criticals are always checked first, setting the warning equal to the -critical is an effective way to turn warnings off and always give a critical.

+ +
check_postgres_timesync
+ +

All the symlinks are created for you in the current directory if use the option --symlinks:

+ +
perl check_postgres.pl --symlinks
+ +

If the file name already exists, it will not be overwritten. If the file exists and is a symlink, you can force it to overwrite by using "--action=build_symlinks_force".

+ +

Most actions take a --warning and a --critical option, indicating at what point we change from OK to WARNING, and what point we go to CRITICAL. Note that because criticals are always checked first, setting the warning equal to the critical is an effective way to turn warnings off and always give a critical.

+

The current supported actions are:

-

-

-

archive_ready

-

(symlink: check_postgres_archive_ready) Checks how many WAL files with extension .ready -exist in the pg_xlog/archive_status directory, which is found -off of your data_directory. This action must be run as a superuser, in order to access the -contents of the pg_xlog/archive_status directory. The minimum version to use this action is -Postgres 8.1. The --warning and --critical options are simply the number of -.ready files in the pg_xlog/archive_status directory. -Usually, these values should be low, turning on the archive mechanism, we usually want it to -archive WAL files as fast as possible.

-

If the archive command fail, number of WAL in your pg_xlog directory will grow until -exhausting all the disk space and force PostgreSQL to stop immediately.

-

Example 1: Check that the number of ready WAL files is 10 or less on host "pluto"

-
-  check_postgres_archive_ready --host=pluto --critical=10
+ +

archive_ready

+ +

(symlink: check_postgres_archive_ready) Checks how many WAL files with extension .ready exist in the pg_xlog/archive_status directory (PostgreSQL 10 and later: pg_wal/archive_status), which is found off of your data_directory. If the --lsfunc option is not used then this action must be run as a superuser, in order to access the contents of the pg_xlog/archive_status directory. The minimum version to use this action is Postgres 8.1. The --warning and --critical options are simply the number of .ready files in the pg_xlog/archive_status directory. Usually, these values should be low, turning on the archive mechanism, we usually want it to archive WAL files as fast as possible.

+ +

If the archive command fail, number of WAL in your pg_xlog directory will grow until exhausting all the disk space and force PostgreSQL to stop immediately.

+ +

To avoid running as a database superuser, a wrapper function around pg_ls_dir() should be defined as a superuser with SECURITY DEFINER, and the --lsfunc option used. This example function, if defined by a superuser, will allow the script to connect as a normal user nagios with --lsfunc=ls_archive_status_dir

+ +
BEGIN;
+CREATE FUNCTION ls_archive_status_dir()
+    RETURNS SETOF TEXT
+    AS $$ SELECT pg_ls_dir('pg_xlog/archive_status') $$
+    LANGUAGE SQL
+    SECURITY DEFINER;
+REVOKE ALL ON FUNCTION ls_archive_status_dir() FROM PUBLIC;
+GRANT EXECUTE ON FUNCTION ls_archive_status_dir() to nagios;
+COMMIT;
+ +

Example 1: Check that the number of ready WAL files is 10 or less on host "pluto", using a wrapper function ls_archive_status_dir to avoid the need for superuser permissions

+ +
check_postgres_archive_ready --host=pluto --critical=10 --lsfunc=ls_archive_status_dir
+

For MRTG output, reports the number of ready WAL files on line 1.

-

-

-

autovac_freeze

-

(symlink: check_postgres_autovac_freeze) Checks how close each database is to the Postgres autovacuum_freeze_max_age setting. This -action will only work for databases version 8.2 or higher. The --warning and ---critical options should be expressed as percentages. The 'age' of the transactions -in each database is compared to the autovacuum_freeze_max_age setting (200 million by default) -to generate a rounded percentage. The default values are 90% for the warning and 95% for -the critical. Databases can be filtered by use of the --include and --exclude options. -See the BASIC FILTERING section for more details.

+ +

autovac_freeze

+ +

(symlink: check_postgres_autovac_freeze) Checks how close each database is to the Postgres autovacuum_freeze_max_age setting. This action will only work for databases version 8.2 or higher. The --warning and --critical options should be expressed as percentages. The 'age' of the transactions in each database is compared to the autovacuum_freeze_max_age setting (200 million by default) to generate a rounded percentage. The default values are 90% for the warning and 95% for the critical. Databases can be filtered by use of the --include and --exclude options. See the "BASIC FILTERING" section for more details.

+

Example 1: Give a warning when any databases on port 5432 are above 97%

-
-  check_postgres_autovac_freeze --port=5432 --warning="97%"
-

For MRTG output, the highest overall percentage is reported on the first line, and the highest age is -reported on the second line. All databases which have the percentage from the first line are reported -on the fourth line, separated by a pipe symbol.

-

-

-

backends

-

(symlink: check_postgres_backends) Checks the current number of connections for one or more databases, and optionally -compares it to the maximum allowed, which is determined by the -Postgres configuration variable max_connections. The --warning and ---critical options can take one of three forms. First, a simple number can be -given, which represents the number of connections at which the alert will be -given. This choice does not use the max_connections setting. Second, the -percentage of available connections can be given. Third, a negative number can -be given which represents the number of connections left until max_connections -is reached. The default values for --warning and --critical are '90%' and '95%'. -You can also filter the databases by use of the --include and --exclude options. -See the BASIC FILTERING section for more details.

-

To view only non-idle processes, you can use the --noidle argument. Note that the -user you are connecting as must be a superuser for this to work properly.

+ +
check_postgres_autovac_freeze --port=5432 --warning="97%"
+ +

For MRTG output, the highest overall percentage is reported on the first line, and the highest age is reported on the second line. All databases which have the percentage from the first line are reported on the fourth line, separated by a pipe symbol.

+ +

backends

+ +

(symlink: check_postgres_backends) Checks the current number of connections for one or more databases, and optionally compares it to the maximum allowed, which is determined by the Postgres configuration variable max_connections. The --warning and --critical options can take one of three forms. First, a simple number can be given, which represents the number of connections at which the alert will be given. This choice does not use the max_connections setting. Second, the percentage of available connections can be given. Third, a negative number can be given which represents the number of connections left until max_connections is reached. The default values for --warning and --critical are '90%' and '95%'. You can also filter the databases by use of the --include and --exclude options. See the "BASIC FILTERING" section for more details.

+ +

To view only non-idle processes, you can use the --noidle argument. Note that the user you are running as (either connecting directly or switching via --role) must be a superuser for this to work properly.

+

Example 1: Give a warning when the number of connections on host quirm reaches 120, and a critical if it reaches 150.

-
-  check_postgres_backends --host=quirm --warning=120 --critical=150
+ +
check_postgres_backends --host=quirm --warning=120 --critical=150
+

Example 2: Give a critical when we reach 75% of our max_connections setting on hosts lancre or lancre2.

-
-  check_postgres_backends --warning='75%' --critical='75%' --host=lancre,lancre2
-

Example 3: Give a warning when there are only 10 more connection slots left on host plasmid, and a critical -when we have only 5 left.

-
-  check_postgres_backends --warning=-10 --critical=-5 --host=plasmid
+ +
check_postgres_backends --warning='75%' --critical='75%' --host=lancre,lancre2
+ +

Example 3: Give a warning when there are only 10 more connection slots left on host plasmid, and a critical when we have only 5 left.

+ +
check_postgres_backends --warning=-10 --critical=-5 --host=plasmid
+

Example 4: Check all databases except those with "test" in their name, but allow ones that are named "pg_greatest". Connect as port 5432 on the first two hosts, and as port 5433 on the third one. We want to always throw a critical when we reach 30 or more connections.

-
- check_postgres_backends --dbhost=hong,kong --dbhost=fooey --dbport=5432 --dbport=5433 --warning=30 --critical=30 --exclude="~test" --include="pg_greatest,~prod"
-

For MRTG output, the number of connections is reported on the first line, and the fourth line gives the name of the database, -plus the current maximum_connections. If more than one database has been queried, the one with the highest number of -connections is output.

-

-

-

bloat

-

(symlink: check_postgres_bloat) Checks the amount of bloat in tables and indexes. (Bloat is generally the amount -of dead unused space taken up in a table or index. This space is usually reclaimed -by use of the VACUUM command.) This action requires that stats collection be -enabled on the target databases, and requires that ANALYZE is run frequently. -The --include and --exclude options can be used to filter out which tables -to look at. See the BASIC FILTERING section for more details.

-

The --warning and --critical options can be specified as sizes, percents, or both. -Valid size units are bytes, kilobytes, megabytes, gigabytes, terabytes, exabytes, -petabytes, and zettabytes. You can abbreviate all of those with the first letter. Items -without units are assumed to be 'bytes'. The default values are '1 GB' and '5 GB'. The value -represents the number of "wasted bytes", or the difference between what is actually -used by the table and index, and what we compute that it should be.

-

Note that this action has two hard-coded values to avoid false alarms on -smaller relations. Tables must have at least 10 pages, and indexes at least 15, -before they can be considered by this test. If you really want to adjust these -values, you can look for the variables $MINPAGES and $MINIPAGES at the top of the -check_bloat subroutine. These values are ignored if either --exclude or ---include is used.

-

Only the top 10 most bloated relations are shown. You can change this number by -using the --perflimit option to set your own limit.

-

The schema named 'information_schema' is excluded from this test, as the only tables -it contains are small and do not change.

-

Please note that the values computed by this action are not precise, and -should be used as a guideline only. Great effort was made to estimate the -correct size of a table, but in the end it is only an estimate. The correct -index size is even more of a guess than the correct table size, but both -should give a rough idea of how bloated things are.

+ +
check_postgres_backends --dbhost=hong,kong --dbhost=fooey --dbport=5432 --dbport=5433 --warning=30 --critical=30 --exclude="~test" --include="pg_greatest,~prod"
+ +

For MRTG output, the number of connections is reported on the first line, and the fourth line gives the name of the database, plus the current maximum_connections. If more than one database has been queried, the one with the highest number of connections is output.

+ +

bloat

+ +

(symlink: check_postgres_bloat) Checks the amount of bloat in tables and indexes. (Bloat is generally the amount of dead unused space taken up in a table or index. This space is usually reclaimed by use of the VACUUM command.) This action requires that stats collection be enabled on the target databases, and requires that ANALYZE is run frequently. The --include and --exclude options can be used to filter out which tables to look at. See the "BASIC FILTERING" section for more details.

+ +

The --warning and --critical options can be specified as sizes, percents, or both. Valid size units are bytes, kilobytes, megabytes, gigabytes, terabytes, exabytes, petabytes, and zettabytes. You can abbreviate all of those with the first letter. Items without units are assumed to be 'bytes'. The default values are '1 GB' and '5 GB'. The value represents the number of "wasted bytes", or the difference between what is actually used by the table and index, and what we compute that it should be.

+ +

Note that this action has two hard-coded values to avoid false alarms on smaller relations. Tables must have at least 10 pages, and indexes at least 15, before they can be considered by this test. If you really want to adjust these values, you can look for the variables $MINPAGES and $MINIPAGES at the top of the check_bloat subroutine. These values are ignored if either --exclude or --include is used.

+ +

Only the top 10 most bloated relations are shown. You can change this number by using the --perflimit option to set your own limit.

+ +

The schema named 'information_schema' is excluded from this test, as the only tables it contains are small and do not change.

+ +

Please note that the values computed by this action are not precise, and should be used as a guideline only. Great effort was made to estimate the correct size of a table, but in the end it is only an estimate. The correct index size is even more of a guess than the correct table size, but both should give a rough idea of how bloated things are.

+

Example 1: Warn if any table on port 5432 is over 100 MB bloated, and critical if over 200 MB

-
-  check_postgres_bloat --port=5432 --warning='100 M' --critical='200 M'
-

Example 2: Give a critical if table 'orders' on host 'sami' has more than 10 megs of bloat

-
-  check_postgres_bloat --host=sami --include=orders --critical='10 MB'
-

Example 3: Give a critical if table 'q4' on database 'sales' is over 50% bloated

-
-  check_postgres_bloat --db=sales --include=q4 --critical='50%'
-

Example 4: Give a critical any table is over 20% bloated and has over 150 -MB of bloat:

-
-  check_postgres_bloat --port=5432 --critical='20% and 150 M'
-

Example 5: Give a critical any table is over 40% bloated or has over 500 MB -of bloat:

-
-  check_postgres_bloat --port=5432 --warning='500 M or 40%'
-

For MRTG output, the first line gives the highest number of wasted bytes for the tables, and the -second line gives the highest number of wasted bytes for the indexes. The fourth line gives the database -name, table name, and index name information. If you want to output the bloat ratio instead (how many -times larger the relation is compared to how large it should be), just pass in --mrtg=ratio.

-

-

-

checkpoint

-

(symlink: check_postgres_checkpoint) Determines how long since the last checkpoint has -been run. This must run on the same server as the database that is being checked (e.g. the -h -flag will not work). This check is meant to run on a "warm standby" server that is actively -processing shipped WAL files, and is meant to check that your warm standby is truly 'warm'. -The data directory must be set, either by the environment variable PGDATA, or passing -the --datadir argument. It returns the number of seconds since the last checkpoint -was run, as determined by parsing the call to pg_controldata. Because of this, the -pg_controldata executable must be available in the current path. Alternatively, -you can specify PGBINDIR as the directory that it lives in. -It is also possible to use the special options --assume-prod or ---assume-standby-mode, if the mode found is not the one expected, a CRITICAL is emitted.

+ +
check_postgres_bloat --port=5432 --warning='100 M' --critical='200 M'
+ +

Example 2: Give a critical if table 'orders' on host 'sami' has more than 10 megs of bloat

+ +
check_postgres_bloat --host=sami --include=orders --critical='10 MB'
+ +

Example 3: Give a critical if table 'q4' on database 'sales' is over 50% bloated

+ +
check_postgres_bloat --db=sales --include=q4 --critical='50%'
+ +

Example 4: Give a critical any table is over 20% bloated and has over 150 MB of bloat:

+ +
check_postgres_bloat --port=5432 --critical='20% and 150 M'
+ +

Example 5: Give a critical any table is over 40% bloated or has over 500 MB of bloat:

+ +
check_postgres_bloat --port=5432 --warning='500 M or 40%'
+ +

For MRTG output, the first line gives the highest number of wasted bytes for the tables, and the second line gives the highest number of wasted bytes for the indexes. The fourth line gives the database name, table name, and index name information. If you want to output the bloat ratio instead (how many times larger the relation is compared to how large it should be), just pass in --mrtg=ratio.

+ +

checkpoint

+ +

(symlink: check_postgres_checkpoint) Determines how long since the last checkpoint has been run. This must run on the same server as the database that is being checked (e.g. the -h flag will not work). This check is meant to run on a "warm standby" server that is actively processing shipped WAL files, and is meant to check that your warm standby is truly 'warm'. The data directory must be set, either by the environment variable PGDATA, or passing the --datadir argument. It returns the number of seconds since the last checkpoint was run, as determined by parsing the call to pg_controldata. Because of this, the pg_controldata executable must be available in the current path. Alternatively, you can specify PGBINDIR as the directory that it lives in. It is also possible to use the special options --assume-prod or --assume-standby-mode, if the mode found is not the one expected, a CRITICAL is emitted.

+

At least one warning or critical argument must be set.

+

This action requires the Date::Parse module.

+

For MRTG or simple output, returns the number of seconds.

-

-

-

cluster_id

-

(symlink: check_postgres_cluster-id) Checks that the Database System Identifier -provided by pg_controldata is the same as last time you checked. This must run on the same -server as the database that is being checked (e.g. the -h flag will not work). -Either the --warning or the --critical option should be given, but not both. The value -of each one is the cluster identifier, an integer value. You can run with the special --critical=0 option -to find out an existing cluster identifier.

+ +

cluster_id

+ +

(symlink: check_postgres_cluster-id) Checks that the Database System Identifier provided by pg_controldata is the same as last time you checked. This must run on the same server as the database that is being checked (e.g. the -h flag will not work). Either the --warning or the --critical option should be given, but not both. The value of each one is the cluster identifier, an integer value. You can run with the special --critical=0 option to find out an existing cluster identifier.

+

Example 1: Find the initial identifier

-
-  check_postgres_cluster_id --critical=0 --datadir=/var//lib/postgresql/9.0/main
+ +
check_postgres_cluster_id --critical=0 --datadir=/var//lib/postgresql/9.0/main
+

Example 2: Make sure the cluster is the same and warn if not, using the result from above.

-
-  check_postgres_cluster_id  --critical=5633695740047915135
-

For MRTG output, returns a 1 or 0 indicating success of failure of the identifier to match. A -identifier must be provided as the --mrtg argument. The fourth line always gives the -current identifier.

-

-

-

commitratio

-

(symlink: check_postgres_commitratio) Checks the commit ratio of all databases and complains when they are too low. -There is no need to run this command more than once per database cluster. -Databases can be filtered with -the --include and --exclude options. See the BASIC FILTERING section -for more details. -They can also be filtered by the owner of the database with the ---includeuser and --excludeuser options. -See the USER NAME FILTERING section for more details.

-

The warning and critical options should be specified as percentages. There are not -defaults for this action: the warning and critical must be specified. The warning value -cannot be greater than the critical value. The output returns all databases sorted by -commitratio, smallest first.

+ +
check_postgres_cluster_id  --critical=5633695740047915135
+ +

For MRTG output, returns a 1 or 0 indicating success of failure of the identifier to match. A identifier must be provided as the --mrtg argument. The fourth line always gives the current identifier.

+ +

commitratio

+ +

(symlink: check_postgres_commitratio) Checks the commit ratio of all databases and complains when they are too low. There is no need to run this command more than once per database cluster. Databases can be filtered with the --include and --exclude options. See the "BASIC FILTERING" section for more details. They can also be filtered by the owner of the database with the --includeuser and --excludeuser options. See the "USER NAME FILTERING" section for more details.

+ +

The warning and critical options should be specified as percentages. There are not defaults for this action: the warning and critical must be specified. The warning value cannot be greater than the critical value. The output returns all databases sorted by commitratio, smallest first.

+

Example: Warn if any database on host flagg is less than 90% in commitratio, and critical if less then 80%.

-
-  check_postgres_database_commitratio --host=flagg --warning='90%' --critical='80%'
-

For MRTG output, returns the percentage of the database with the smallest commitratio on the first line, -and the name of the database on the fourth line.

-

-

-

connection

-

(symlink: check_postgres_connection) Simply connects, issues a 'SELECT version()', and leaves. -Takes no --warning or --critical options.

+ +
check_postgres_database_commitratio --host=flagg --warning='90%' --critical='80%'
+ +

For MRTG output, returns the percentage of the database with the smallest commitratio on the first line, and the name of the database on the fourth line.

+ +

connection

+ +

(symlink: check_postgres_connection) Simply connects, issues a 'SELECT version()', and leaves. Takes no --warning or --critical options.

+

For MRTG output, simply outputs a 1 (good connection) or a 0 (bad connection) on the first line.

-

-

-

custom_query

-

(symlink: check_postgres_custom_query) Runs a custom query of your choosing, and parses the results. -The query itself is passed in through the query argument, and should be kept as simple as possible. -If at all possible, wrap it in a view or a function to keep things easier to manage. The query should -return one or two columns. It is required that one of the columns be named "result" and is the item -that will be checked against your warning and critical values. The second column is for the performance -data and any name can be used: this will be the 'value' inside the performance data section.

-

At least one warning or critical argument must be specified. What these are set to depends on the type of -query you are running. There are four types of custom_queries that can be run, specified by the valtype -argument. If none is specified, this action defaults to 'integer'. The four types are:

-

integer: -Does a simple integer comparison. The first column should be a simple integer, and the warning and -critical values should be the same.

-

string: -The warning and critical are strings, and are triggered only if the value in the first column matches -it exactly. This is case-sensitive.

-

time: -The warning and the critical are times, and can have units of seconds, minutes, hours, or days. -Each may be written singular or abbreviated to just the first letter. If no units are given, -seconds are assumed. The first column should be an integer representing the number of seconds -to check.

-

size: -The warning and the critical are sizes, and can have units of bytes, kilobytes, megabytes, gigabytes, -terabytes, or exabytes. Each may be abbreviated to the first letter. If no units are given, -bytes are assumed. The first column should be an integer representing the number of bytes to check.

-

Normally, an alert is triggered if the values returned are greater than or equal to the critical or warning -value. However, an option of --reverse will trigger the alert if the returned value is -lower than or equal to the critical or warning value.

-

Example 1: Warn if any relation over 100 pages is named "rad", put the number of pages -inside the performance data section.

-
-  check_postgres_custom_query --valtype=string -w "rad" --query=
-    "SELECT relname AS result, relpages AS pages FROM pg_class WHERE relpages > 100"
+ +

custom_query

+ +

(symlink: check_postgres_custom_query) Runs a custom query of your choosing, and parses the results. The query itself is passed in through the query argument, and should be kept as simple as possible. If at all possible, wrap it in a view or a function to keep things easier to manage. The query should return one or two columns. It is required that one of the columns be named "result" and is the item that will be checked against your warning and critical values. The second column is for the performance data and any name can be used: this will be the 'value' inside the performance data section.

+ +

At least one warning or critical argument must be specified. What these are set to depends on the type of query you are running. There are four types of custom_queries that can be run, specified by the valtype argument. If none is specified, this action defaults to 'integer'. The four types are:

+ +

integer: Does a simple integer comparison. The first column should be a simple integer, and the warning and critical values should be the same.

+ +

string: The warning and critical are strings, and are triggered only if the value in the first column matches it exactly. This is case-sensitive.

+ +

time: The warning and the critical are times, and can have units of seconds, minutes, hours, or days. Each may be written singular or abbreviated to just the first letter. If no units are given, seconds are assumed. The first column should be an integer representing the number of seconds to check.

+ +

size: The warning and the critical are sizes, and can have units of bytes, kilobytes, megabytes, gigabytes, terabytes, or exabytes. Each may be abbreviated to the first letter. If no units are given, bytes are assumed. The first column should be an integer representing the number of bytes to check.

+ +

Normally, an alert is triggered if the values returned are greater than or equal to the critical or warning value. However, an option of --reverse will trigger the alert if the returned value is lower than or equal to the critical or warning value.

+ +

Example 1: Warn if any relation over 100 pages is named "rad", put the number of pages inside the performance data section.

+ +
check_postgres_custom_query --valtype=string -w "rad" --query=
+  "SELECT relname AS result, relpages AS pages FROM pg_class WHERE relpages > 100"
+

Example 2: Give a critical if the "foobar" function returns a number over 5MB:

-
-  check_postgres_custom_query --critical='5MB'--valtype=size --query="SELECT foobar() AS result"
+ +
check_postgres_custom_query --critical='5MB'--valtype=size --query="SELECT foobar() AS result"
+

Example 2: Warn if the function "snazzo" returns less than 42:

-
-  check_postgres_custom_query --critical=42 --query="SELECT snazzo() AS result" --reverse
-

If you come up with a useful custom_query, consider sending in a patch to this program -to make it into a standard action that other people can use.

+ +
check_postgres_custom_query --critical=42 --query="SELECT snazzo() AS result" --reverse
+ +

If you come up with a useful custom_query, consider sending in a patch to this program to make it into a standard action that other people can use.

+

This action does not support MRTG or simple output yet.

-

-

-

database_size

-

(symlink: check_postgres_database_size) Checks the size of all databases and complains when they are too big. -There is no need to run this command more than once per database cluster. -Databases can be filtered with -the --include and --exclude options. See the BASIC FILTERING section -for more details. -They can also be filtered by the owner of the database with the ---includeuser and --excludeuser options. -See the USER NAME FILTERING section for more details.

-

The warning and critical options can be specified as bytes, kilobytes, megabytes, -gigabytes, terabytes, or exabytes. Each may be abbreviated to the first letter as well. -If no unit is given, the units are assumed to be bytes. There are not defaults for this -action: the warning and critical must be specified. The warning value cannot be greater -than the critical value. The output returns all databases sorted by size largest first, -showing both raw bytes and a "pretty" version of the size.

+ +

database_size

+ +

(symlink: check_postgres_database_size) Checks the size of all databases and complains when they are too big. There is no need to run this command more than once per database cluster. Databases can be filtered with the --include and --exclude options. See the "BASIC FILTERING" section for more details. They can also be filtered by the owner of the database with the --includeuser and --excludeuser options. See the "USER NAME FILTERING" section for more details.

+ +

The warning and critical options can be specified as bytes, kilobytes, megabytes, gigabytes, terabytes, or exabytes. Each may be abbreviated to the first letter as well. If no unit is given, the units are assumed to be bytes. There are not defaults for this action: the warning and critical must be specified. The warning value cannot be greater than the critical value. The output returns all databases sorted by size largest first, showing both raw bytes and a "pretty" version of the size.

+

Example 1: Warn if any database on host flagg is over 1 TB in size, and critical if over 1.1 TB.

-
-  check_postgres_database_size --host=flagg --warning='1 TB' --critical='1.1 t'
+ +
check_postgres_database_size --host=flagg --warning='1 TB' --critical='1.1 t'
+

Example 2: Give a critical if the database template1 on port 5432 is over 10 MB.

-
-  check_postgres_database_size --port=5432 --include=template1 --warning='10MB' --critical='10MB'
-

Example 3: Give a warning if any database on host 'tardis' owned by the user 'tom' is over 5 GB

-
-  check_postgres_database_size --host=tardis --includeuser=tom --warning='5 GB' --critical='10 GB'
-

For MRTG output, returns the size in bytes of the largest database on the first line, -and the name of the database on the fourth line.

-

-

-

dbstats

-

(symlink: check_postgres_dbstats) Reports information from the pg_stat_database view, -and outputs it in a Cacti-friendly manner. No other output is supported, as the output -is informational and does not lend itself to alerts, such as used with Nagios. If no -options are given, all databases are returned, one per line. You can include a specific -database by use of the --include option, or you can use the --dbname option.

-

Eleven items are returned on each line, in the format name:value, separated by a single -space. The items are:

+ +
check_postgres_database_size --port=5432 --include=template1 --warning='10MB' --critical='10MB'
+ +

Example 3: Give a warning if any database on host 'tardis' owned by the user 'tom' is over 5 GB

+ +
check_postgres_database_size --host=tardis --includeuser=tom --warning='5 GB' --critical='10 GB'
+ +

For MRTG output, returns the size in bytes of the largest database on the first line, and the name of the database on the fourth line.

+ +

dbstats

+ +

(symlink: check_postgres_dbstats) Reports information from the pg_stat_database view, and outputs it in a Cacti-friendly manner. No other output is supported, as the output is informational and does not lend itself to alerts, such as used with Nagios. If no options are given, all databases are returned, one per line. You can include a specific database by use of the --include option, or you can use the --dbname option.

+ +

Eleven items are returned on each line, in the format name:value, separated by a single space. The items are:

+
-
backends
+
backends
+

The number of currently running backends for this database.

-
-
commits
+ +
commits
+

The total number of commits for this database since it was created or reset.

-
-
rollbacks
+ +
rollbacks
+

The total number of rollbacks for this database since it was created or reset.

-
-
read
+ +
read
+

The total number of disk blocks read.

-
-
hit
+ +
hit
+

The total number of buffer hits.

-
-
ret
+ +
ret
+

The total number of rows returned.

-
-
fetch
+ +
fetch
+

The total number of rows fetched.

-
-
ins
+ +
ins
+

The total number of rows inserted.

-
-
upd
+ +
upd
+

The total number of rows updated.

-
-
del
+ +
del
+

The total number of rows deleted.

-
-
dbname
+ +
dbname
+

The name of the database.

+
-

Note that ret, fetch, ins, upd, and del items will always be 0 if Postgres is version 8.2 or lower, as those stats were -not available in those versions.

+ +

Note that ret, fetch, ins, upd, and del items will always be 0 if Postgres is version 8.2 or lower, as those stats were not available in those versions.

+

If the dbname argument is given, seven additional items are returned:

+
-
idxscan
+
idxscan
+

Total number of user index scans.

-
-
idxtupread
+ +
idxtupread
+

Total number of user index entries returned.

-
-
idxtupfetch
+ +
idxtupfetch
+

Total number of rows fetched by simple user index scans.

-
-
idxblksread
+ +
idxblksread
+

Total number of disk blocks read for all user indexes.

-
-
idxblkshit
+ +
idxblkshit
+

Total number of buffer hits for all user indexes.

-
-
seqscan
+ +
seqscan
+

Total number of sequential scans against all user tables.

-
-
seqtupread
+ +
seqtupread
+

Total number of tuples returned from all user tables.

+
+

Example 1: Grab the stats for a database named "products" on host "willow":

-
-  check_postgres_dbstats --dbhost willow --dbname products
+ +
check_postgres_dbstats --dbhost willow --dbname products
+

The output returned will be like this (all on one line, not wrapped):

-
-    backends:82 commits:58374408 rollbacks:1651 read:268435543 hit:2920381758 idxscan:310931294 idxtupread:2777040927
-    idxtupfetch:1840241349 idxblksread:62860110 idxblkshit:1107812216 seqscan:5085305 seqtupread:5370500520
-    ret:0 fetch:0 ins:0 upd:0 del:0 dbname:willow
-

-

-

disabled_triggers

-

(symlink: check_postgres_disabled_triggers) Checks on the number of disabled triggers inside the database. -The --warning and --critical options are the number of such triggers found, and both -default to "1", as in normal usage having disabled triggers is a dangerous event. If the -database being checked is 8.3 or higher, the check is for the number of triggers that are -in a 'disabled' status (as opposed to being 'always' or 'replica'). The output will show -the name of the table and the name of the trigger for each disabled trigger.

+ +
backends:82 commits:58374408 rollbacks:1651 read:268435543 hit:2920381758 idxscan:310931294 idxtupread:2777040927
+idxtupfetch:1840241349 idxblksread:62860110 idxblkshit:1107812216 seqscan:5085305 seqtupread:5370500520
+ret:0 fetch:0 ins:0 upd:0 del:0 dbname:willow
+ +

disabled_triggers

+ +

(symlink: check_postgres_disabled_triggers) Checks on the number of disabled triggers inside the database. The --warning and --critical options are the number of such triggers found, and both default to "1", as in normal usage having disabled triggers is a dangerous event. If the database being checked is 8.3 or higher, the check is for the number of triggers that are in a 'disabled' status (as opposed to being 'always' or 'replica'). The output will show the name of the table and the name of the trigger for each disabled trigger.

+

Example 1: Make sure that there are no disabled triggers

-
-  check_postgres_disabled_triggers
+ +
check_postgres_disabled_triggers
+

For MRTG output, returns the number of disabled triggers on the first line.

-

-

-

disk_space

-

(symlink: check_postgres_disk_space) Checks on the available physical disk space used by Postgres. This action requires -that you have the executable "/bin/df" available to report on disk sizes, and it -also needs to be run as a superuser, so it can examine the data_directory -setting inside of Postgres. The --warning and --critical options are -given in either sizes or percentages or both. If using sizes, the standard unit types -are allowed: bytes, kilobytes, gigabytes, megabytes, gigabytes, terabytes, or -exabytes. Each may be abbreviated to the first letter only; no units at all -indicates 'bytes'. The default values are '90%' and '95%'.

-

This command checks the following things to determine all of the different -physical disks being used by Postgres.

-

data_directory - The disk that the main data directory is on.

-

log directory - The disk that the log files are on.

-

WAL file directory - The disk that the write-ahead logs are on (e.g. symlinked pg_xlog)

-

tablespaces - Each tablespace that is on a separate disk.

-

The output shows the total size used and available on each disk, as well as -the percentage, ordered by highest to lowest percentage used. Each item above -maps to a file system: these can be included or excluded. See the -BASIC FILTERING section for more details.

+ +

disk_space

+ +

(symlink: check_postgres_disk_space) Checks on the available physical disk space used by Postgres. This action requires that you have the executable "/bin/df" available to report on disk sizes, and it also needs to be run as a superuser (either connecting directly or switching via --role), so it can examine the data_directory setting inside of Postgres. The --warning and --critical options are given in either sizes or percentages or both. If using sizes, the standard unit types are allowed: bytes, kilobytes, gigabytes, megabytes, gigabytes, terabytes, or exabytes. Each may be abbreviated to the first letter only; no units at all indicates 'bytes'. The default values are '90%' and '95%'.

+ +

This command checks the following things to determine all of the different physical disks being used by Postgres.

+ +

data_directory - The disk that the main data directory is on.

+ +

log directory - The disk that the log files are on.

+ +

WAL file directory - The disk that the write-ahead logs are on (e.g. symlinked pg_xlog or pg_wal)

+ +

tablespaces - Each tablespace that is on a separate disk.

+ +

The output shows the total size used and available on each disk, as well as the percentage, ordered by highest to lowest percentage used. Each item above maps to a file system: these can be included or excluded. See the "BASIC FILTERING" section for more details.

+

Example 1: Make sure that no file system is over 90% for the database on port 5432.

-
-  check_postgres_disk_space --port=5432 --warning='90%' --critical='90%'
-

Example 2: Check that all file systems starting with /dev/sda are smaller than 10 GB and 11 GB (warning and critical)

-
-  check_postgres_disk_space --port=5432 --warning='10 GB' --critical='11 GB' --include="~^/dev/sda"
-

Example 4: Make sure that no file system is both over 50% and has over 15 GB

-
-  check_postgres_disk_space --critical='50% and 15 GB'
-

Example 5: Issue a warning if any file system is either over 70% full or has -more than 1T

-
-  check_postgres_disk_space --warning='1T or 75'
-

For MRTG output, returns the size in bytes of the file system on the first line, -and the name of the file system on the fourth line.

-

-

-

fsm_pages

-

(symlink: check_postgres_fsm_pages) Checks how close a cluster is to the Postgres max_fsm_pages setting. -This action will only work for databases of 8.2 or higher, and it requires the contrib -module pg_freespacemap be installed. The --warning and --critical options should be expressed -as percentages. The number of used pages in the free-space-map is determined by looking in the -pg_freespacemap_relations view, and running a formula based on the formula used for -outputting free-space-map pageslots in the vacuum verbose command. The default values are 85% for the -warning and 95% for the critical.

+ +
check_postgres_disk_space --port=5432 --warning='90%' --critical='90%'
+ +

Example 2: Check that all file systems starting with /dev/sda are smaller than 10 GB and 11 GB (warning and critical)

+ +
check_postgres_disk_space --port=5432 --warning='10 GB' --critical='11 GB' --include="~^/dev/sda"
+ +

Example 4: Make sure that no file system is both over 50% and has over 15 GB

+ +
check_postgres_disk_space --critical='50% and 15 GB'
+ +

Example 5: Issue a warning if any file system is either over 70% full or has more than 1T

+ +
check_postgres_disk_space --warning='1T or 75'
+ +

For MRTG output, returns the size in bytes of the file system on the first line, and the name of the file system on the fourth line.

+ +

fsm_pages

+ +

(symlink: check_postgres_fsm_pages) Checks how close a cluster is to the Postgres max_fsm_pages setting. This action will only work for databases of 8.2 or higher, and it requires the contrib module pg_freespacemap be installed. The --warning and --critical options should be expressed as percentages. The number of used pages in the free-space-map is determined by looking in the pg_freespacemap_relations view, and running a formula based on the formula used for outputting free-space-map pageslots in the vacuum verbose command. The default values are 85% for the warning and 95% for the critical.

+

Example 1: Give a warning when our cluster has used up 76% of the free-space pageslots, with pg_freespacemap installed in database robert

-
-  check_postgres_fsm_pages --dbname=robert --warning="76%"
+ +
check_postgres_fsm_pages --dbname=robert --warning="76%"
+

While you need to pass in the name of the database where pg_freespacemap is installed, you only need to run this check once per cluster. Also, checking this information does require obtaining special locks on the free-space-map, so it is recommend you do not run this check with short intervals.

-

For MRTG output, returns the percent of free-space-map on the first line, and the number of pages currently used on -the second line.

-

-

-

fsm_relations

-

(symlink: check_postgres_fsm_relations) Checks how close a cluster is to the Postgres max_fsm_relations setting. -This action will only work for databases of 8.2 or higher, and it requires the contrib module pg_freespacemap be -installed. The --warning and --critical options should be expressed as percentages. The number of used relations -in the free-space-map is determined by looking in the pg_freespacemap_relations view. The default values are 85% for -the warning and 95% for the critical.

+ +

For MRTG output, returns the percent of free-space-map on the first line, and the number of pages currently used on the second line.

+ +

fsm_relations

+ +

(symlink: check_postgres_fsm_relations) Checks how close a cluster is to the Postgres max_fsm_relations setting. This action will only work for databases of 8.2 or higher, and it requires the contrib module pg_freespacemap be installed. The --warning and --critical options should be expressed as percentages. The number of used relations in the free-space-map is determined by looking in the pg_freespacemap_relations view. The default values are 85% for the warning and 95% for the critical.

+

Example 1: Give a warning when our cluster has used up 80% of the free-space relations, with pg_freespacemap installed in database dylan

-
-  check_postgres_fsm_relations --dbname=dylan --warning="75%"
-

While you need to pass in the name of the database where pg_freespacemap is installed, you only need to run this check -once per cluster. Also, -checking this information does require obtaining special locks on the free-space-map, so it is recommend you do not -run this check with short intervals.

-

For MRTG output, returns the percent of free-space-map on the first line, the number of relations currently used on -the second line.

-

-

-

hitratio

-

(symlink: check_postgres_hitratio) Checks the hit ratio of all databases and complains when they are too low. -There is no need to run this command more than once per database cluster. -Databases can be filtered with -the --include and --exclude options. See the BASIC FILTERING section -for more details. -They can also be filtered by the owner of the database with the ---includeuser and --excludeuser options. -See the USER NAME FILTERING section for more details.

-

The warning and critical options should be specified as percentages. There are not -defaults for this action: the warning and critical must be specified. The warning value -cannot be greater than the critical value. The output returns all databases sorted by -hitratio, smallest first.

+ +
check_postgres_fsm_relations --dbname=dylan --warning="75%"
+ +

While you need to pass in the name of the database where pg_freespacemap is installed, you only need to run this check once per cluster. Also, checking this information does require obtaining special locks on the free-space-map, so it is recommend you do not run this check with short intervals.

+ +

For MRTG output, returns the percent of free-space-map on the first line, the number of relations currently used on the second line.

+ +

hitratio

+ +

(symlink: check_postgres_hitratio) Checks the hit ratio of all databases and complains when they are too low. There is no need to run this command more than once per database cluster. Databases can be filtered with the --include and --exclude options. See the "BASIC FILTERING" section for more details. They can also be filtered by the owner of the database with the --includeuser and --excludeuser options. See the "USER NAME FILTERING" section for more details.

+ +

The warning and critical options should be specified as percentages. There are not defaults for this action: the warning and critical must be specified. The warning value cannot be greater than the critical value. The output returns all databases sorted by hitratio, smallest first.

+

Example: Warn if any database on host flagg is less than 90% in hitratio, and critical if less then 80%.

-
-  check_postgres_hitratio --host=flagg --warning='90%' --critical='80%'
-

For MRTG output, returns the percentage of the database with the smallest hitratio on the first line, -and the name of the database on the fourth line.

-

-

-

hot_standby_delay

-

(symlink: check_hot_standby_delay) Checks the streaming replication lag by computing the delta -between the xlog position of a master server and the one of the slaves connected to it. The slave_ -server must be in hot_standby (e.g. read only) mode, therefore the minimum version to use this_ -action is Postgres 9.0. The --warning and --critical options are the delta between xlog -location. These values should match the volume of transactions needed to have the streaming -replication disconnect from the master because of too much lag.

-

You must provide information on how to reach the second database by a connection -parameter ending in the number 2, such as "--dbport2=5543". If if it not given, -the action fails.

-

-

-

index_size

-

-

-

table_size

-

-

-

relation_size

-

(symlinks: check_postgres_index_size, check_postgres_table_size, and check_postgres_relation_size) -The actions table_size and index_size are simply variations of the -relation_size action, which checks for a relation that has grown too big. -Relations (in other words, tables and indexes) can be filtered with the ---include and --exclude options. See the BASIC FILTERING section -for more details. Relations can also be filtered by the user that owns them, -by using the --includeuser and --excludeuser options. -See the USER NAME FILTERING section for more details.

-

The values for the --warning and --critical options are file sizes, and -may have units of bytes, kilobytes, megabytes, gigabytes, terabytes, or exabytes. -Each can be abbreviated to the first letter. If no units are given, bytes are -assumed. There are no default values: both the warning and the critical option -must be given. The return text shows the size of the largest relation found.

-

If the --showperf option is enabled, all of the relations with their sizes -will be given. To prevent this, it is recommended that you set the ---perflimit option, which will cause the query to do a -ORDER BY size DESC LIMIT (perflimit).

+ +
check_postgres_hitratio --host=flagg --warning='90%' --critical='80%'
+ +

For MRTG output, returns the percentage of the database with the smallest hitratio on the first line, and the name of the database on the fourth line.

+ +

hot_standby_delay

+ +

(symlink: check_hot_standby_delay) Checks the streaming replication lag by computing the delta between the current xlog position of a master server and the replay location of a slave connected to it. The slave server must be in hot_standby (e.g. read only) mode, therefore the minimum version to use this action is Postgres 9.0. The --warning and --critical options are the delta between the xlog locations. Since these values are byte offsets in the WAL they should match the expected transaction volume of your application to prevent false positives or negatives.

+ +

The first "--dbname", "--host", and "--port", etc. options are considered the master; the second belongs to the slave.

+ +

Byte values should be based on the volume of transactions needed to have the streaming replication disconnect from the master because of too much lag, determined by the Postgres configuration variable wal_keep_segments. For units of time, valid units are 'seconds', 'minutes', 'hours', or 'days'. Each may be written singular or abbreviated to just the first letter. When specifying both, in the form 'bytes and time', both conditions must be true for the threshold to be met.

+ +

You must provide information on how to reach the databases by providing a comma separated list to the --dbhost and --dbport parameters, such as "--dbport=5432,5543". If not given, the action fails.

+ +

Example 1: Warn a database with a local replica on port 5433 is behind on any xlog replay at all

+ +
check_hot_standby_delay --dbport=5432,5433 --warning='1'
+ +

Example 2: Give a critical if the last transaction replica1 receives is more than 10 minutes ago

+ +
check_hot_standby_delay --dbhost=master,replica1 --critical='10 min'
+ +

Example 3: Allow replica1 to be 1 WAL segment behind, if the master is momentarily seeing more activity than the streaming replication connection can handle, or 10 minutes behind, if the master is seeing very little activity and not processing any transactions, but not both, which would indicate a lasting problem with the replication connection.

+ +
check_hot_standby_delay --dbhost=master,replica1 --warning='1048576 and 2 min' --critical='16777216 and 10 min'
+ +

relation_size

+ +

index_size

+ +

table_size

+ +

indexes_size

+ +

total_relation_size

+ +

(symlinks: check_postgres_relation_size, check_postgres_index_size, check_postgres_table_size, check_postgres_indexes_size, and check_postgres_total_relation_size)

+ +

The actions relation_size and index_size check for a relation (table, index, materialized view), respectively an index that has grown too big, using the pg_relation_size() function.

+ +

The action table_size checks tables and materialized views using pg_table_size(), i.e. including relation forks and TOAST table.

+ +

The action indexes_size checks tables and materialized views for the size of the attached indexes using pg_indexes_size().

+ +

The action total_relation_size checks relations using pg_total_relation_size(), i.e. including relation forks, indexes and TOAST table.

+ +

Relations can be filtered with the --include and --exclude options. See the "BASIC FILTERING" section for more details. Relations can also be filtered by the user that owns them, by using the --includeuser and --excludeuser options. See the "USER NAME FILTERING" section for more details.

+ +

The values for the --warning and --critical options are file sizes, and may have units of bytes, kilobytes, megabytes, gigabytes, terabytes, or exabytes. Each can be abbreviated to the first letter. If no units are given, bytes are assumed. There are no default values: both the warning and the critical option must be given. The return text shows the size of the largest relation found.

+ +

If the --showperf option is enabled, all of the relations with their sizes will be given. To prevent this, it is recommended that you set the --perflimit option, which will cause the query to do a ORDER BY size DESC LIMIT (perflimit).

+

Example 1: Give a critical if any table is larger than 600MB on host burrick.

-
-  check_postgres_table_size --critical='600 MB' --warning='600 MB' --host=burrick
+ +
check_postgres_table_size --critical='600 MB' --warning='600 MB' --host=burrick
+

Example 2: Warn if the table products is over 4 GB in size, and give a critical at 4.5 GB.

-
-  check_postgres_table_size --host=burrick --warning='4 GB' --critical='4.5 GB' --include=products
+ +
check_postgres_table_size --host=burrick --warning='4 GB' --critical='4.5 GB' --include=products
+

Example 3: Warn if any index not owned by postgres goes over 500 MB.

-
-  check_postgres_index_size --port=5432 --excludeuser=postgres -w 500MB -c 600MB
-

For MRTG output, returns the size in bytes of the largest relation, and the name of the database -and relation as the fourth line.

-

-

-

last_analyze

-

-

-

last_vacuum

-

-

-

last_autoanalyze

-

-

-

last_autovacuum

-

(symlinks: check_postgres_last_analyze, check_postgres_last_vacuum, -check_postgres_last_autoanalyze, and check_postgres_last_autovacuum) -Checks how long it has been since vacuum (or analyze) was last run on each -table in one or more databases. Use of these actions requires that the target -database is version 8.3 or greater, or that the version is 8.2 and the -configuration variable stats_row_level has been enabled. Tables can be filtered with the ---include and --exclude options. See the BASIC FILTERING section -for more details. -Tables can also be filtered by their owner by use of the ---includeuser and --excludeuser options. -See the USER NAME FILTERING section for more details.

-

The units for --warning and --critical are specified as times. -Valid units are seconds, minutes, hours, and days; all can be abbreviated -to the first letter. If no units are given, 'seconds' are assumed. The -default values are '1 day' and '2 days'. Please note that there are cases -in which this field does not get automatically populated. If certain tables -are giving you problems, make sure that they have dead rows to vacuum, -or just exclude them from the test.

-

The schema named 'information_schema' is excluded from this test, as the only tables -it contains are small and do not change.

-

Note that the non-'auto' versions will also check on the auto versions as well. In other words, -using last_vacuum will report on the last vacuum, whether it was a normal vacuum, or -one run by the autovacuum daemon.

-

Example 1: Warn if any table has not been vacuumed in 3 days, and give a -critical at a week, for host wormwood

-
-  check_postgres_last_vacuum --host=wormwood --warning='3d' --critical='7d'
-

Example 2: Same as above, but skip tables belonging to the users 'eve' or 'mallory'

-
-  check_postgres_last_vacuum --host=wormwood --warning='3d' --critical='7d' --excludeusers=eve,mallory
-

For MRTG output, returns (on the first line) the LEAST amount of time in seconds since a table was -last vacuumed or analyzed. The fourth line returns the name of the database and name of the table.

-

-

-

listener

-

(symlink: check_postgres_listener) Confirm that someone is listening for one or more -specific strings (using the LISTEN/NOTIFY system), by looking at the pg_listener table. -Only one of warning or critical is needed. The format is a simple string representing the -LISTEN target, or a tilde character followed by a string for a regular expression check. -Note that this check will not work on versions of Postgres 9.0 or higher.

+ +
check_postgres_index_size --port=5432 --excludeuser=postgres -w 500MB -c 600MB
+ +

For MRTG output, returns the size in bytes of the largest relation, and the name of the database and relation as the fourth line.

+ +

last_analyze

+ +

last_vacuum

+ +

last_autoanalyze

+ +

last_autovacuum

+ +

(symlinks: check_postgres_last_analyze, check_postgres_last_vacuum, check_postgres_last_autoanalyze, and check_postgres_last_autovacuum) Checks how long it has been since vacuum (or analyze) was last run on each table in one or more databases. Use of these actions requires that the target database is version 8.3 or greater, or that the version is 8.2 and the configuration variable stats_row_level has been enabled. Tables can be filtered with the --include and --exclude options. See the "BASIC FILTERING" section for more details. Tables can also be filtered by their owner by use of the --includeuser and --excludeuser options. See the "USER NAME FILTERING" section for more details.

+ +

The units for --warning and --critical are specified as times. Valid units are seconds, minutes, hours, and days; all can be abbreviated to the first letter. If no units are given, 'seconds' are assumed. The default values are '1 day' and '2 days'. Please note that there are cases in which this field does not get automatically populated. If certain tables are giving you problems, make sure that they have dead rows to vacuum, or just exclude them from the test.

+ +

The schema named 'information_schema' is excluded from this test, as the only tables it contains are small and do not change.

+ +

Note that the non-'auto' versions will also check on the auto versions as well. In other words, using last_vacuum will report on the last vacuum, whether it was a normal vacuum, or one run by the autovacuum daemon.

+ +

Example 1: Warn if any table has not been vacuumed in 3 days, and give a critical at a week, for host wormwood

+ +
check_postgres_last_vacuum --host=wormwood --warning='3d' --critical='7d'
+ +

Example 2: Same as above, but skip tables belonging to the users 'eve' or 'mallory'

+ +
check_postgres_last_vacuum --host=wormwood --warning='3d' --critical='7d' --excludeuser=eve,mallory
+ +

For MRTG output, returns (on the first line) the LEAST amount of time in seconds since a table was last vacuumed or analyzed. The fourth line returns the name of the database and name of the table.

+ +

listener

+ +

(symlink: check_postgres_listener) Confirm that someone is listening for one or more specific strings (using the LISTEN/NOTIFY system), by looking at the pg_listener table. Only one of warning or critical is needed. The format is a simple string representing the LISTEN target, or a tilde character followed by a string for a regular expression check. Note that this check will not work on versions of Postgres 9.0 or higher.

+

Example 1: Give a warning if nobody is listening for the string bucardo_mcp_ping on ports 5555 and 5556

-
-  check_postgres_listener --port=5555,5556 --warning=bucardo_mcp_ping
-

Example 2: Give a critical if there are no active LISTEN requests matching 'grimm' on database oskar

-
-  check_postgres_listener --db oskar --critical=~grimm
-

For MRTG output, returns a 1 or a 0 on the first, indicating success or failure. The name of the notice must -be provided via the --mrtg option.

-

-

-

locks

-

(symlink: check_postgres_locks) Check the total number of locks on one or more databases. There is no -need to run this more than once per database cluster. Databases can be filtered -with the --include and --exclude options. See the BASIC FILTERING section -for more details.

-

The --warning and --critical options can be specified as simple numbers, -which represent the total number of locks, or they can be broken down by type of lock. -Valid lock names are 'total', 'waiting', or the name of a lock type used by Postgres. -These names are case-insensitive and do not need the "lock" part on the end, -so exclusive will match 'ExclusiveLock'. The format is name=number, with different -items separated by colons or semicolons (or any other symbol).

+ +
check_postgres_listener --port=5555,5556 --warning=bucardo_mcp_ping
+ +

Example 2: Give a critical if there are no active LISTEN requests matching 'grimm' on database oskar

+ +
check_postgres_listener --db oskar --critical=~grimm
+ +

For MRTG output, returns a 1 or a 0 on the first, indicating success or failure. The name of the notice must be provided via the --mrtg option.

+ +

locks

+ +

(symlink: check_postgres_locks) Check the total number of locks on one or more databases. There is no need to run this more than once per database cluster. Databases can be filtered with the --include and --exclude options. See the "BASIC FILTERING" section for more details.

+ +

The --warning and --critical options can be specified as simple numbers, which represent the total number of locks, or they can be broken down by type of lock. Valid lock names are 'total', 'waiting', or the name of a lock type used by Postgres. These names are case-insensitive and do not need the "lock" part on the end, so exclusive will match 'ExclusiveLock'. The format is name=number, with different items separated by colons or semicolons (or any other symbol).

+

Example 1: Warn if the number of locks is 100 or more, and critical if 200 or more, on host garrett

-
-  check_postgres_locks --host=garrett --warning=100 --critical=200
+ +
check_postgres_locks --host=garrett --warning=100 --critical=200
+

Example 2: On the host artemus, warn if 200 or more locks exist, and give a critical if over 250 total locks exist, or if over 20 exclusive locks exist, or if over 5 connections are waiting for a lock.

-
-  check_postgres_locks --host=artemus --warning=200 --critical="total=250:waiting=5:exclusive=20"
+ +
check_postgres_locks --host=artemus --warning=200 --critical="total=250:waiting=5:exclusive=20"
+

For MRTG output, returns the number of locks on the first line, and the name of the database on the fourth line.

-

-

-

logfile

-

(symlink: check_postgres_logfile) Ensures that the logfile is in the expected location and is being logged to. -This action issues a command that throws an error on each database it is -checking, and ensures that the message shows up in the logs. It scans the -various log_* settings inside of Postgres to figure out where the logs should be. -If you are using syslog, it does a rough (but not foolproof) scan of -/etc/syslog.conf. Alternatively, you can provide the name of the logfile -with the --logfile option. This is especially useful if the logs have a -custom rotation scheme driven be an external program. The --logfile option -supports the following escape characters: %Y %m %d %H, which represent -the current year, month, date, and hour respectively. An error is always -reported as critical unless the warning option has been passed in as a non-zero -value. Other than that specific usage, the --warning and --critical -options should not be used.

+ +

lockwait

+ +

(symlink: check_postgres_lockwait) Check if there are blocking blocks and for how long. There is no need to run this more than once per database cluster. Databases can be filtered with the --include and --exclude options. See the "BASIC FILTERING" section for more details.

+ +

The --warning and --critical options is time, which represent the time for which the lock has been blocking.

+ +

Example 1: Warn if a lock has been blocking for more than a minute, critcal if for more than 2 minutes

+ +
check_postgres_lockwait --host=garrett --warning='1 min' --critical='2 min'
+ +

For MRTG output, returns the number of blocked sessions.

+ +

logfile

+ +

(symlink: check_postgres_logfile) Ensures that the logfile is in the expected location and is being logged to. This action issues a command that throws an error on each database it is checking, and ensures that the message shows up in the logs. It scans the various log_* settings inside of Postgres to figure out where the logs should be. If you are using syslog, it does a rough (but not foolproof) scan of /etc/syslog.conf. Alternatively, you can provide the name of the logfile with the --logfile option. This is especially useful if the logs have a custom rotation scheme driven be an external program. The --logfile option supports the following escape characters: %Y %m %d %H, which represent the current year, month, date, and hour respectively. An error is always reported as critical unless the warning option has been passed in as a non-zero value. Other than that specific usage, the --warning and --critical options should not be used.

+

Example 1: On port 5432, ensure the logfile is being written to the file /home/greg/pg8.2.log

-
-  check_postgres_logfile --port=5432 --logfile=/home/greg/pg8.2.log
+ +
check_postgres_logfile --port=5432 --logfile=/home/greg/pg8.2.log
+

Example 2: Same as above, but raise a warning, not a critical

-
-  check_postgres_logfile --port=5432 --logfile=/home/greg/pg8.2.log -w 1
-

For MRTG output, returns a 1 or 0 on the first line, indicating success or failure. In case of a -failure, the fourth line will provide more detail on the failure encountered.

-

-

-

new_version_bc

-

(symlink: check_postgres_new_version_bc) Checks if a newer version of the Bucardo -program is available. The current version is obtained by running bucardo_ctl --version. -If a major upgrade is available, a warning is returned. If a revision upgrade is -available, a critical is returned. (Bucardo is a master to slave, and master to master -replication system for Postgres: see http://bucardo.org for more information). -See also the information on the --get_method option.

-

-

-

new_version_box

-

(symlink: check_postgres_new_version_box) Checks if a newer version of the boxinfo -program is available. The current version is obtained by running boxinfo.pl --version. -If a major upgrade is available, a warning is returned. If a revision upgrade is -available, a critical is returned. (boxinfo is a program for grabbing important -information from a server and putting it into a HTML format: see -http://bucardo.org/wiki/boxinfo for more information). See also the information on -the --get_method option.

-

-

-

new_version_cp

-

(symlink: check_postgres_new_version_cp) Checks if a newer version of this program -(check_postgres.pl) is available, by grabbing the version from a small text file -on the main page of the home page for the project. Returns a warning if the returned -version does not match the one you are running. Recommended interval to check is -once a day. See also the information on the --get_method option.

-

-

-

new_version_pg

-

(symlink: check_postgres_new_version_pg) Checks if a newer revision of Postgres -exists for each database connected to. Note that this only checks for revision, e.g. -going from 8.3.6 to 8.3.7. Revisions are always 100% binary compatible and involve no -dump and restore to upgrade. Revisions are made to address bugs, so upgrading as soon -as possible is always recommended. Returns a warning if you do not have the latest revision. -It is recommended this check is run at least once a day. See also the information on -the --get_method option.

-

-

-

new_version_tnm

-

(symlink: check_postgres_new_version_tnm) Checks if a newer version of the -tail_n_mail program is available. The current version is obtained by running -tail_n_mail --version. If a major upgrade is available, a warning is returned. If a -revision upgrade is available, a critical is returned. (tail_n_mail is a log monitoring -tool that can send mail when interesting events appear in your Postgres logs. -See: http://bucardo.org/wiki/Tail_n_mail for more information). -See also the information on the --get_method option.

-

-

-

pgb_pool_cl_active

-

-

-

pgb_pool_cl_waiting

-

-

-

pgb_pool_sv_active

-

-

-

pgb_pool_sv_idle

-

-

-

pgb_pool_sv_used

-

-

-

pgb_pool_sv_tested

-

-

-

pgb_pool_sv_login

-

-

-

pgb_pool_maxwait

-

(symlinks: check_postgres_pgb_pool_cl_active, check_postgres_pgb_pool_cl_waiting, -check_postgres_pgb_pool_sv_active, check_postgres_pgb_pool_sv_idle, -check_postgres_pgb_pool_sv_used, check_postgres_pgb_pool_sv_tested, -check_postgres_pgb_pool_sv_login, and check_postgres_pgb_pool_maxwait)

-

Examines pgbouncer's pool statistics. Each pool has a set of "client" -connections, referring to connections from external clients, and "server" -connections, referring to connections to PostgreSQL itself. The related -check_postgres actions are prefixed by "cl_" and "sv_", respectively. Active -client connections are those connections currently linked with an active server -connection. Client connections may also be "waiting", meaning they have not yet -been allocated a server connection. Server connections are "active" (linked to -a client), "idle" (standing by for a client connection to link with), "used" -(just unlinked from a client, and not yet returned to the idle pool), "tested" -(currently being tested) and "login" (in the process of logging in). The -maxwait value shows how long in seconds the oldest waiting client connection -has been waiting.

-

-

-

pgbouncer_backends

-

(symlink: check_postgres_pgbouncer_backends) Checks the current number of -connections for one or more databases through pgbouncer, and optionally -compares it to the maximum allowed, which is determined by the pgbouncer -configuration variable max_client_conn. The --warning and --critical -options can take one of three forms. First, a simple number can be given, -which represents the number of connections at which the alert will be given. -This choice does not use the max_connections setting. Second, the -percentage of available connections can be given. Third, a negative number can -be given which represents the number of connections left until -max_connections is reached. The default values for --warning and ---critical are '90%' and '95%'. You can also filter the databases by use -of the --include and --exclude options. See the BASIC FILTERING -section for more details.

-

To view only non-idle processes, you can use the --noidle argument. Note -that the user you are connecting as must be a superuser for this to work -properly.

-

Example 1: Give a warning when the number of connections on host quirm reaches -120, and a critical if it reaches 150.

-
-  check_postgres_pgbouncer_backends --host=quirm --warning=120 --critical=150 -p 6432 -u pgbouncer
-

Example 2: Give a critical when we reach 75% of our max_connections setting on -hosts lancre or lancre2.

-
-  check_postgres_pgbouncer_backends --warning='75%' --critical='75%' --host=lancre,lancre2 -p 6432 -u pgbouncer
-

Example 3: Give a warning when there are only 10 more connection slots left on -host plasmid, and a critical when we have only 5 left.

-
-  check_postgres_pgbouncer_backends --warning=-10 --critical=-5 --host=plasmid -p 6432 -u pgbouncer
-

For MRTG output, the number of connections is reported on the first line, and -the fourth line gives the name of the database, plus the current -max_client_conn. If more than one database has been queried, the one with the -highest number of connections is output.

-

-

-

pgbouncer_checksum

-

(symlink: check_postgres_pgbouncer_checksum) Checks that all the -pgBouncer settings are the same as last time you checked. -This is done by generating a checksum of a sorted list of setting names and -their values. Note that you shouldn't specify the database name, it will -automatically default to pgbouncer. Either the --warning or the --critical option -should be given, but not both. The value of each one is the checksum, a -32-character hexadecimal value. You can run with the special --critical=0 option -to find out an existing checksum.

-

This action requires the Digest::MD5 module.

-

Example 1: Find the initial checksum for pgbouncer configuration on port 6432 using the default user (usually postgres)

-
-  check_postgres_pgbouncer_checksum --port=6432 --critical=0
-

Example 2: Make sure no settings have changed and warn if so, using the checksum from above.

-
-  check_postgres_pgbouncer_checksum --port=6432 --warning=cd2f3b5e129dc2b4f5c0f6d8d2e64231
-

For MRTG output, returns a 1 or 0 indicating success of failure of the checksum to match. A -checksum must be provided as the --mrtg argument. The fourth line always gives the -current checksum.

-

-

-

pgagent_jobs

-

(symlink: check_postgres_pgagent_jobs) Checks that all the pgAgent jobs -that have executed in the preceding interval of time have succeeded. This is -done by checking for any steps that have a non-zero result.

-

Either --warning or --critical, or both, may be specified as times, and -jobs will be checked for failures withing the specified periods of time before -the current time. Valid units are seconds, minutes, hours, and days; all can -be abbreviated to the first letter. If no units are given, 'seconds' are -assumed.

-

Example 1: Give a critical when any jobs executed in the last day have failed.

-
-  check_postgres_pgagent_jobs --critical=1d
-

Example 2: Give a warning when any jobs executed in the last week have failed.

-
-  check_postgres_pgagent_jobs --warning=7d
-

Example 3: Give a critical for jobs that have failed in the last 2 hours and a -warning for jobs that have failed in the last 4 hours:

-
-  check_postgres_pgagent_jobs --critical=2h --warning=4h
-

-

-

prepared_txns

-

(symlink: check_postgres_prepared_txns) Check on the age of any existing prepared transactions. -Note that most people will NOT use prepared transactions, as they are part of two-part commit -and complicated to maintain. They should also not be confused with prepared STATEMENTS, which is -what most people think of when they hear prepare. The default value for a warning is 1 second, to -detect any use of prepared transactions, which is probably a mistake on most systems. Warning and -critical are the number of seconds a prepared transaction has been open before an alert is given.

-

Example 1: Give a warning on detecting any prepared transactions:

-
-  check_postgres_prepared_txns -w 0
-

Example 2: Give a critical if any prepared transaction has been open longer than 10 seconds, but allow -up to 360 seconds for the database 'shrike':

-
-  check_postgres_prepared_txns --critical=10 --exclude=shrike
-  check_postgres_prepared_txns --critical=360 --include=shrike
-

For MRTG output, returns the number of seconds the oldest transaction has been open as the first line, -and which database is came from as the final line.

-

-

-

query_runtime

-

(symlink: check_postgres_query_runtime) Checks how long a specific query takes to run, by executing a "EXPLAIN ANALYZE" -against it. The --warning and --critical options are the maximum amount of -time the query should take. Valid units are seconds, minutes, and hours; any can be -abbreviated to the first letter. If no units are given, 'seconds' are assumed. -Both the warning and the critical option must be given. The name of the view or -function to be run must be passed in to the --queryname option. It must consist -of a single word (or schema.word), with optional parens at the end.

-

Example 1: Give a critical if the function named "speedtest" fails to run in 10 seconds or less.

-
-  check_postgres_query_runtime --queryname='speedtest()' --critical=10 --warning=10
-

For MRTG output, reports the time in seconds for the query to complete on the first line. The fourth -line lists the database.

-

-

-

query_time

-

(symlink: check_postgres_query_time) Checks the length of running queries on one or more databases. -There is no need to run this more than once on the same database cluster. Note that -this already excludes queries that are "idle in transaction". Databases can be filtered -by using the --include and --exclude options. See the BASIC FILTERING -section for more details. You can also filter on the user running the -query with the --includeuser and --excludeuser options. -See the USER NAME FILTERING section for more details.

-

The values for the --warning and --critical options are amounts of -time, and default to '2 minutes' and '5 minutes' respectively. Valid units -are 'seconds', 'minutes', 'hours', or 'days'. Each may be written singular or -abbreviated to just the first letter. If no units are given, the unit is -assumed to be seconds.

-

This action requires Postgres 8.1 or better.

-

Example 1: Give a warning if any query has been running longer than 3 minutes, and a critical if longer than 5 minutes.

-
-  check_postgres_query_time --port=5432 --warning='3 minutes' --critical='5 minutes'
-

Example 2: Using default values (2 and 5 minutes), check all databases except those starting with 'template'.

-
-  check_postgres_query_time --port=5432 --exclude=~^template
-

Example 3: Warn if user 'don' has a query running over 20 seconds

-
-  check_postgres_query_time --port=5432 --includeuser=don --warning=20s
-

For MRTG output, returns the length in seconds of the longest running query on the first line. The fourth -line gives the name of the database.

-

-

-

replicate_row

-

(symlink: check_postgres_replicate_row) Checks that master-slave replication is working to one or more slaves. -The slaves are specified the same as the normal databases, except with -the number 2 at the end of them, so "--port2" instead of "--port", etc. -The values or the --warning and --critical options are units of time, and -at least one must be provided (no defaults). Valid units are 'seconds', 'minutes', 'hours', -or 'days'. Each may be written singular or abbreviated to just the first letter. -If no units are given, the units are assumed to be seconds.

-

This check updates a single row on the master, and then measures how long it -takes to be applied to the slaves. To do this, you need to pick a table that -is being replicated, then find a row that can be changed, and is not going -to be changed by any other process. A specific column of this row will be changed -from one value to another. All of this is fed to the repinfo option, and should -contain the following options, separated by commas: table name, primary key, key id, -column, first value, second value.

-

Example 1: Slony is replicating a table named 'orders' from host 'alpha' to -host 'beta', in the database 'sales'. The primary key of the table is named -id, and we are going to test the row with an id of 3 (which is historical and -never changed). There is a column named 'salesrep' that we are going to toggle -from a value of 'slon' to 'nols' to check on the replication. We want to throw -a warning if the replication does not happen within 10 seconds.

-
-  check_postgres_replicate_row --host=alpha --dbname=sales --host2=beta 
-  --dbname2=sales --warning=10 --repinfo=orders,id,3,salesrep,slon,nols
-

Example 2: Bucardo is replicating a table named 'receipt' from host 'green' -to hosts 'red', 'blue', and 'yellow'. The database for both sides is 'public'. -The slave databases are running on port 5455. The primary key is named 'receipt_id', -the row we want to use has a value of 9, and the column we want to change for the -test is called 'zone'. We'll toggle between 'north' and 'south' for the value of -this column, and throw a critical if the change is not on all three slaves within 5 seconds.

-
- check_postgres_replicate_row --host=green --port2=5455 --host2=red,blue,yellow
-  --critical=5 --repinfo=receipt,receipt_id,9,zone,north,south
-

For MRTG output, returns on the first line the time in seconds the replication takes to finish. -The maximum time is set to 4 minutes 30 seconds: if no replication has taken place in that long -a time, an error is thrown.

-

-

-

same_schema

-

(symlink: check_postgres_same_schema) Verifies that two or more databases are identical as far as their -schema (but not the data within). This is particularly handy for making sure your slaves have not -been modified or corrupted in any way when using master to slave replication. Unlike most other -actions, this has no warning or critical criteria - the databases are either in sync, or are not. -If they are different, a detailed list of the differences is presented.

-

You may want to exclude or filter out certain differences. The way to do this is to add strings -to the --filter option. To exclude a type of object, use "noname", where 'name' is the type of -object, for example, "noschema". To exclude objects of a certain type by a regular expression against -their name, use "noname=regex". See the examples below for a better understanding.

-

The types of objects that can be filtered include:

-
-
user
-
schema
+
check_postgres_logfile --port=5432 --logfile=/home/greg/pg8.2.log -w 1
-
table
+

For MRTG output, returns a 1 or 0 on the first line, indicating success or failure. In case of a failure, the fourth line will provide more detail on the failure encountered.

-
view
+

new_version_bc

-
index
+

(symlink: check_postgres_new_version_bc) Checks if a newer version of the Bucardo program is available. The current version is obtained by running bucardo_ctl --version. If a major upgrade is available, a warning is returned. If a revision upgrade is available, a critical is returned. (Bucardo is a master to slave, and master to master replication system for Postgres: see https://bucardo.org/ for more information). See also the information on the --get_method option.

-
sequence
+

new_version_box

-
constraint
+

(symlink: check_postgres_new_version_box) Checks if a newer version of the boxinfo program is available. The current version is obtained by running boxinfo.pl --version. If a major upgrade is available, a warning is returned. If a revision upgrade is available, a critical is returned. (boxinfo is a program for grabbing important information from a server and putting it into a HTML format: see https://bucardo.org/Boxinfo/ for more information). See also the information on the --get_method option.

-
trigger
+

new_version_cp

-
function
+

(symlink: check_postgres_new_version_cp) Checks if a newer version of this program (check_postgres.pl) is available, by grabbing the version from a small text file on the main page of the home page for the project. Returns a warning if the returned version does not match the one you are running. Recommended interval to check is once a day. See also the information on the --get_method option.

+ +

new_version_pg

+ +

(symlink: check_postgres_new_version_pg) Checks if a newer revision of Postgres exists for each database connected to. Note that this only checks for revision, e.g. going from 8.3.6 to 8.3.7. Revisions are always 100% binary compatible and involve no dump and restore to upgrade. Revisions are made to address bugs, so upgrading as soon as possible is always recommended. Returns a warning if you do not have the latest revision. It is recommended this check is run at least once a day. See also the information on the --get_method option.

+ +

new_version_tnm

+ +

(symlink: check_postgres_new_version_tnm) Checks if a newer version of the tail_n_mail program is available. The current version is obtained by running tail_n_mail --version. If a major upgrade is available, a warning is returned. If a revision upgrade is available, a critical is returned. (tail_n_mail is a log monitoring tool that can send mail when interesting events appear in your Postgres logs. See: https://bucardo.org/tail_n_mail/ for more information). See also the information on the --get_method option.

+ +

partman_premake

+ +

(symlink: check_postgres_partman_premake) Checks if all partitions that pg_parman's maintenance routine should have created are actually present. Monthly and daily intervals are supported.

+ +

pgb_pool_cl_active

+ +

pgb_pool_cl_waiting

+ +

pgb_pool_sv_active

+ +

pgb_pool_sv_idle

+ +

pgb_pool_sv_used

+ +

pgb_pool_sv_tested

+ +

pgb_pool_sv_login

+ +

pgb_pool_maxwait

+ +

(symlinks: check_postgres_pgb_pool_cl_active, check_postgres_pgb_pool_cl_waiting, check_postgres_pgb_pool_sv_active, check_postgres_pgb_pool_sv_idle, check_postgres_pgb_pool_sv_used, check_postgres_pgb_pool_sv_tested, check_postgres_pgb_pool_sv_login, and check_postgres_pgb_pool_maxwait)

+ +

Examines pgbouncer's pool statistics. Each pool has a set of "client" connections, referring to connections from external clients, and "server" connections, referring to connections to PostgreSQL itself. The related check_postgres actions are prefixed by "cl_" and "sv_", respectively. Active client connections are those connections currently linked with an active server connection. Client connections may also be "waiting", meaning they have not yet been allocated a server connection. Server connections are "active" (linked to a client), "idle" (standing by for a client connection to link with), "used" (just unlinked from a client, and not yet returned to the idle pool), "tested" (currently being tested) and "login" (in the process of logging in). The maxwait value shows how long in seconds the oldest waiting client connection has been waiting.

+ +

pgbouncer_backends

+ +

(symlink: check_postgres_pgbouncer_backends) Checks the current number of connections for one or more databases through pgbouncer, and optionally compares it to the maximum allowed, which is determined by the pgbouncer configuration variable max_client_conn. The --warning and --critical options can take one of three forms. First, a simple number can be given, which represents the number of connections at which the alert will be given. This choice does not use the max_connections setting. Second, the percentage of available connections can be given. Third, a negative number can be given which represents the number of connections left until max_connections is reached. The default values for --warning and --critical are '90%' and '95%'. You can also filter the databases by use of the --include and --exclude options. See the "BASIC FILTERING" section for more details.

+ +

To view only non-idle processes, you can use the --noidle argument. Note that the user you are running as (either connecting directly or switching via --role) must be a superuser for this to work properly.

+ +

Example 1: Give a warning when the number of connections on host quirm reaches 120, and a critical if it reaches 150.

+ +
check_postgres_pgbouncer_backends --host=quirm --warning=120 --critical=150 -p 6432 -u pgbouncer
+ +

Example 2: Give a critical when we reach 75% of our max_connections setting on hosts lancre or lancre2.

+ +
check_postgres_pgbouncer_backends --warning='75%' --critical='75%' --host=lancre,lancre2 -p 6432 -u pgbouncer
+ +

Example 3: Give a warning when there are only 10 more connection slots left on host plasmid, and a critical when we have only 5 left.

+ +
check_postgres_pgbouncer_backends --warning=-10 --critical=-5 --host=plasmid -p 6432 -u pgbouncer
+ +

For MRTG output, the number of connections is reported on the first line, and the fourth line gives the name of the database, plus the current max_client_conn. If more than one database has been queried, the one with the highest number of connections is output.

+ +

pgbouncer_checksum

+ +

(symlink: check_postgres_pgbouncer_checksum) Checks that all the pgBouncer settings are the same as last time you checked. This is done by generating a checksum of a sorted list of setting names and their values. Note that you shouldn't specify the database name, it will automatically default to pgbouncer. Either the --warning or the --critical option should be given, but not both. The value of each one is the checksum, a 32-character hexadecimal value. You can run with the special --critical=0 option to find out an existing checksum.

-
-

The filter option "noposition" prevents verification of the position of -columns within a table.

-

The filter option "nofuncbody" prevents comparison of the bodies of all -functions.

-

The filter option "noperm" prevents comparison of object permissions.

-

To provide the second database, just append the differences to the first one -by a call to the appropriate connection argument. For example, to compare -databases on hosts alpha and bravo, use "--dbhost=alpha,bravo". Also see the -examples below.

-

If only a single host is given, it is assumed we are doing a "time-based" report. -The first time this is run a snapshot of all the items in the database is -saved to a local file. When you run it again, that snapshot is read in and -becomes "database #2" and is compared to the current database.

-

To replace the old stored file with the new version, use the --replace argument.

-

To enable snapshots at various points in time, you can use the "--suffix" -argument to make the filenames unique to each run. See the examples below.

-

Example 1: Verify that two databases on hosts star and line are the same:

-
-  check_postgres_same_schema --dbhost=star,line
-

Example 2: Same as before, but exclude any triggers with "slony" in their name

-
-  check_postgres_same_schema --dbhost=star,line --filter="notrigger=slony"
-

Example 3: Same as before, but also exclude all indexes

-
-  check_postgres_same_schema --dbhost=star,line --filter="notrigger=slony noindexes"
-

Example 4: Check differences for the database "battlestar" on different ports

-
-  check_postgres_same_schema --dbname=battlestar --dbport=5432,5544
-

Example 5: Create a daily and weekly snapshot file

-
-  check_postgres_same_schema --dbname=cylon --suffix=daily
-  check_postgres_same_schema --dbname=cylon --suffix=weekly
-

Example 6: Run a historical comparison, then replace the file

-
-  check_postgres_same_schema --dbname=cylon --suffix=daily --replace
-

-

-

sequence

-

(symlink: check_postgres_sequence) Checks how much room is left on all sequences in the database. -This is measured as the percent of total possible values that have been used for each sequence. -The --warning and --critical options should be expressed as percentages. The default values -are 85% for the warning and 95% for the critical. You may use --include and --exclude to -control which sequences are to be checked. Note that this check does account for unusual minvalue -and increment by values, but does not care if the sequence is set to cycle or not.

-

The output for Nagios gives the name of the sequence, the percentage used, and the number of 'calls' -left, indicating how many more times nextval can be called on that sequence before running into -the maximum value.

-

The output for MRTG returns the highest percentage across all sequences on the first line, and -the name of each sequence with that percentage on the fourth line, separated by a "|" (pipe) -if there are more than one sequence at that percentage.

-

Example 1: Give a warning if any sequences are approaching 95% full.

-
-  check_postgres_sequence --dbport=5432 --warning=95%
-

Example 2: Check that the sequence named "orders_id_seq" is not more than half full.

-
-  check_postgres_sequence --dbport=5432 --critical=50% --include=orders_id_seq
-

-

-

settings_checksum

-

(symlink: check_postgres_settings_checksum) Checks that all the Postgres settings are the same as last time you checked. -This is done by generating a checksum of a sorted list of setting names and -their values. Note that different users in the same database may have different -checksums, due to ALTER USER usage, and due to the fact that superusers see more -settings than ordinary users. Either the --warning or the --critical option -should be given, but not both. The value of each one is the checksum, a -32-character hexadecimal value. You can run with the special --critical=0 option -to find out an existing checksum.

This action requires the Digest::MD5 module.

-

Example 1: Find the initial checksum for the database on port 5555 using the default user (usually postgres)

-
-  check_postgres_settings_checksum --port=5555 --critical=0
+ +

Example 1: Find the initial checksum for pgbouncer configuration on port 6432 using the default user (usually postgres)

+ +
check_postgres_pgbouncer_checksum --port=6432 --critical=0
+

Example 2: Make sure no settings have changed and warn if so, using the checksum from above.

-
-  check_postgres_settings_checksum --port=5555 --warning=cd2f3b5e129dc2b4f5c0f6d8d2e64231
-

For MRTG output, returns a 1 or 0 indicating success of failure of the checksum to match. A -checksum must be provided as the --mrtg argument. The fourth line always gives the -current checksum.

-

-

-

slony_status

-

(symlink: check_postgres_slony_status) Checks in the status of a Slony cluster by looking -at the results of Slony's sl_status view. This is returned as the number of seconds of "lag time". -The --warning and --critical options should be expressed as times. The default values -are 60 seconds for the warning and 300 seconds for the critical.

-

The optional argument --schema indicated the schema that Slony is installed under. If it is -not given, the schema will be determined automatically each time this check is run.

-

Example 1: Give a warning if any Slony is lagged by more than 20 seconds

-
-  check_postgres_slony_status --warning 20
-

Example 2: Give a critical if Slony, installed under the schema "_slony", is over 10 minutes lagged

-
-  check_postgres_slony_status --schema=_slony --critical=600
-

-

-

timesync

-

(symlink: check_postgres_timesync) Compares the local system time with the time reported by one or more databases. -The --warning and --critical options represent the number of seconds between -the two systems before an alert is given. If neither is specified, the default values -are used, which are '2' and '5'. The warning value cannot be greater than the critical -value. Due to the non-exact nature of this test, values of '0' or '1' are not recommended.

-

The string returned shows the time difference as well as the time on each side written out.

-

Example 1: Check that databases on hosts ankh, morpork, and klatch are no more than 3 seconds off from the local time:

-
-  check_postgres_timesync --host=ankh,morpork,klatch --critical=3
-

For MRTG output, returns one the first line the number of seconds difference between the local -time and the database time. The fourth line returns the name of the database.

-

-

-

txn_idle

-

(symlink: check_postgres_txn_idle) Checks the number and duration of "idle -in transaction" queries on one or more databases. There is no need to run this -more than once on the same database cluster. Databases can be filtered by -using the --include and --exclude options. See the BASIC FILTERING -section below for more details.

-

The --warning and --critical options are given as units of time, signed -integers, or integers for units of time, and both must be provided (there are -no defaults). Valid units are 'seconds', 'minutes', 'hours', or 'days'. Each -may be written singular or abbreviated to just the first letter. If no units -are given and the numbers are unsigned, the units are assumed to be seconds.

-

This action requires Postgres 8.3 or better.

-

Example 1: Give a warning if any connection has been idle in transaction for more than 15 seconds:

-
-  check_postgres_txn_idle --port=5432 --warning='15 seconds'
-

Example 2: Give a warning if there are 50 or more transactions

-
-  check_postgres_txn_idle --port=5432 --warning='+50'
-

Example 3: Give a critical if 5 or more connections have been idle in -transaction for more than 10 seconds:

-
-  check_postgres_txn_idle --port=5432 --critical='5 for 10 seconds'
-

For MRTG output, returns the time in seconds the longest idle transaction has been running. The fourth -line returns the name of the database and other information about the longest transaction.

-

-

-

txn_time

-

(symlink: check_postgres_txn_time) Checks the length of open transactions on one or more databases. -There is no need to run this command more than once per database cluster. -Databases can be filtered by use of the ---include and --exclude options. See the BASIC FILTERING section -for more details. The owner of the transaction can also be filtered, by use of -the --includeuser and --excludeuser options. -See the USER NAME FILTERING section for more details.

-

The values or the --warning and --critical options are units of time, and -must be provided (no default). Valid units are 'seconds', 'minutes', 'hours', -or 'days'. Each may be written singular or abbreviated to just the first letter. -If no units are given, the units are assumed to be seconds.

+ +
check_postgres_pgbouncer_checksum --port=6432 --warning=cd2f3b5e129dc2b4f5c0f6d8d2e64231
+ +

For MRTG output, returns a 1 or 0 indicating success of failure of the checksum to match. A checksum must be provided as the --mrtg argument. The fourth line always gives the current checksum.

+ +

pgbouncer_maxwait

+ +

(symlink: check_postgres_pgbouncer_maxwait) Checks how long the first (oldest) client in the queue has been waiting, in seconds. If this starts increasing, then the current pool of servers does not handle requests quick enough. Reason may be either overloaded server or just too small of a pool_size setting in pbouncer config file. Databases can be filtered by use of the --include and --exclude options. See the "BASIC FILTERING" section for more details. The values or the --warning and --critical options are units of time, and must be provided (no default). Valid units are 'seconds', 'minutes', 'hours', or 'days'. Each may be written singular or abbreviated to just the first letter. If no units are given, the units are assumed to be seconds.

+

This action requires Postgres 8.3 or better.

+

Example 1: Give a critical if any transaction has been open for more than 10 minutes:

-
-  check_postgres_txn_time --port=5432 --critical='10 minutes'
-

Example 1: Warn if user 'warehouse' has a transaction open over 30 seconds

-
-  check_postgres_txn_time --port-5432 --warning=30s --includeuser=warehouse
-

For MRTG output, returns the maximum time in seconds a transaction has been open on the -first line. The fourth line gives the name of the database.

-

-

-

txn_wraparound

-

(symlink: check_postgres_txn_wraparound) Checks how close to transaction wraparound one or more databases are getting. -The --warning and --critical options indicate the number of transactions done, and must be a positive integer. -If either option is not given, the default values of 1.3 and 1.4 billion are used. There is no need to run this command -more than once per database cluster. For a more detailed discussion of what this number represents and what to do about -it, please visit the page -http://www.postgresql.org/docs/current/static/routine-vacuuming.html#VACUUM-FOR-WRAPAROUND

-

The warning and critical values can have underscores in the number for legibility, as Perl does.

-

Example 1: Check the default values for the localhost database

-
-  check_postgres_txn_wraparound --host=localhost
-

Example 2: Check port 6000 and give a critical when 1.7 billion transactions are hit:

-
-  check_postgres_txn_wraparound --port=6000 --critical=1_700_000_000
-

For MRTG output, returns the highest number of transactions for all databases on line one, -while line 4 indicates which database it is.

-

-

-

version

-

(symlink: check_postgres_version) Checks that the required version of Postgres is running. The ---warning and --critical options (only one is required) must be of -the format X.Y or X.Y.Z where X is the major version number, -Y is the minor version number, and Z is the revision.

-

Example 1: Give a warning if the database on port 5678 is not version 8.4.10:

-
-  check_postgres_version --port=5678 -w=8.4.10
-

Example 2: Give a warning if any databases on hosts valley,grain, or sunshine is not 8.3:

-
-  check_postgres_version -H valley,grain,sunshine --critical=8.3
-

For MRTG output, reports a 1 or a 0 indicating success or failure on the first line. The -fourth line indicates the current version. The version must be provided via the --mrtg option.

-

-

-

wal_files

-

(symlink: check_postgres_wal_files) Checks how many WAL files exist in the pg_xlog directory, which is found -off of your data_directory, sometimes as a symlink to another physical disk for -performance reasons. This action must be run as a superuser, in order to access the -contents of the pg_xlog directory. The minimum version to use this action is -Postgres 8.1. The --warning and --critical options are simply the number of -files in the pg_xlog directory. What number to set this to will vary, but a general -guideline is to put a number slightly higher than what is normally there, to catch -problems early.

-

Normally, WAL files are closed and then re-used, but a long-running open -transaction, or a faulty archive_command script, may cause Postgres to -create too many files. Ultimately, this will cause the disk they are on to run -out of space, at which point Postgres will shut down.

-

Example 1: Check that the number of WAL files is 20 or less on host "pluto"

-
-  check_postgres_wal_files --host=pluto --critical=20
-

For MRTG output, reports the number of WAL files on line 1.

-

-

-

rebuild_symlinks

-

-

-

rebuild_symlinks_force

-

This action requires no other arguments, and does not connect to any databases, -but simply creates symlinks in the current directory for each action, in the form -check_postgres_<action_name>. -If the file already exists, it will not be overwritten. If the action is rebuild_symlinks_force, -then symlinks will be overwritten. The option --symlinks is a shorter way of saying ---action=rebuild_symlinks

-

-

-
-

BASIC FILTERING

-

The options --include and --exclude can be combined to limit which -things are checked, depending on the action. The name of the database can -be filtered when using the following actions: -backends, database_size, locks, query_time, txn_idle, and txn_time. -The name of a relation can be filtered when using the following actions: -bloat, index_size, table_size, relation_size, last_vacuum, last_autovacuum, -last_analyze, and last_autoanalyze. -The name of a setting can be filtered when using the settings_checksum action. -The name of a file system can be filtered when using the disk_space action.

-

If only an include option is given, then ONLY those entries that match will be -checked. However, if given both exclude and include, the exclusion is done first, -and the inclusion after, to reinstate things that may have been excluded. Both ---include and --exclude can be given multiple times, -and/or as comma-separated lists. A leading tilde will match the following word -as a regular expression.

-

To match a schema, end the search term with a single period. Leading tildes can -be used for schemas as well.

-

Be careful when using filtering: an inclusion rule on the backends, for example, -may report no problems not only because the matching database had no backends, -but because you misspelled the name of the database!

-

Examples:

-

Only checks items named pg_class:

-
- --include=pg_class
-

Only checks items containing the letters 'pg_':

-
- --include=~pg_
-

Only check items beginning with 'pg_':

-
- --include=~^pg_
-

Exclude the item named 'test':

-
- --exclude=test
-

Exclude all items containing the letters 'test:

-
- --exclude=~test
-

Exclude all items in the schema 'pg_catalog':

-
- --exclude='pg_catalog.'
-

Exclude all items containing the letters 'ace', but allow the item 'faceoff':

-
- --exclude=~ace --include=faceoff
-

Exclude all items which start with the letters 'pg_', which contain the letters 'slon', -or which are named 'sql_settings' or 'green'. Specifically check items with the letters 'prod' in their names, and always check the item named 'pg_relname':

-
- --exclude=~^pg_,~slon,sql_settings --exclude=green --include=~prod,pg_relname
-

-

-
-

USER NAME FILTERING

-

The options --includeuser and --excludeuser can be used on some actions -to only examine database objects owned by (or not owned by) one or more users. -An --includeuser option always trumps an --excludeuser option. You can -give each option more than once for multiple users, or you can give a -comma-separated list. The actions that currently use these options are:

-
-
database_size
-
last_analyze
+
check_postgres_pgbouncer_maxwait -p 6432 -u pgbouncer --critical='10 minutes'
-
last_autoanalyze
+

For MRTG output, returns the maximum time in seconds a transaction has been open on the first line. The fourth line gives the name of the database.

-
last_vacuum
+

pgagent_jobs

-
last_autovacuum
+

(symlink: check_postgres_pgagent_jobs) Checks that all the pgAgent jobs that have executed in the preceding interval of time have succeeded. This is done by checking for any steps that have a non-zero result.

-
query_time
+

Either --warning or --critical, or both, may be specified as times, and jobs will be checked for failures withing the specified periods of time before the current time. Valid units are seconds, minutes, hours, and days; all can be abbreviated to the first letter. If no units are given, 'seconds' are assumed.

-
relation_size
+

Example 1: Give a critical when any jobs executed in the last day have failed.

-
txn_time
+
check_postgres_pgagent_jobs --critical=1d
-
-

Examples:

-

Only check items owned by the user named greg:

-
- --includeuser=greg
-

Only check items owned by either watson or crick:

-
- --includeuser=watson,crick
-

Only check items owned by crick,franklin, watson, or wilkins:

-
- --includeuser=watson --includeuser=franklin --includeuser=crick,wilkins
-

Check all items except for those belonging to the user scott:

-
- --excludeuser=scott
-

-

-
-

TEST MODE

-

To help in setting things up, this program can be run in a "test mode" by -specifying the --test option. This will perform some basic tests to -make sure that the databases can be contacted, and that certain per-action -prerequisites are met, such as whether the user is a superuser, if the version -of Postgres is new enough, and if stats_row_level is enabled.

-

-

-
-

FILES

-

In addition to command-line configurations, you can put any options inside of a file. The file -.check_postgresrc in the current directory will be used if found. If not found, then the file -~/.check_postgresrc will be used. Finally, the file /etc/check_postgresrc will be used if available. -The format of the file is option = value, one per line. Any line starting with a '#' will be skipped. -Any values loaded from a check_postgresrc file will be overwritten by command-line options. All -check_postgresrc files can be ignored by supplying a --no-checkpostgresrc argument.

-

-

-
-

ENVIRONMENT VARIABLES

-

The environment variable $ENV{HOME} is used to look for a .check_postgresrc file. -The environment variable $ENV{PGBINDIR} is used to look for PostgreSQL binaries.

-

-

-
-

TIPS AND TRICKS

-

Since this program uses the psql program, make sure it is accessible to the -user running the script. If run as a cronjob, this often means modifying the -PATH environment variable.

-

If you are using Nagios in embedded Perl mode, use the --action argument -instead of symlinks, so that the plugin only gets compiled one time.

-

-

-
-

DEPENDENCIES

-

Access to a working version of psql, and the following very standard Perl modules:

-
-
Cwd
+

Example 2: Give a warning when any jobs executed in the last week have failed.

+ +
check_postgres_pgagent_jobs --warning=7d
+ +

Example 3: Give a critical for jobs that have failed in the last 2 hours and a warning for jobs that have failed in the last 4 hours:

+ +
check_postgres_pgagent_jobs --critical=2h --warning=4h
+ +

prepared_txns

-
Getopt::Long
+

(symlink: check_postgres_prepared_txns) Check on the age of any existing prepared transactions. Note that most people will NOT use prepared transactions, as they are part of two-part commit and complicated to maintain. They should also not be confused with prepared STATEMENTS, which is what most people think of when they hear prepare. The default value for a warning is 1 second, to detect any use of prepared transactions, which is probably a mistake on most systems. Warning and critical are the number of seconds a prepared transaction has been open before an alert is given.

+ +

Example 1: Give a warning on detecting any prepared transactions:

+ +
check_postgres_prepared_txns -w 0
+ +

Example 2: Give a critical if any prepared transaction has been open longer than 10 seconds, but allow up to 360 seconds for the database 'shrike':

+ +
check_postgres_prepared_txns --critical=10 --exclude=shrike
+check_postgres_prepared_txns --critical=360 --include=shrike
+ +

For MRTG output, returns the number of seconds the oldest transaction has been open as the first line, and which database is came from as the final line.

+ +

query_runtime

+ +

(symlink: check_postgres_query_runtime) Checks how long a specific query takes to run, by executing a "EXPLAIN ANALYZE" against it. The --warning and --critical options are the maximum amount of time the query should take. Valid units are seconds, minutes, and hours; any can be abbreviated to the first letter. If no units are given, 'seconds' are assumed. Both the warning and the critical option must be given. The name of the view or function to be run must be passed in to the --queryname option. It must consist of a single word (or schema.word), with optional parens at the end.

+ +

Example 1: Give a critical if the function named "speedtest" fails to run in 10 seconds or less.

+ +
check_postgres_query_runtime --queryname='speedtest()' --critical=10 --warning=10
+ +

For MRTG output, reports the time in seconds for the query to complete on the first line. The fourth line lists the database.

+ +

query_time

+ +

(symlink: check_postgres_query_time) Checks the length of running queries on one or more databases. There is no need to run this more than once on the same database cluster. Note that this already excludes queries that are "idle in transaction". Databases can be filtered by using the --include and --exclude options. See the "BASIC FILTERING" section for more details. You can also filter on the user running the query with the --includeuser and --excludeuser options. See the "USER NAME FILTERING" section for more details.

+ +

The values for the --warning and --critical options are amounts of time, and at least one must be provided (no defaults). Valid units are 'seconds', 'minutes', 'hours', or 'days'. Each may be written singular or abbreviated to just the first letter. If no units are given, the unit is assumed to be seconds.

+ +

This action requires Postgres 8.1 or better.

+ +

Example 1: Give a warning if any query has been running longer than 3 minutes, and a critical if longer than 5 minutes.

+ +
check_postgres_query_time --port=5432 --warning='3 minutes' --critical='5 minutes'
+ +

Example 2: Using default values (2 and 5 minutes), check all databases except those starting with 'template'.

+ +
check_postgres_query_time --port=5432 --exclude=~^template
+ +

Example 3: Warn if user 'don' has a query running over 20 seconds

+ +
check_postgres_query_time --port=5432 --includeuser=don --warning=20s
+ +

For MRTG output, returns the length in seconds of the longest running query on the first line. The fourth line gives the name of the database.

+ +

replicate_row

+ +

(symlink: check_postgres_replicate_row) Checks that master-slave replication is working to one or more slaves.

+ +

The first "--dbname", "--host", and "--port", etc. options are considered the master; subsequent uses are the slaves. The values or the --warning and --critical options are units of time, and at least one must be provided (no defaults). Valid units are 'seconds', 'minutes', 'hours', or 'days'. Each may be written singular or abbreviated to just the first letter. If no units are given, the units are assumed to be seconds.

+ +

This check updates a single row on the master, and then measures how long it takes to be applied to the slaves. To do this, you need to pick a table that is being replicated, then find a row that can be changed, and is not going to be changed by any other process. A specific column of this row will be changed from one value to another. All of this is fed to the repinfo option, and should contain the following options, separated by commas: table name, primary key, key id, column, first value, second value.

+ +

Example 1: Slony is replicating a table named 'orders' from host 'alpha' to host 'beta', in the database 'sales'. The primary key of the table is named id, and we are going to test the row with an id of 3 (which is historical and never changed). There is a column named 'salesrep' that we are going to toggle from a value of 'slon' to 'nols' to check on the replication. We want to throw a warning if the replication does not happen within 10 seconds.

+ +
check_postgres_replicate_row --host=alpha --dbname=sales --host=beta
+--dbname=sales --warning=10 --repinfo=orders,id,3,salesrep,slon,nols
+ +

Example 2: Bucardo is replicating a table named 'receipt' from host 'green' to hosts 'red', 'blue', and 'yellow'. The database for both sides is 'public'. The slave databases are running on port 5455. The primary key is named 'receipt_id', the row we want to use has a value of 9, and the column we want to change for the test is called 'zone'. We'll toggle between 'north' and 'south' for the value of this column, and throw a critical if the change is not on all three slaves within 5 seconds.

+ +
check_postgres_replicate_row --host=green --port=5455 --host=red,blue,yellow
+ --critical=5 --repinfo=receipt,receipt_id,9,zone,north,south
+ +

For MRTG output, returns on the first line the time in seconds the replication takes to finish. The maximum time is set to 4 minutes 30 seconds: if no replication has taken place in that long a time, an error is thrown.

+ +

replication_slots

+ +

(symlink: check_postgres_replication_slots) Check the quantity of WAL retained for any replication slots in the target database cluster. This is handy for monitoring environments where all WAL archiving and replication is taking place over replication slots.

+ +

Warning and critical are total bytes retained for the slot. E.g:

+ +
check_postgres_replication_slots --port=5432 --host=yellow -warning=32M -critical=64M
+ +

Specific named slots can be monitored using --include/--exclude

+ +

same_schema

+ +

(symlink: check_postgres_same_schema) Verifies that two or more databases are identical as far as their schema (but not the data within). Unlike most other actions, this has no warning or critical criteria - the databases are either in sync, or are not. If they are different, a detailed list of the differences is presented.

+ +

You may want to exclude or filter out certain differences. The way to do this is to add strings to the --filter option. To exclude a type of object, use "noname", where 'name' is the type of object, for example, "noschema". To exclude objects of a certain type by a regular expression against their name, use "noname=regex". See the examples below for a better understanding.

+ +

The types of objects that can be filtered include:

+ +
+ +
user
+
+ +
+
schema
+
+ +
+
table
+
+ +
+
view
+
+ +
+
index
+
+ +
+
sequence
+
+ +
+
constraint
+
+ +
+
trigger
+
+ +
+
function
+
+ +
+
+ +

The filter option "noposition" prevents verification of the position of columns within a table.

+ +

The filter option "nofuncbody" prevents comparison of the bodies of all functions.

+ +

The filter option "noperm" prevents comparison of object permissions.

+ +

To provide the second database, just append the differences to the first one by a call to the appropriate connection argument. For example, to compare databases on hosts alpha and bravo, use "--dbhost=alpha,bravo". Also see the examples below.

+ +

If only a single host is given, it is assumed we are doing a "time-based" report. The first time this is run a snapshot of all the items in the database is saved to a local file. When you run it again, that snapshot is read in and becomes "database #2" and is compared to the current database.

+ +

To replace the old stored file with the new version, use the --replace argument.

+ +

If you need to write the stored file to a specific directory, use the --audit-file-dir argument.

+ +

To avoid false positives on value based checks caused by replication lag on asynchronous replicas, use the --assume-async option.

+ +

To enable snapshots at various points in time, you can use the "--suffix" argument to make the filenames unique to each run. See the examples below.

+ +

Example 1: Verify that two databases on hosts star and line are the same:

+ +
check_postgres_same_schema --dbhost=star,line
+ +

Example 2: Same as before, but exclude any triggers with "slony" in their name

+ +
check_postgres_same_schema --dbhost=star,line --filter="notrigger=slony"
+ +

Example 3: Same as before, but also exclude all indexes

+ +
check_postgres_same_schema --dbhost=star,line --filter="notrigger=slony noindexes"
+ +

Example 4: Check differences for the database "battlestar" on different ports

+ +
check_postgres_same_schema --dbname=battlestar --dbport=5432,5544
+ +

Example 5: Create a daily and weekly snapshot file

+ +
check_postgres_same_schema --dbname=cylon --suffix=daily
+check_postgres_same_schema --dbname=cylon --suffix=weekly
+ +

Example 6: Run a historical comparison, then replace the file

+ +
check_postgres_same_schema --dbname=cylon --suffix=daily --replace
+ +

Example 7: Verify that two databases on hosts star and line are the same, excluding value data (i.e. sequence last_val):

+ +
check_postgres_same_schema --dbhost=star,line --assume-async
+ +

sequence

+ +

(symlink: check_postgres_sequence) Checks how much room is left on all sequences in the database. This is measured as the percent of total possible values that have been used for each sequence. The --warning and --critical options should be expressed as percentages. The default values are 85% for the warning and 95% for the critical. You may use --include and --exclude to control which sequences are to be checked. Note that this check does account for unusual minvalue and increment by values. By default it does not care if the sequence is set to cycle or not, and by passing --skipcycled sequenced set to cycle are reported with 0% usage.

+ +

The output for Nagios gives the name of the sequence, the percentage used, and the number of 'calls' left, indicating how many more times nextval can be called on that sequence before running into the maximum value.

+ +

The output for MRTG returns the highest percentage across all sequences on the first line, and the name of each sequence with that percentage on the fourth line, separated by a "|" (pipe) if there are more than one sequence at that percentage.

+ +

Example 1: Give a warning if any sequences are approaching 95% full.

+ +
check_postgres_sequence --dbport=5432 --warning=95%
+ +

Example 2: Check that the sequence named "orders_id_seq" is not more than half full.

+ +
check_postgres_sequence --dbport=5432 --critical=50% --include=orders_id_seq
+ +

settings_checksum

+ +

(symlink: check_postgres_settings_checksum) Checks that all the Postgres settings are the same as last time you checked. This is done by generating a checksum of a sorted list of setting names and their values. Note that different users in the same database may have different checksums, due to ALTER USER usage, and due to the fact that superusers see more settings than ordinary users. Either the --warning or the --critical option should be given, but not both. The value of each one is the checksum, a 32-character hexadecimal value. You can run with the special --critical=0 option to find out an existing checksum.

+ +

This action requires the Digest::MD5 module.

+ +

Example 1: Find the initial checksum for the database on port 5555 using the default user (usually postgres)

+ +
check_postgres_settings_checksum --port=5555 --critical=0
+ +

Example 2: Make sure no settings have changed and warn if so, using the checksum from above.

+ +
check_postgres_settings_checksum --port=5555 --warning=cd2f3b5e129dc2b4f5c0f6d8d2e64231
+ +

For MRTG output, returns a 1 or 0 indicating success of failure of the checksum to match. A checksum must be provided as the --mrtg argument. The fourth line always gives the current checksum.

+ +

slony_status

+ +

(symlink: check_postgres_slony_status) Checks in the status of a Slony cluster by looking at the results of Slony's sl_status view. This is returned as the number of seconds of "lag time". The --warning and --critical options should be expressed as times. The default values are 60 seconds for the warning and 300 seconds for the critical.

+ +

The optional argument --schema indicated the schema that Slony is installed under. If it is not given, the schema will be determined automatically each time this check is run.

+ +

Example 1: Give a warning if any Slony is lagged by more than 20 seconds

+ +
check_postgres_slony_status --warning 20
+ +

Example 2: Give a critical if Slony, installed under the schema "_slony", is over 10 minutes lagged

+ +
check_postgres_slony_status --schema=_slony --critical=600
+ +

timesync

+ +

(symlink: check_postgres_timesync) Compares the local system time with the time reported by one or more databases. The --warning and --critical options represent the number of seconds between the two systems before an alert is given. If neither is specified, the default values are used, which are '2' and '5'. The warning value cannot be greater than the critical value. Due to the non-exact nature of this test, values of '0' or '1' are not recommended.

+ +

The string returned shows the time difference as well as the time on each side written out.

+ +

Example 1: Check that databases on hosts ankh, morpork, and klatch are no more than 3 seconds off from the local time:

+ +
check_postgres_timesync --host=ankh,morpork,klatch --critical=3
+ +

For MRTG output, returns one the first line the number of seconds difference between the local time and the database time. The fourth line returns the name of the database.

+ +

txn_idle

+ +

(symlink: check_postgres_txn_idle) Checks the number and duration of "idle in transaction" queries on one or more databases. There is no need to run this more than once on the same database cluster. Databases can be filtered by using the --include and --exclude options. See the "BASIC FILTERING" section below for more details.

+ +

The --warning and --critical options are given as units of time, signed integers, or integers for units of time, and at least one must be provided (there are no defaults). Valid units are 'seconds', 'minutes', 'hours', or 'days'. Each may be written singular or abbreviated to just the first letter. If no units are given and the numbers are unsigned, the units are assumed to be seconds.

+ +

This action requires Postgres 8.3 or better.

+ +

As of PostgreSQL 10, you can just GRANT pg_read_all_stats to an unprivileged user account. In all earlier versions, superuser privileges are required to see the queries of all users in the system; UNKNOWN is returned if queries cannot be checked. To only include queries by the connecting user, use --includeuser.

+ +

Example 1: Give a warning if any connection has been idle in transaction for more than 15 seconds:

+ +
check_postgres_txn_idle --port=5432 --warning='15 seconds'
+ +

Example 2: Give a warning if there are 50 or more transactions

+ +
check_postgres_txn_idle --port=5432 --warning='+50'
+ +

Example 3: Give a critical if 5 or more connections have been idle in transaction for more than 10 seconds:

+ +
check_postgres_txn_idle --port=5432 --critical='5 for 10 seconds'
+ +

For MRTG output, returns the time in seconds the longest idle transaction has been running. The fourth line returns the name of the database and other information about the longest transaction.

+ +

txn_time

+ +

(symlink: check_postgres_txn_time) Checks the length of open transactions on one or more databases. There is no need to run this command more than once per database cluster. Databases can be filtered by use of the --include and --exclude options. See the "BASIC FILTERING" section for more details. The owner of the transaction can also be filtered, by use of the --includeuser and --excludeuser options. See the "USER NAME FILTERING" section for more details.

+ +

The values or the --warning and --critical options are units of time, and at least one must be provided (no default). Valid units are 'seconds', 'minutes', 'hours', or 'days'. Each may be written singular or abbreviated to just the first letter. If no units are given, the units are assumed to be seconds.

+ +

This action requires Postgres 8.3 or better.

+ +

Example 1: Give a critical if any transaction has been open for more than 10 minutes:

+ +
check_postgres_txn_time --port=5432 --critical='10 minutes'
+ +

Example 1: Warn if user 'warehouse' has a transaction open over 30 seconds

+ +
check_postgres_txn_time --port-5432 --warning=30s --includeuser=warehouse
+ +

For MRTG output, returns the maximum time in seconds a transaction has been open on the first line. The fourth line gives the name of the database.

+ +

txn_wraparound

+ +

(symlink: check_postgres_txn_wraparound) Checks how close to transaction wraparound one or more databases are getting. The --warning and --critical options indicate the number of transactions done, and must be a positive integer. If either option is not given, the default values of 1.3 and 1.4 billion are used. There is no need to run this command more than once per database cluster. For a more detailed discussion of what this number represents and what to do about it, please visit the page https://www.postgresql.org/docs/current/static/routine-vacuuming.html#VACUUM-FOR-WRAPAROUND

+ +

The warning and critical values can have underscores in the number for legibility, as Perl does.

+ +

Example 1: Check the default values for the localhost database

+ +
check_postgres_txn_wraparound --host=localhost
+ +

Example 2: Check port 6000 and give a critical when 1.7 billion transactions are hit:

+ +
check_postgres_txn_wraparound --port=6000 --critical=1_700_000_000
+ +

For MRTG output, returns the highest number of transactions for all databases on line one, while line 4 indicates which database it is.

+ +

version

+ +

(symlink: check_postgres_version) Checks that the required version of Postgres is running. The --warning and --critical options (only one is required) must be of the format X.Y or X.Y.Z where X is the major version number, Y is the minor version number, and Z is the revision.

+ +

Example 1: Give a warning if the database on port 5678 is not version 8.4.10:

+ +
check_postgres_version --port=5678 -w=8.4.10
+ +

Example 2: Give a warning if any databases on hosts valley,grain, or sunshine is not 8.3:

+ +
check_postgres_version -H valley,grain,sunshine --critical=8.3
+ +

For MRTG output, reports a 1 or a 0 indicating success or failure on the first line. The fourth line indicates the current version. The version must be provided via the --mrtg option.

+ +

wal_files

+ +

(symlink: check_postgres_wal_files) Checks how many WAL files exist in the pg_xlog directory (PostgreSQL 10 and later" pg_wal), which is found off of your data_directory, sometimes as a symlink to another physical disk for performance reasons. If the --lsfunc option is not used then this action must be run as a superuser, in order to access the contents of the pg_xlog directory. The minimum version to use this action is Postgres 8.1. The --warning and --critical options are simply the number of files in the pg_xlog directory. What number to set this to will vary, but a general guideline is to put a number slightly higher than what is normally there, to catch problems early.

+ +

Normally, WAL files are closed and then re-used, but a long-running open transaction, or a faulty archive_command script, may cause Postgres to create too many files. Ultimately, this will cause the disk they are on to run out of space, at which point Postgres will shut down.

+ +

To avoid connecting as a database superuser, a wrapper function around pg_ls_dir() should be defined as a superuser with SECURITY DEFINER, and the --lsfunc option used. This example function, if defined by a superuser, will allow the script to connect as a normal user nagios with --lsfunc=ls_xlog_dir

+ +
BEGIN;
+CREATE FUNCTION ls_xlog_dir()
+    RETURNS SETOF TEXT
+    AS $$ SELECT pg_ls_dir('pg_xlog') $$
+    LANGUAGE SQL
+    SECURITY DEFINER;
+REVOKE ALL ON FUNCTION ls_xlog_dir() FROM PUBLIC;
+GRANT EXECUTE ON FUNCTION ls_xlog_dir() to nagios;
+COMMIT;
+ +

Example 1: Check that the number of ready WAL files is 10 or less on host "pluto", using a wrapper function ls_xlog_dir to avoid the need for superuser permissions

+ +
check_postgres_archive_ready --host=pluto --critical=10 --lsfunc=ls_xlog_dir
+ +

For MRTG output, reports the number of WAL files on line 1.

+ + + + + +

This action requires no other arguments, and does not connect to any databases, but simply creates symlinks in the current directory for each action, in the form check_postgres_<action_name>. If the file already exists, it will not be overwritten. If the action is rebuild_symlinks_force, then symlinks will be overwritten. The option --symlinks is a shorter way of saying --action=rebuild_symlinks

+ +

BASIC FILTERING

+ +

The options --include and --exclude can be combined to limit which things are checked, depending on the action. The name of the database can be filtered when using the following actions: backends, database_size, locks, query_time, txn_idle, and txn_time. The name of a relation can be filtered when using the following actions: bloat, index_size, table_size, relation_size, last_vacuum, last_autovacuum, last_analyze, and last_autoanalyze. The name of a setting can be filtered when using the settings_checksum action. The name of a file system can be filtered when using the disk_space action.

+ +

If only an include option is given, then ONLY those entries that match will be checked. However, if given both exclude and include, the exclusion is done first, and the inclusion after, to reinstate things that may have been excluded. Both --include and --exclude can be given multiple times, and/or as comma-separated lists. A leading tilde will match the following word as a regular expression.

+ +

To match a schema, end the search term with a single period. Leading tildes can be used for schemas as well.

+ +

Be careful when using filtering: an inclusion rule on the backends, for example, may report no problems not only because the matching database had no backends, but because you misspelled the name of the database!

+ +

Examples:

+ +

Only checks items named pg_class:

+ +
--include=pg_class
+ +

Only checks items containing the letters 'pg_':

+ +
--include=~pg_
+ +

Only check items beginning with 'pg_':

+ +
--include=~^pg_
+ +

Exclude the item named 'test':

+ +
--exclude=test
+ +

Exclude all items containing the letters 'test:

+ +
--exclude=~test
+ +

Exclude all items in the schema 'pg_catalog':

+ +
--exclude='pg_catalog.'
+ +

Exclude all items in the 'pg_temp_nnn' per-session temporary schemas:

+ +
--exclude=~^pg_temp_.
+ +

Exclude all items containing the letters 'ace', but allow the item 'faceoff':

+ +
--exclude=~ace --include=faceoff
+ +

Exclude all items which start with the letters 'pg_', which contain the letters 'slon', or which are named 'sql_settings' or 'green'. Specifically check items with the letters 'prod' in their names, and always check the item named 'pg_relname':

+ +
--exclude=~^pg_,~slon,sql_settings --exclude=green --include=~prod,pg_relname
+ +

USER NAME FILTERING

+ +

The options --includeuser and --excludeuser can be used on some actions to only examine database objects owned by (or not owned by) one or more users. An --includeuser option always trumps an --excludeuser option. You can give each option more than once for multiple users, or you can give a comma-separated list. The actions that currently use these options are:

+ +
+ +
database_size
+
+ +
+
last_analyze
+
+ +
+
last_autoanalyze
+
+ +
+
last_vacuum
+
+ +
+
last_autovacuum
+
+ +
+
query_time
+
+ +
+
relation_size
+
+ +
+
txn_time
+
+ +
+
+ +

Examples:

+ +

Only check items owned by the user named greg:

+ +
--includeuser=greg
+ +

Only check items owned by either watson or crick:

+ +
--includeuser=watson,crick
+ +

Only check items owned by crick,franklin, watson, or wilkins:

+ +
--includeuser=watson --includeuser=franklin --includeuser=crick,wilkins
+ +

Check all items except for those belonging to the user scott:

+ +
--excludeuser=scott
+ +

TEST MODE

+ +

To help in setting things up, this program can be run in a "test mode" by specifying the --test option. This will perform some basic tests to make sure that the databases can be contacted, and that certain per-action prerequisites are met, such as whether the user is a superuser, if the version of Postgres is new enough, and if stats_row_level is enabled.

+ +

FILES

+ +

In addition to command-line configurations, you can put any options inside of a file. The file .check_postgresrc in the current directory will be used if found. If not found, then the file ~/.check_postgresrc will be used. Finally, the file /etc/check_postgresrc will be used if available. The format of the file is option = value, one per line. Any line starting with a '#' will be skipped. Any values loaded from a check_postgresrc file will be overwritten by command-line options. All check_postgresrc files can be ignored by supplying a --no-checkpostgresrc argument.

+ +

ENVIRONMENT VARIABLES

+ +

The environment variable $ENV{HOME} is used to look for a .check_postgresrc file. The environment variable $ENV{PGBINDIR} is used to look for PostgreSQL binaries.

+ +

TIPS AND TRICKS

+ +

Since this program uses the psql program, make sure it is accessible to the user running the script. If run as a cronjob, this often means modifying the PATH environment variable.

+ +

If you are using Nagios in embedded Perl mode, use the --action argument instead of symlinks, so that the plugin only gets compiled one time.

+ +

DEPENDENCIES

+ +

Access to a working version of psql, and the following very standard Perl modules:

+ +
+ +
Cwd
+
+ +
+
Getopt::Long
+
+ +
+
File::Basename
+
+ +
+
File::Temp
+
+ +
+
Time::HiRes (if $opt{showtime} is set to true, which is the default)
+
+ +
+
+ +

The "settings_checksum" action requires the Digest::MD5 module.

+ +

The "checkpoint" action requires the Date::Parse module.

+ +

Some actions require access to external programs. If psql is not explicitly specified, the command which is used to find it. The program /bin/df is needed by the "disk_space" action.

+ +

DEVELOPMENT

+ +

Development happens using the git system. You can clone the latest version by doing:

+ +
https://github.com/bucardo/check_postgres
+git clone https://github.com/bucardo/check_postgres.git
+ +

MAILING LIST

+ +

Three mailing lists are available. For discussions about the program, bug reports, feature requests, and commit notices, send email to check_postgres@bucardo.org

+ +

https://bucardo.org/mailman/listinfo/check_postgres

+ +

A low-volume list for announcement of new versions and important notices is the 'check_postgres-announce' list:

+ +

https://bucardo.org/mailman/listinfo/check_postgres-announce

+ +

HISTORY

+ +

Items not specifically attributed are by GSM (Greg Sabino Mullane).

+ +
+ +
Version 2.26.1 not yet released
+
+ +
Add new action "lockwait" showing details of blocked queries
+(Github user miraclesvenni)
+[Github issue #154]
+
+Raise minimum version or Perl to 5.10.0
+
+Allow commas in passwords via --dbpass for one-connection queries (Greg Sabino Mullane) [Github issue #133]
+
+Fix undefined variable error (Greg Sabino Mullane) [Github issue #141]
+ +
+
Version 2.26.0 Released April 3, 2023
+
+ +
Add new action "pgbouncer_maxwait" (Ruslan Kabalin) [Github pull #59]
+
+For the bloat check, add option to populate all known databases, 
+  as well as includsion and exclusion regexes. (Giles Westwood) [Github pull #86]
+
+Add Partman premake check (Jens Wilke) [Github pull #196]
+
+Add --role flag to explicitly set the role of the user after connecting (David Christensen)
+
+Fix check_replication_slots on recently promoted servers (Christoph Berg)
+
+Allow the check_disk_space action to handle relative log_directory paths (jacksonfoz) [Github pull #174]
+
+Fix MINPAGES and MINIPAGES in the "check_bloat" action (Christoph Moench-Tegeder) [Github pull #82]
+
+Replace 'which' with 'command -v' (Christoph Berg)
+
+Fix check_replication_slots on recently promoted servers (Christoph Berg)
+
+Fix undefined variable warning (Michael van Bracht) [Github pull #158]
+
+In the tests, force log_destination to stderr (Christoph Moench-Tegeder) [Github pull #185]
+
+Add to docs how to exclude all items in the 'pg_temp_nnn' per-session temporary schemas (Michael Banck)
+
+Various fixes for the CI system (Emre Hasegeli) [Github pull #181]
+
+Various improvements to the tests (Christoph Berg, Emre Hasegeli)
+ +
+
Version 2.25.0 Released February 3, 2020
+
+ +
Allow same_schema objects to be included or excluded with --object and --skipobject
+  (Greg Sabino Mullane)
+
+Fix to allow mixing service names and other connection parameters for same_schema
+  (Greg Sabino Mullane)
+ +
+
Version 2.24.0 Released May 30, 2018
+
+ +
Support new_version_pg for PG10
+  (Michael Pirogov)
+
+Option to skip CYCLE sequences in action sequence
+  (Christoph Moench-Tegeder)
+
+Output per-database perfdata for pgbouncer pool checks
+  (George Hansper)
+
+German message translations
+  (Holger Jacobs)
+
+Consider only client backends in query_time and friends
+  (David Christensen)
+ +
+
Version 2.23.0 Released October 31, 2017
+
+ +
Support PostgreSQL 10.
+  (David Christensen, Christoph Berg)
+
+Change table_size to use pg_table_size() on 9.0+, i.e. include the TOAST
+table size in the numbers reported. Add new actions indexes_size and
+total_relation_size, using the respective pg_indexes_size() and
+pg_total_relation_size() functions. All size checks will now also check
+materialized views where applicable.
+  (Christoph Berg)
+
+Connection errors are now always critical, not unknown.
+  (Christoph Berg)
+
+New action replication_slots checking if logical or physical replication
+slots have accumulated too much data
+  (Glyn Astill)
+
+Multiple same_schema improvements
+  (Glyn Astill)
+
+Add Spanish message translations
+  (Luis Vazquez)
+
+Allow a wrapper function to run wal_files and archive_ready actions as
+non-superuser
+  (Joshua Elsasser)
+
+Add some defensive casting to the bloat query
+  (Greg Sabino Mullane)
+
+Invoke psql with option -X
+  (Peter Eisentraut)
+
+Update postgresql.org URLs to use https.
+  (Magnus Hagander)
+
+check_txn_idle: Don't fail when query contains 'disabled' word
+  (Marco Nenciarini)
+
+check_txn_idle: Use state_change instead of query_start.
+  (Sebastian Webber)
+
+check_hot_standby_delay: Correct extra space in perfdata
+  (Adrien Nayrat)
+
+Remove \r from psql output as it can confuse some regexes
+  (Greg Sabino Mullane)
+
+Sort failed jobs in check_pgagent_jobs for stable output.
+  (Christoph Berg)
+ +
+
Version 2.22.0 June 30, 2015
+
+ +
Add xact timestamp support to hot_standby_delay.
+Allow the hot_standby_delay check to accept xlog byte position or
+timestamp lag intervals as thresholds, or even both at the same time.
+  (Josh Williams)
+
+Query all sequences per DB in parallel for action=sequence.
+  (Christoph Berg)
+
+Fix bloat check to use correct SQL depending on the server version.
+  (Adrian Vondendriesch)
+
+Show actual long-running query in query_time output
+  (Peter Eisentraut)
+
+Add explicit ORDER BY to the slony_status check to get the most lagged server.
+  (Jeff Frost)
+
+Improved multi-slave support in replicate_row.
+  (Andrew Yochum)
+
+Change the way tables are quoted in replicate_row.
+  (Glyn Astill)
+
+Don't swallow space before the -c flag when reporting errors
+  (Jeff Janes)
+
+Fix and extend hot_standby_delay documentation
+  (Michael Renner)
+
+Declare POD encoding to be utf8.
+  (Christoph Berg)
+ +
+
Version 2.21.0 September 24, 2013
+
+ +
Fix issue with SQL steps in check_pgagent_jobs for sql steps which perform deletes
+  (Rob Emery via github pull)
+
+Install man page in section 1.
+  (Peter Eisentraut, bug 53, github issue 26)
+
+Order lock types in check_locks output to make the ordering predictable;
+setting SKIP_NETWORK_TESTS will skip the new_version tests; other minor test
+suite fixes.
+  (Christoph Berg)
+
+Fix same_schema check on 9.3 by ignoring relminmxid differences in pg_class
+  (Christoph Berg)
+ +
+
Version 2.20.1 June 24, 2013
+
+ +
Make connection check failures return CRITICAL not UNKNOWN
+  (Dominic Hargreaves)
+
+Fix --reverse option when using string comparisons in custom queries
+  (Nathaniel Waisbrot)
+
+Compute correct 'totalwastedbytes' in the bloat query
+  (Michael Renner)
+
+Do not use pg_stats "inherited" column in bloat query, if the
+  database is 8.4 or older. (Greg Sabino Mullane, per bug 121)
+
+Remove host reordering in hot_standby_delay check
+  (Josh Williams, with help from Jacobo Blasco)
+
+Better output for the "simple" flag
+  (Greg Sabino Mullane)
+
+Force same_schema to ignore the 'relallvisible' column
+  (Greg Sabino Mullane)
+ +
+
Version 2.20.0 March 13, 2013
+
+ +
Add check for pgagent jobs (David E. Wheeler)
+
+Force STDOUT to use utf8 for proper output
+  (Greg Sabino Mullane; reported by Emmanuel Lesouef)
+
+Fixes for Postgres 9.2: new pg_stat_activity view,
+  and use pg_tablespace_location, (Josh Williams)
+
+Allow for spaces in item lists when doing same_schema.
+
+Allow txn_idle to work again for < 8.3 servers by switching to query_time.
+
+Fix the check_bloat SQL to take inherited tables into account,
+  and assume 2k for non-analyzed columns. (Geert Pante)
+
+Cache sequence information to speed up same_schema runs.
+
+Fix --excludeuser in check_txn_idle (Mika Eloranta)
+
+Fix user clause handling in check_txn_idle (Michael van Bracht)
+
+Adjust docs to show colon as a better separator inside args for locks
+  (Charles Sprickman)
+
+Fix undefined $SQL2 error in check_txn_idle [github issue 16] (Patric Bechtel)
+
+Prevent "uninitialized value" warnings when showing the port (Henrik Ahlgren)
+
+Do not assume everyone has a HOME [github issue 23]
+ +
+
Version 2.19.0 January 17, 2012
+
+ +
Add the --assume-prod option (Cédric Villemain)
+
+Add the cluster_id check (Cédric Villemain)
+
+Improve settings_checksum and checkpoint tests (Cédric Villemain)
+
+Do not do an inner join to pg_user when checking database size
+  (Greg Sabino Mullane; reported by Emmanuel Lesouef)
+
+Use the full path when getting sequence information for same_schema.
+  (Greg Sabino Mullane; reported by Cindy Wise)
+
+Fix the formula for calculating xlog positions (Euler Taveira de Oliveira)
+
+Better ordering of output for bloat check - make indexes as important
+  as tables (Greg Sabino Mullane; reported by Jens Wilke)
+
+Show the dbservice if it was used at top of same_schema output
+  (Mike Blackwell)
+
+Better installation paths (Greg Sabino Mullane, per bug 53)
+ +
+
Version 2.18.0 October 2, 2011
+
+ +
Redo the same_schema action. Use new --filter argument for all filtering.
+Allow comparisons between any number of databases.
+Remove the dbname2, dbport2, etc. arguments.
+Allow comparison of the same db over time.
+
+Swap db1 and db2 if the slave is 1 for the hot standby check (David E. Wheeler)
+
+Allow multiple --schema arguments for the slony_status action (GSM and Jehan-Guillaume de Rorthais)
+
+Fix ORDER BY in the last vacuum/analyze action (Nicolas Thauvin)
+
+Fix check_hot_standby_delay perfdata output (Nicolas Thauvin)
+
+Look in the correct place for the .ready files with the archive_ready action (Nicolas Thauvin)
+
+New action: commitratio (Guillaume Lelarge)
+
+New action: hitratio (Guillaume Lelarge)
+
+Make sure --action overrides the symlink naming trick.
+
+Set defaults for archive_ready and wal_files (Thomas Guettler, GSM)
+
+Better output for wal_files and archive_ready (GSM)
+
+Fix warning when client_port set to empty string (bug #79)
+
+Account for "empty row" in -x output (i.e. source of functions).
+
+Fix some incorrectly named data fields (Andy Lester)
+
+Expand the number of pgbouncer actions (Ruslan Kabalin)
+
+Give detailed information and refactor txn_idle, txn_time, and query_time
+  (Per request from bug #61)
+
+Set maxalign to 8 in the bloat check if box identified as '64-bit'
+  (Michel Sijmons, bug #66)
+
+Support non-standard version strings in the bloat check.
+  (Michel Sijmons and Gurjeet Singh, bug #66)
+
+Do not show excluded databases in some output (Ruslan Kabalin)
+
+Allow "and", "or" inside arguments (David E. Wheeler)
+
+Add the "new_version_box" action.
+
+Fix psql version regex (Peter Eisentraut, bug #69)
+
+Add the --assume-standby-mode option (Ruslan Kabalin)
+
+Note that txn_idle and query_time require 8.3 (Thomas Guettler)
+
+Standardize and clean up all perfdata output (bug #52)
+
+Exclude "idle in transaction" from the query_time check (bug #43)
+
+Fix the perflimit for the bloat action (bug #50)
+
+Clean up the custom_query action a bit.
+
+Fix space in perfdata for hot_standby_delay action (Nicolas Thauvin)
+
+Handle undef percents in check_fsm_relations (Andy Lester)
+
+Fix typo in dbstats action (Stas Vitkovsky)
+
+Fix MRTG for last vacuum and last_analyze actions.
+ +
+
Version 2.17.0 no public release
+
+ +
+
Version 2.16.0 January 20, 2011
+
+ +
Add new action 'hot_standby_delay' (Nicolas Thauvin)
+Add cache-busting for the version-grabbing utilities.
+Fix problem with going to next method for new_version_pg
+  (Greg Sabino Mullane, reported by Hywel Mallett in bug #65)
+Allow /usr/local/etc as an alternative location for the 
+  check_postgresrc file (Hywel Mallett)
+Do not use tgisconstraint in same_schema if Postgres >= 9
+  (Guillaume Lelarge)
+ +
+
Version 2.15.4 January 3, 2011
+
+ +
Fix warning when using symlinks
+  (Greg Sabino Mullane, reported by Peter Eisentraut in bug #63)
+ +
+
Version 2.15.3 December 30, 2010
+
+ +
Show OK for no matching txn_idle entries.
+ +
+
Version 2.15.2 December 28, 2010
+
+ +
Better formatting of sizes in the bloat action output.
+
+Remove duplicate perfs in bloat action output.
+ +
+
Version 2.15.1 December 27, 2010
+
+ +
Fix problem when examining items in pg_settings (Greg Sabino Mullane)
+
+For connection test, return critical, not unknown, on FATAL errors
+  (Greg Sabino Mullane, reported by Peter Eisentraut in bug #62)
+ +
+
Version 2.15.0 November 8, 2010
+
+ +
Add --quiet argument to suppress output on OK Nagios results
+Add index comparison for same_schema (Norman Yamada and Greg Sabino Mullane)
+Use $ENV{PGSERVICE} instead of "service=" to prevent problems (Guillaume Lelarge)
+Add --man option to show the entire manual. (Andy Lester)
+Redo the internal run_command() sub to use -x and hashes instead of regexes.
+Fix error in custom logic (Andreas Mager)
+Add the "pgbouncer_checksum" action (Guillaume Lelarge)
+Fix regex to work on WIN32 for check_fsm_relations and check_fsm_pages (Luke Koops)
+Don't apply a LIMIT when using --exclude on the bloat action (Marti Raudsepp)
+Change the output of query_time to show pid,user,port, and address (Giles Westwood)
+Fix to show database properly when using slony_status (Guillaume Lelarge)
+Allow warning items for same_schema to be comma-separated (Guillaume Lelarge)
+Constraint definitions across Postgres versions match better in same_schema.
+Work against "EnterpriseDB" databases (Sivakumar Krishnamurthy and Greg Sabino Mullane)
+Separate perfdata with spaces (Jehan-Guillaume (ioguix) de Rorthais)
+Add new action "archive_ready" (Jehan-Guillaume (ioguix) de Rorthais)
+ +
+
Version 2.14.3 (March 1, 2010)
+
+ +
Allow slony_status action to handle more than one slave.
+Use commas to separate function args in same_schema output (Robert Treat)
+ +
+
Version 2.14.2 (February 18, 2010)
+
+ +
Change autovac_freeze default warn/critical back to 90%/95% (Robert Treat)
+Put all items one-per-line for relation size actions if --verbose=1
+ +
+
Version 2.14.1 (February 17, 2010)
+
+ +
Don't use $^T in logfile check, as script may be long-running
+Change the error string for the logfile action for easier exclusion
+  by programs like tail_n_mail
+ +
+
Version 2.14.0 (February 11, 2010)
+
+ +
Added the 'slony_status' action.
+Changed the logfile sleep from 0.5 to 1, as 0.5 gets rounded to 0 on some boxes!
+ +
+
Version 2.13.2 (February 4, 2010)
+
+ +
Allow timeout option to be used for logtime 'sleep' time.
+ +
+
Version 2.13.2 (February 4, 2010)
+
+ +
Show offending database for query_time action.
+Apply perflimit to main output for sequence action.
+Add 'noowner' option to same_schema action.
+Raise sleep timeout for logfile check to 15 seconds.
+ +
+
Version 2.13.1 (February 2, 2010)
+
+ +
Fix bug preventing column constraint differences from 2 > 1 for same_schema from being shown.
+Allow aliases 'dbname1', 'dbhost1', 'dbport1',etc.
+Added "nolanguage" as a filter for the same_schema option.
+Don't track "generic" table constraints (e.. $1, $2) using same_schema
+ +
+
Version 2.13.0 (January 29, 2010)
+
+ +
Allow "nofunctions" as a filter for the same_schema option.
+Added "noperm" as a filter for the same_schema option.
+Ignore dropped columns when considered positions for same_schema (Guillaume Lelarge)
+ +
+
Version 2.12.1 (December 3, 2009)
+
+ +
Change autovac_freeze default warn/critical from 90%/95% to 105%/120% (Marti Raudsepp)
+ +
+
Version 2.12.0 (December 3, 2009)
+
+ +
Allow the temporary directory to be specified via the "tempdir" argument,
+  for systems that need it (e.g. /tmp is not owned by root).
+Fix so old versions of Postgres (< 8.0) use the correct default database (Giles Westwood)
+For "same_schema" trigger mismatches, show the attached table.
+Add the new_version_bc check for Bucardo version checking.
+Add database name to perf output for last_vacuum|analyze (Guillaume Lelarge)
+Fix for bloat action against old versions of Postgres without the 'block_size' param.
+ +
+
Version 2.11.1 (August 27, 2009)
+
+ +
Proper Nagios output for last_vacuum|analyze actions. (Cédric Villemain)
+Proper Nagios output for locks action. (Cédric Villemain)
+Proper Nagios output for txn_wraparound action. (Cédric Villemain)
+Fix for constraints with embedded newlines for same_schema.
+Allow --exclude for all items when using same_schema.
+ +
+
Version 2.11.0 (August 23, 2009)
+
+ +
Add Nagios perf output to the wal_files check (Cédric Villemain)
+Add support for .check_postgresrc, per request from Albe Laurenz.
+Allow list of web fetch methods to be changed with the --get_method option.
+Add support for the --language argument, which overrides any ENV.
+Add the --no-check_postgresrc flag.
+Ensure check_postgresrc options are completely overridden by command-line options.
+Fix incorrect warning > critical logic in replicate_rows (Glyn Astill)
+ +
+
Version 2.10.0 (August 3, 2009)
+
-
File::Basename
+
For same_schema, compare view definitions, and compare languages.
+Make script into a global executable via the Makefile.PL file.
+Better output when comparing two databases.
+Proper Nagios output syntax for autovac_freeze and backends checks (Cédric Villemain)
-
File::Temp
+ +
Version 2.9.5 (July 24, 2009)
+
-
Time::HiRes (if $opt{showtime} is set to true, which is the default)
+
Don't use a LIMIT in check_bloat if --include is used. Per complaint from Jeff Frost.
-
-

The settings_checksum action requires the Digest::MD5 module.

-

The checkpoint action requires the Date::Parse module.

-

Some actions require access to external programs. If psql is not explicitly -specified, the command which is used to find it. The program /bin/df -is needed by the disk_space action.

-

-

-
-

DEVELOPMENT

-

Development happens using the git system. You can clone the latest version by doing:

-
- git clone git://bucardo.org/check_postgres.git
-

-

-
-

MAILING LIST

-

Three mailing lists are available. For discussions about the program, bug reports, -feature requests, and commit notices, send email to check_postgres@bucardo.org

-

https://mail.endcrypt.com/mailman/listinfo/check_postgres

-

A low-volume list for announcement of new versions and important notices is the -'check_postgres-announce' list:

-

https://mail.endcrypt.com/mailman/listinfo/check_postgres-announce

-

Source code changes (via git-commit) are sent to the -'check_postgres-commit' list:

-

https://mail.endcrypt.com/mailman/listinfo/check_postgres-commit

-

-

-
-

HISTORY

-

Items not specifically attributed are by GSM (Greg Sabino Mullane).

-
-
Version 2.21.0 September 24, 2013
- -
-
-  Fix issue with SQL steps in check_pgagent_jobs for sql steps which perform deletes
-    (Rob Emery via github pull)
-
-  Install man page in section 1. (Peter Eisentraut, bug 53, github issue 26)
-
-  Order lock types in check_locks output to make the ordering predictable;
-  setting SKIP_NETWORK_TESTS will skip the new_version tests; other minor test
-  suite fixes. (Christoph Berg)
-
-  Fix same_schema check on 9.3 by ignoring relminmxid differences in pg_class
-    (Christoph Berg)
-
-
Version 2.20.1 June 24, 2013
- -
-
-  Make connection check failures return CRITICAL not UNKNOWN
-    (Dominic Hargreaves)
-
-  Fix --reverse option when using string comparisons in custom queries
-    (Nathaniel Waisbrot)
-
-  Compute correct 'totalwastedbytes' in the bloat query
-    (Michael Renner)
-
-  Do not use pg_stats "inherited" column in bloat query, if the
-    database is 8.4 or older. (Greg Sabino Mullane, per bug 121)
-
-  Remove host reordering in hot_standby_delay check
-    (Josh Williams, with help from Jacobo Blasco)
-
-  Better output for the "simple" flag
-    (Greg Sabino Mullane)
-
-  Force same_schema to ignore the 'relallvisible' column
-    (Greg Sabino Mullane)
-
-
Version 2.20.0 March 13, 2013
- -
-
-  Add check for pgagent jobs (David E. Wheeler)
-
-  Force STDOUT to use utf8 for proper output
-    (Greg Sabino Mullane; reported by Emmanuel Lesouef)
-
-  Fixes for Postgres 9.2: new pg_stat_activity view,
-    and use pg_tablespace_location, (Josh Williams)
-
-  Allow for spaces in item lists when doing same_schema.
-
-  Allow txn_idle to work again for < 8.3 servers by switching to query_time.
-
-  Fix the check_bloat SQL to take inherited tables into account,
-    and assume 2k for non-analyzed columns. (Geert Pante)
-
-  Cache sequence information to speed up same_schema runs.
-
-  Fix --excludeuser in check_txn_idle (Mika Eloranta)
-
-  Fix user clause handling in check_txn_idle (Michael van Bracht)
-
-  Adjust docs to show colon as a better separator inside args for locks
-    (Charles Sprickman)
-
-  Fix undefined $SQL2 error in check_txn_idle [github issue 16] (Patric Bechtel)
-
-  Prevent "uninitialized value" warnings when showing the port (Henrik Ahlgren)
-
-  Do not assume everyone has a HOME [github issue 23]
-
-
Version 2.19.0 January 17, 2012
- -
-
-  Add the --assume-prod option (Cédric Villemain)
-
-  Add the cluster_id check (Cédric Villemain)
-
-  Improve settings_checksum and checkpoint tests (Cédric Villemain)
-
-  Do not do an inner join to pg_user when checking database size
-    (Greg Sabino Mullane; reported by Emmanuel Lesouef)
-
-  Use the full path when getting sequence information for same_schema.
-    (Greg Sabino Mullane; reported by Cindy Wise)
-
-  Fix the formula for calculating xlog positions (Euler Taveira de Oliveira)
-
-  Better ordering of output for bloat check - make indexes as important
-    as tables (Greg Sabino Mullane; reported by Jens Wilke)
-
-  Show the dbservice if it was used at top of same_schema output
-    (Mike Blackwell)
-
-  Better installation paths (Greg Sabino Mullane, per bug 53)
-
-
Version 2.18.0 October 2, 2011
- -
-
-  Redo the same_schema action. Use new --filter argument for all filtering.
-  Allow comparisons between any number of databases.
-  Remove the dbname2, dbport2, etc. arguments.
-  Allow comparison of the same db over time.
-
-  Swap db1 and db2 if the slave is 1 for the hot standby check (David E. Wheeler)
-
-  Allow multiple --schema arguments for the slony_status action (GSM and Jehan-Guillaume de Rorthais)
-
-  Fix ORDER BY in the last vacuum/analyze action (Nicolas Thauvin)
-
-  Fix check_hot_standby_delay perfdata output (Nicolas Thauvin)
-
-  Look in the correct place for the .ready files with the archive_ready action (Nicolas Thauvin)
-
-  New action: commitratio (Guillaume Lelarge)
-
-  New action: hitratio (Guillaume Lelarge)
-
-  Make sure --action overrides the symlink naming trick.
-
-  Set defaults for archive_ready and wal_files (Thomas Guettler, GSM)
-
-  Better output for wal_files and archive_ready (GSM)
-
-  Fix warning when client_port set to empty string (bug #79)
-
-  Account for "empty row" in -x output (i.e. source of functions).
-
-  Fix some incorrectly named data fields (Andy Lester)
-
-  Expand the number of pgbouncer actions (Ruslan Kabalin)
-
-  Give detailed information and refactor txn_idle, txn_time, and query_time
-    (Per request from bug #61)
-
-  Set maxalign to 8 in the bloat check if box identified as '64-bit'
-    (Michel Sijmons, bug #66)
-
-  Support non-standard version strings in the bloat check.
-    (Michel Sijmons and Gurjeet Singh, bug #66)
-
-  Do not show excluded databases in some output (Ruslan Kabalin)
-
-  Allow "and", "or" inside arguments (David E. Wheeler)
-
-  Add the "new_version_box" action.
-
-  Fix psql version regex (Peter Eisentraut, bug #69)
-
-  Add the --assume-standby-mode option (Ruslan Kabalin)
-
-  Note that txn_idle and query_time require 8.3 (Thomas Guettler)
-
-  Standardize and clean up all perfdata output (bug #52)
-
-  Exclude "idle in transaction" from the query_time check (bug #43)
-
-  Fix the perflimit for the bloat action (bug #50)
-
-  Clean up the custom_query action a bit.
-
-  Fix space in perfdata for hot_standby_delay action (Nicolas Thauvin)
-
-  Handle undef percents in check_fsm_relations (Andy Lester)
-
-  Fix typo in dbstats action (Stas Vitkovsky)
-
-  Fix MRTG for last vacuum and last_analyze actions.
-
-
Version 2.17.0 no public release
- -
Version 2.16.0 January 20, 2011
- -
-
-  Add new action 'hot_standby_delay' (Nicolas Thauvin)
-  Add cache-busting for the version-grabbing utilities.
-  Fix problem with going to next method for new_version_pg
-    (Greg Sabino Mullane, reported by Hywel Mallett in bug #65)
-  Allow /usr/local/etc as an alternative location for the 
-    check_postgresrc file (Hywel Mallett)
-  Do not use tgisconstraint in same_schema if Postgres >= 9
-    (Guillaume Lelarge)
-
-
Version 2.15.4 January 3, 2011
- -
-
-  Fix warning when using symlinks
-    (Greg Sabino Mullane, reported by Peter Eisentraut in bug #63)
-
-
Version 2.15.3 December 30, 2010
- -
-
-  Show OK for no matching txn_idle entries.
-
-
Version 2.15.2 December 28, 2010
- -
-
-  Better formatting of sizes in the bloat action output.
-
-  Remove duplicate perfs in bloat action output.
-
-
Version 2.15.1 December 27, 2010
- -
-
-  Fix problem when examining items in pg_settings (Greg Sabino Mullane)
-
-  For connection test, return critical, not unknown, on FATAL errors
-    (Greg Sabino Mullane, reported by Peter Eisentraut in bug #62)
-
Version 2.15.0 November 8, 2010
- +
Version 2.9.4 (July 21, 2009)
-
-  Add --quiet argument to suppress output on OK Nagios results
-  Add index comparison for same_schema (Norman Yamada and Greg Sabino Mullane)
-  Use $ENV{PGSERVICE} instead of "service=" to prevent problems (Guillaume Lelarge)
-  Add --man option to show the entire manual. (Andy Lester)
-  Redo the internal run_command() sub to use -x and hashes instead of regexes.
-  Fix error in custom logic (Andreas Mager)
-  Add the "pgbouncer_checksum" action (Guillaume Lelarge)
-  Fix regex to work on WIN32 for check_fsm_relations and check_fsm_pages (Luke Koops)
-  Don't apply a LIMIT when using --exclude on the bloat action (Marti Raudsepp)
-  Change the output of query_time to show pid,user,port, and address (Giles Westwood)
-  Fix to show database properly when using slony_status (Guillaume Lelarge)
-  Allow warning items for same_schema to be comma-separated (Guillaume Lelarge)
-  Constraint definitions across Postgres versions match better in same_schema.
-  Work against "EnterpriseDB" databases (Sivakumar Krishnamurthy and Greg Sabino Mullane)
-  Separate perfdata with spaces (Jehan-Guillaume (ioguix) de Rorthais)
-  Add new action "archive_ready" (Jehan-Guillaume (ioguix) de Rorthais)
-
-
Version 2.14.3 (March 1, 2010)
-
-
-  Allow slony_status action to handle more than one slave.
-  Use commas to separate function args in same_schema output (Robert Treat)
-
-
Version 2.14.2 (February 18, 2010)
+
More French translations (Guillaume Lelarge)
-
-
-  Change autovac_freeze default warn/critical back to 90%/95% (Robert Treat)
-  Put all items one-per-line for relation size actions if --verbose=1
-
Version 2.14.1 (February 17, 2010)
- +
Version 2.9.3 (July 14, 2009)
-
-  Don't use $^T in logfile check, as script may be long-running
-  Change the error string for the logfile action for easier exclusion
-    by programs like tail_n_mail
-
-
Version 2.14.0 (February 11, 2010)
-
-
-  Added the 'slony_status' action.
-  Changed the logfile sleep from 0.5 to 1, as 0.5 gets rounded to 0 on some boxes!
-
-
Version 2.13.2 (February 4, 2010)
+
Quote dbname in perf output for the backends check. (Davide Abrigo)
+Add 'fetch' as an alternative method for new_version checks, as this 
+  comes by default with FreeBSD. (Hywel Mallett)
-
-
-  Allow timeout option to be used for logtime 'sleep' time.
-
Version 2.13.2 (February 4, 2010)
- +
Version 2.9.2 (July 12, 2009)
-
-  Show offending database for query_time action.
-  Apply perflimit to main output for sequence action.
-  Add 'noowner' option to same_schema action.
-  Raise sleep timeout for logfile check to 15 seconds.
-
-
Version 2.13.1 (February 2, 2010)
-
-
-  Fix bug preventing column constraint differences from 2 > 1 for same_schema from being shown.
-  Allow aliases 'dbname1', 'dbhost1', 'dbport1',etc.
-  Added "nolanguage" as a filter for the same_schema option.
-  Don't track "generic" table constraints (e.. $1, $2) using same_schema
-
-
Version 2.13.0 (January 29, 2010)
+
Allow dots and dashes in database name for the backends check (Davide Abrigo)
+Check and display the database for each match in the bloat check (Cédric Villemain)
+Handle 'too many connections' FATAL error in the backends check with a critical,
+  rather than a generic error (Greg, idea by Jürgen Schulz-Brüssel)
+Do not allow perflimit to interfere with exclusion rules in the vacuum and 
+  analyze tests. (Greg, bug reported by Jeff Frost)
-
-
-  Allow "nofunctions" as a filter for the same_schema option.
-  Added "noperm" as a filter for the same_schema option.
-  Ignore dropped columns when considered positions for same_schema (Guillaume Lelarge)
-
Version 2.12.1 (December 3, 2009)
- +
Version 2.9.1 (June 12, 2009)
-
-  Change autovac_freeze default warn/critical from 90%/95% to 105%/120% (Marti Raudsepp)
-
-
Version 2.12.0 (December 3, 2009)
-
-
-  Allow the temporary directory to be specified via the "tempdir" argument,
-    for systems that need it (e.g. /tmp is not owned by root).
-  Fix so old versions of Postgres (< 8.0) use the correct default database (Giles Westwood)
-  For "same_schema" trigger mismatches, show the attached table.
-  Add the new_version_bc check for Bucardo version checking.
-  Add database name to perf output for last_vacuum|analyze (Guillaume Lelarge)
-  Fix for bloat action against old versions of Postgres without the 'block_size' param.
-
-
Version 2.11.1 (August 27, 2009)
+
Fix for multiple databases with the check_bloat action (Mark Kirkwood)
+Fixes and improvements to the same_schema action (Jeff Boes)
+Write tests for same_schema, other minor test fixes (Jeff Boes)
-
-
-  Proper Nagios output for last_vacuum|analyze actions. (Cédric Villemain)
-  Proper Nagios output for locks action. (Cédric Villemain)
-  Proper Nagios output for txn_wraparound action. (Cédric Villemain)
-  Fix for constraints with embedded newlines for same_schema.
-  Allow --exclude for all items when using same_schema.
-
Version 2.11.0 (August 23, 2009)
- +
Version 2.9.0 (May 28, 2009)
-
-  Add Nagios perf output to the wal_files check (Cédric Villemain)
-  Add support for .check_postgresrc, per request from Albe Laurenz.
-  Allow list of web fetch methods to be changed with the --get_method option.
-  Add support for the --language argument, which overrides any ENV.
-  Add the --no-check_postgresrc flag.
-  Ensure check_postgresrc options are completely overridden by command-line options.
-  Fix incorrect warning > critical logic in replicate_rows (Glyn Astill)
-
-
Version 2.10.0 (August 3, 2009)
-
-
-  For same_schema, compare view definitions, and compare languages.
-  Make script into a global executable via the Makefile.PL file.
-  Better output when comparing two databases.
-  Proper Nagios output syntax for autovac_freeze and backends checks (Cédric Villemain)
-
-
Version 2.9.5 (July 24, 2009)
+
Added the same_schema action (Greg)
-
-
-  Don't use a LIMIT in check_bloat if --include is used. Per complaint from Jeff Frost.
-
Version 2.9.4 (July 21, 2009)
- +
Version 2.8.1 (May 15, 2009)
-
-  More French translations (Guillaume Lelarge)
-
-
Version 2.9.3 (July 14, 2009)
-
-
-  Quote dbname in perf output for the backends check. (Davide Abrigo)
-  Add 'fetch' as an alternative method for new_version checks, as this 
-    comes by default with FreeBSD. (Hywel Mallett)
-
-
Version 2.9.2 (July 12, 2009)
+
Added timeout via statement_timeout in addition to perl alarm (Greg)
-
-
-  Allow dots and dashes in database name for the backends check (Davide Abrigo)
-  Check and display the database for each match in the bloat check (Cédric Villemain)
-  Handle 'too many connections' FATAL error in the backends check with a critical,
-    rather than a generic error (Greg, idea by Jürgen Schulz-Brüssel)
-  Do not allow perflimit to interfere with exclusion rules in the vacuum and 
-    analyze tests. (Greg, bug reported by Jeff Frost)
-
Version 2.9.1 (June 12, 2009)
- +
Version 2.8.0 (May 4, 2009)
-
-  Fix for multiple databases with the check_bloat action (Mark Kirkwood)
-  Fixes and improvements to the same_schema action (Jeff Boes)
-  Write tests for same_schema, other minor test fixes (Jeff Boes)
-
-
Version 2.9.0 (May 28, 2009)
-
-
-  Added the same_schema action (Greg)
-
-
Version 2.8.1 (May 15, 2009)
+
Added internationalization support (Greg)
+Added the 'disabled_triggers' check (Greg)
+Added the 'prepared_txns' check (Greg)
+Added the 'new_version_cp' and 'new_version_pg' checks (Greg)
+French translations (Guillaume Lelarge)
+Make the backends search return ok if no matches due to inclusion rules,
+  per report by Guillaume Lelarge (Greg)
+Added comprehensive unit tests (Greg, Jeff Boes, Selena Deckelmann)
+Make fsm_pages and fsm_relations handle 8.4 servers smoothly. (Greg)
+Fix missing 'upd' field in show_dbstats (Andras Fabian)
+Allow ENV{PGCONTROLDATA} and ENV{PGBINDIR}. (Greg)
+Add various Perl module infrastructure (e.g. Makefile.PL) (Greg)
+Fix incorrect regex in txn_wraparound (Greg)
+For txn_wraparound: consistent ordering and fix duplicates in perf output (Andras Fabian)
+Add in missing exabyte regex check (Selena Deckelmann)
+Set stats to zero if we bail early due to USERWHERECLAUSE (Andras Fabian)
+Add additional items to dbstats output (Andras Fabian)
+Remove --schema option from the fsm_ checks. (Greg Mullane and Robert Treat)
+Handle case when ENV{PGUSER} is set. (Andy Lester)
+Many various fixes. (Jeff Boes)
+Fix --dbservice: check version and use ENV{PGSERVICE} for old versions (Cédric Villemain)
-
-
-  Added timeout via statement_timeout in addition to perl alarm (Greg)
-
Version 2.8.0 (May 4, 2009)
- +
Version 2.7.3 (February 10, 2009)
-
-  Added internationalization support (Greg)
-  Added the 'disabled_triggers' check (Greg)
-  Added the 'prepared_txns' check (Greg)
-  Added the 'new_version_cp' and 'new_version_pg' checks (Greg)
-  French translations (Guillaume Lelarge)
-  Make the backends search return ok if no matches due to inclusion rules,
-    per report by Guillaume Lelarge (Greg)
-  Added comprehensive unit tests (Greg, Jeff Boes, Selena Deckelmann)
-  Make fsm_pages and fsm_relations handle 8.4 servers smoothly. (Greg)
-  Fix missing 'upd' field in show_dbstats (Andras Fabian)
-  Allow ENV{PGCONTROLDATA} and ENV{PGBINDIR}. (Greg)
-  Add various Perl module infrastructure (e.g. Makefile.PL) (Greg)
-  Fix incorrect regex in txn_wraparound (Greg)
-  For txn_wraparound: consistent ordering and fix duplicates in perf output (Andras Fabian)
-  Add in missing exabyte regex check (Selena Deckelmann)
-  Set stats to zero if we bail early due to USERWHERECLAUSE (Andras Fabian)
-  Add additional items to dbstats output (Andras Fabian)
-  Remove --schema option from the fsm_ checks. (Greg Mullane and Robert Treat)
-  Handle case when ENV{PGUSER} is set. (Andy Lester)
-  Many various fixes. (Jeff Boes)
-  Fix --dbservice: check version and use ENV{PGSERVICE} for old versions (Cédric Villemain)
-
-
Version 2.7.3 (February 10, 2009)
-
-
-  Make the sequence action check if sequence being used for a int4 column and
-  react appropriately. (Michael Glaesemann)
-
-
Version 2.7.2 (February 9, 2009)
+
Make the sequence action check if sequence being used for a int4 column and
+react appropriately. (Michael Glaesemann)
-
-
-  Fix to prevent multiple groupings if db arguments given.
-
Version 2.7.1 (February 6, 2009)
- +
Version 2.7.2 (February 9, 2009)
-
-  Allow the -p argument for port to work again.
-
-
Version 2.7.0 (February 4, 2009)
-
-
-  Do not require a connection argument, but use defaults and ENV variables when 
-    possible: PGHOST, PGPORT, PGUSER, PGDATABASE.
-
-
Version 2.6.1 (February 4, 2009)
+
Fix to prevent multiple groupings if db arguments given.
-
-
-  Only require Date::Parse to be loaded if using the checkpoint action.
-
Version 2.6.0 (January 26, 2009)
- +
Version 2.7.1 (February 6, 2009)
-
-  Add the 'checkpoint' action.
-
-
Version 2.5.4 (January 7, 2009)
-
-
-  Better checking of $opt{dbservice} structure (Cédric Villemain)
-  Fix time display in timesync action output (Selena Deckelmann)
-  Fix documentation typos (Josh Tolley)
-
-
Version 2.5.3 (December 17, 2008)
+
Allow the -p argument for port to work again.
-
-
-  Minor fix to regex in verify_version (Lee Jensen)
-
Version 2.5.2 (December 16, 2008)
- +
Version 2.7.0 (February 4, 2009)
-
-  Minor documentation tweak.
-
-
Version 2.5.1 (December 11, 2008)
-
-
-  Add support for --noidle flag to prevent backends action from counting idle processes.
-  Patch by Selena Deckelmann.
-
-  Fix small undefined warning when not using --dbservice.
-
-
Version 2.5.0 (December 4, 2008)
+
Do not require a connection argument, but use defaults and ENV variables when 
+  possible: PGHOST, PGPORT, PGUSER, PGDATABASE.
-
-
-  Add support for the pg_Service.conf file with the --dbservice option.
-
Version 2.4.3 (November 7, 2008)
- +
Version 2.6.1 (February 4, 2009)
-
-  Fix options for replicate_row action, per report from Jason Gordon.
-
-
Version 2.4.2 (November 6, 2008)
-
-
-  Wrap File::Temp::cleanup() calls in eval, in case File::Temp is an older version.
-  Patch by Chris Butler.
-
-
Version 2.4.1 (November 5, 2008)
+
Only require Date::Parse to be loaded if using the checkpoint action.
-
-
-  Cast numbers to numeric to support sequences ranges > bigint in check_sequence action.
-  Thanks to Scott Marlowe for reporting this.
-
Version 2.4.0 (October 26, 2008)
- +
Version 2.6.0 (January 26, 2009)
-
- Add Cacti support with the dbstats action.
- Pretty up the time output for last vacuum and analyze actions.
- Show the percentage of backends on the check_backends action.
-
-
Version 2.3.10 (October 23, 2008)
-
-
- Fix minor warning in action check_bloat with multiple databases.
- Allow warning to be greater than critical when using the --reverse option.
- Support the --perflimit option for the check_sequence action.
-
-
Version 2.3.9 (October 23, 2008)
+
Add the 'checkpoint' action.
-
-
- Minor tweak to way we store the default port.
-
Version 2.3.8 (October 21, 2008)
- +
Version 2.5.4 (January 7, 2009)
-
- Allow the default port to be changed easily.
- Allow transform of simple output by MB, GB, etc.
-
-
Version 2.3.7 (October 14, 2008)
-
-
- Allow multiple databases in 'sequence' action. Reported by Christoph Zwerschke.
-
-
Version 2.3.6 (October 13, 2008)
+
Better checking of $opt{dbservice} structure (Cédric Villemain)
+Fix time display in timesync action output (Selena Deckelmann)
+Fix documentation typos (Josh Tolley)
-
-
- Add missing $schema to check_fsm_pages. (Robert Treat)
-
Version 2.3.5 (October 9, 2008)
- +
Version 2.5.3 (December 17, 2008)
-
- Change option 'checktype' to 'valtype' to prevent collisions with -c[ritical]
- Better handling of errors.
-
-
Version 2.3.4 (October 9, 2008)
-
-
- Do explicit cleanups of the temp directory, per problems reported by sb@nnx.com.
-
-
Version 2.3.3 (October 8, 2008)
+
Minor fix to regex in verify_version (Lee Jensen)
-
-
- Account for cases where some rounding queries give -0 instead of 0.
- Thanks to Glyn Astill for helping to track this down.
-
Version 2.3.2 (October 8, 2008)
- +
Version 2.5.2 (December 16, 2008)
-
- Always quote identifiers in check_replicate_row action.
-
-
Version 2.3.1 (October 7, 2008)
-
-
- Give a better error if one of the databases cannot be reached.
-
-
Version 2.3.0 (October 4, 2008)
+
Minor documentation tweak.
-
-
- Add the "sequence" action, thanks to Gavin M. Roy for the idea.
- Fix minor problem with autovac_freeze action when using MRTG output.
- Allow output argument to be case-insensitive.
- Documentation fixes.
-
Version 2.2.4 (October 3, 2008)
- +
Version 2.5.1 (December 11, 2008)
-
- Fix some minor typos
-
-
Version 2.2.3 (October 1, 2008)
-
-
- Expand range of allowed names for --repinfo argument (Glyn Astill)
- Documentation tweaks.
-
-
Version 2.2.2 (September 30, 2008)
+
Add support for --noidle flag to prevent backends action from counting idle processes.
+Patch by Selena Deckelmann.
 
-

-
- Fixes for minor output and scoping problems.

-

-
Version 2.2.1 (September 28, 2008)

+Fix small undefined warning when not using --dbservice.
-
-
- Add MRTG output to fsm_pages and fsm_relations.
- Force error messages to one-line for proper Nagios output.
- Check for invalid prereqs on failed command. From conversations with Euler Taveira de Oliveira.
- Tweak the fsm_pages formula a little.
-
Version 2.2.0 (September 25, 2008)
- +
Version 2.5.0 (December 4, 2008)
-
- Add fsm_pages and fsm_relations actions. (Robert Treat)
-
-
Version 2.1.4 (September 22, 2008)
-
-
- Fix for race condition in txn_time action.
- Add --debugoutput option.
-
-
Version 2.1.3 (September 22, 2008)
+
Add support for the pg_Service.conf file with the --dbservice option.
-
-
- Allow alternate arguments "dbhost" for "host" and "dbport" for "port".
- Output a zero as default value for second line of MRTG output.
-
Version 2.1.2 (July 28, 2008)
- +
Version 2.4.3 (November 7, 2008)
-
- Fix sorting error in the "disk_space" action for non-Nagios output.
- Allow --simple as a shortcut for --output=simple.
-
-
Version 2.1.1 (July 22, 2008)
-
-
- Don't check databases with datallowconn false for the "autovac_freeze" action.
-
-
Version 2.1.0 (July 18, 2008)
+
Fix options for replicate_row action, per report from Jason Gordon.
-
-
- Add the "autovac_freeze" action, thanks to Robert Treat for the idea and design.
- Put an ORDER BY on the "txn_wraparound" action.
-
Version 2.0.1 (July 16, 2008)
- +
Version 2.4.2 (November 6, 2008)
-
- Optimizations to speed up the "bloat" action quite a bit.
- Fix "version" action to not always output in mrtg mode.
-
-
Version 2.0.0 (July 15, 2008)
-
-
- Add support for MRTG and "simple" output options.
- Many small improvements to nearly all actions.
-
-
Version 1.9.1 (June 24, 2008)
+
Wrap File::Temp::cleanup() calls in eval, in case File::Temp is an older version.
+Patch by Chris Butler.
+ +
Version 2.4.1 (November 5, 2008)
-
- Fix an error in the bloat SQL in 1.9.0
- Allow percentage arguments to be over 99%
- Allow percentages in the bloat --warning and --critical (thanks to Robert Treat for the idea)
+ +
Cast numbers to numeric to support sequences ranges > bigint in check_sequence action.
+Thanks to Scott Marlowe for reporting this.
+
-
Version 1.9.0 (June 22, 2008)
+
Version 2.4.0 (October 26, 2008)
+
+ +
Add Cacti support with the dbstats action.
+Pretty up the time output for last vacuum and analyze actions.
+Show the percentage of backends on the check_backends action.
+
+
Version 2.3.10 (October 23, 2008)
-
- Don't include information_schema in certain checks. (Jeff Frost)
- Allow --include and --exclude to use schemas by using a trailing period.
+ +
Fix minor warning in action check_bloat with multiple databases.
+Allow warning to be greater than critical when using the --reverse option.
+Support the --perflimit option for the check_sequence action.
+
-
Version 1.8.5 (June 22, 2008)
+
Version 2.3.9 (October 23, 2008)
+
+
Minor tweak to way we store the default port.
+ +
+
Version 2.3.8 (October 21, 2008)
-
- Output schema name before table name where appropriate.
- Thanks to Jeff Frost.
+ +
Allow the default port to be changed easily.
+Allow transform of simple output by MB, GB, etc.
+
-
Version 1.8.4 (June 19, 2008)
+
Version 2.3.7 (October 14, 2008)
+
+ +
Allow multiple databases in 'sequence' action. Reported by Christoph Zwerschke.
+
+
Version 2.3.6 (October 13, 2008)
-
- Better detection of problems in --replicate_row.
+ +
Add missing $schema to check_fsm_pages. (Robert Treat)
+
-
Version 1.8.3 (June 18, 2008)
+
Version 2.3.5 (October 9, 2008)
+
+
Change option 'checktype' to 'valtype' to prevent collisions with -c[ritical]
+Better handling of errors.
+ +
+
Version 2.3.4 (October 9, 2008)
-
- Fix 'backends' action: there may be no rows in pg_stat_activity, so run a second
-   query if needed to find the max_connections setting.
- Thanks to Jeff Frost for the bug report.
+ +
Do explicit cleanups of the temp directory, per problems reported by sb@nnx.com.
+
-
Version 1.8.2 (June 10, 2008)
+
Version 2.3.3 (October 8, 2008)
+
+ +
Account for cases where some rounding queries give -0 instead of 0.
+Thanks to Glyn Astill for helping to track this down.
+
+
Version 2.3.2 (October 8, 2008)
-
- Changes to allow working under Nagios' embedded Perl mode. (Ioannis Tambouras)
+ +
Always quote identifiers in check_replicate_row action.
+
-
Version 1.8.1 (June 9, 2008)
+
Version 2.3.1 (October 7, 2008)
+
+
Give a better error if one of the databases cannot be reached.
+ +
+
Version 2.3.0 (October 4, 2008)
-
- Allow 'bloat' action to work on Postgres version 8.0.
- Allow for different commands to be run for each action depending on the server version.
- Give better warnings when running actions not available on older Postgres servers.
+ +
Add the "sequence" action, thanks to Gavin M. Roy for the idea.
+Fix minor problem with autovac_freeze action when using MRTG output.
+Allow output argument to be case-insensitive.
+Documentation fixes.
+
-
Version 1.8.0 (June 3, 2008)
+
Version 2.2.4 (October 3, 2008)
+
+ +
Fix some minor typos
+
+
Version 2.2.3 (October 1, 2008)
-
- Add the --reverse option to the custom_query action.
+ +
Expand range of allowed names for --repinfo argument (Glyn Astill)
+Documentation tweaks.
+
-
Version 1.7.1 (June 2, 2008)
+
Version 2.2.2 (September 30, 2008)
+
+
Fixes for minor output and scoping problems.
+ +
+
Version 2.2.1 (September 28, 2008)
-
- Fix 'query_time' action: account for race condition in which zero rows appear in pg_stat_activity.
- Thanks to Dustin Black for the bug report.
+ +
Add MRTG output to fsm_pages and fsm_relations.
+Force error messages to one-line for proper Nagios output.
+Check for invalid prereqs on failed command. From conversations with Euler Taveira de Oliveira.
+Tweak the fsm_pages formula a little.
+
-
Version 1.7.0 (May 11, 2008)
+
Version 2.2.0 (September 25, 2008)
+
+ +
Add fsm_pages and fsm_relations actions. (Robert Treat)
+
+
Version 2.1.4 (September 22, 2008)
-
- Add --replicate_row action
+ +
Fix for race condition in txn_time action.
+Add --debugoutput option.
+
-
Version 1.6.1 (May 11, 2008)
+
Version 2.1.3 (September 22, 2008)
+
+
Allow alternate arguments "dbhost" for "host" and "dbport" for "port".
+Output a zero as default value for second line of MRTG output.
+ +
+
Version 2.1.2 (July 28, 2008)
-
- Add --symlinks option as a shortcut to --action=rebuild_symlinks
+ +
Fix sorting error in the "disk_space" action for non-Nagios output.
+Allow --simple as a shortcut for --output=simple.
+
-
Version 1.6.0 (May 11, 2008)
+
Version 2.1.1 (July 22, 2008)
+
+ +
Don't check databases with datallowconn false for the "autovac_freeze" action.
+
+
Version 2.1.0 (July 18, 2008)
-
- Add the custom_query action.
+ +
Add the "autovac_freeze" action, thanks to Robert Treat for the idea and design.
+Put an ORDER BY on the "txn_wraparound" action.
+
-
Version 1.5.2 (May 2, 2008)
+
Version 2.0.1 (July 16, 2008)
+
+
Optimizations to speed up the "bloat" action quite a bit.
+Fix "version" action to not always output in mrtg mode.
+ +
+
Version 2.0.0 (July 15, 2008)
-
- Fix problem with too eager creation of custom pgpass file.
+ +
Add support for MRTG and "simple" output options.
+Many small improvements to nearly all actions.
+
-
Version 1.5.1 (April 17, 2008)
+
Version 1.9.1 (June 24, 2008)
+
+ +
Fix an error in the bloat SQL in 1.9.0
+Allow percentage arguments to be over 99%
+Allow percentages in the bloat --warning and --critical (thanks to Robert Treat for the idea)
+
+
Version 1.9.0 (June 22, 2008)
-
- Add example Nagios configuration settings (Brian A. Seklecki)
+ +
Don't include information_schema in certain checks. (Jeff Frost)
+Allow --include and --exclude to use schemas by using a trailing period.
+
-
Version 1.5.0 (April 16, 2008)
+
Version 1.8.5 (June 22, 2008)
+
+
Output schema name before table name where appropriate.
+Thanks to Jeff Frost.
+ +
+
Version 1.8.4 (June 19, 2008)
-
- Add the --includeuser and --excludeuser options. Documentation cleanup.
+ +
Better detection of problems in --replicate_row.
+
-
Version 1.4.3 (April 16, 2008)
+
Version 1.8.3 (June 18, 2008)
+
+ +
Fix 'backends' action: there may be no rows in pg_stat_activity, so run a second
+  query if needed to find the max_connections setting.
+Thanks to Jeff Frost for the bug report.
+
+
Version 1.8.2 (June 10, 2008)
-
- Add in the 'output' concept for future support of non-Nagios programs.
+ +
Changes to allow working under Nagios' embedded Perl mode. (Ioannis Tambouras)
+
-
Version 1.4.2 (April 8, 2008)
+
Version 1.8.1 (June 9, 2008)
+
+
Allow 'bloat' action to work on Postgres version 8.0.
+Allow for different commands to be run for each action depending on the server version.
+Give better warnings when running actions not available on older Postgres servers.
+ +
+
Version 1.8.0 (June 3, 2008)
-
- Fix bug preventing --dbpass argument from working (Robert Treat).
+ +
Add the --reverse option to the custom_query action.
+
-
Version 1.4.1 (April 4, 2008)
+
Version 1.7.1 (June 2, 2008)
+
+ +
Fix 'query_time' action: account for race condition in which zero rows appear in pg_stat_activity.
+Thanks to Dustin Black for the bug report.
+
+
Version 1.7.0 (May 11, 2008)
-
- Minor documentation fixes.
+ +
Add --replicate_row action
+
-
Version 1.4.0 (April 2, 2008)
+
Version 1.6.1 (May 11, 2008)
+
+
Add --symlinks option as a shortcut to --action=rebuild_symlinks
+ +
+
Version 1.6.0 (May 11, 2008)
-
- Have 'wal_files' action use pg_ls_dir (idea by Robert Treat).
- For last_vacuum and last_analyze, respect autovacuum effects, add separate 
-   autovacuum checks (ideas by Robert Treat).
+ +
Add the custom_query action.
+
-
Version 1.3.1 (April 2, 2008)
+
Version 1.5.2 (May 2, 2008)
+
+ +
Fix problem with too eager creation of custom pgpass file.
+
+
Version 1.5.1 (April 17, 2008)
-
- Have txn_idle use query_start, not xact_start.
+ +
Add example Nagios configuration settings (Brian A. Seklecki)
+
-
Version 1.3.0 (March 23, 2008)
+
Version 1.5.0 (April 16, 2008)
+
+
Add the --includeuser and --excludeuser options. Documentation cleanup.
+ +
+
Version 1.4.3 (April 16, 2008)
-
- Add in txn_idle and txn_time actions.
+ +
Add in the 'output' concept for future support of non-Nagios programs.
+
-
Version 1.2.0 (February 21, 2008)
+
Version 1.4.2 (April 8, 2008)
+
+ +
Fix bug preventing --dbpass argument from working (Robert Treat).
+
+
Version 1.4.1 (April 4, 2008)
-
- Add the 'wal_files' action, which counts the number of WAL files
-   in your pg_xlog directory.
- Fix some typos in the docs.
- Explicitly allow -v as an argument.
- Allow for a null syslog_facility in the 'logfile' action.
+ +
Minor documentation fixes.
+
-
Version 1.1.2 (February 5, 2008)
+
Version 1.4.0 (April 2, 2008)
+
+
Have 'wal_files' action use pg_ls_dir (idea by Robert Treat).
+For last_vacuum and last_analyze, respect autovacuum effects, add separate 
+  autovacuum checks (ideas by Robert Treat).
+ +
+
Version 1.3.1 (April 2, 2008)
-
- Fix error preventing --action=rebuild_symlinks from working.
+ +
Have txn_idle use query_start, not xact_start.
+
-
Version 1.1.1 (February 3, 2008)
+
Version 1.3.0 (March 23, 2008)
+
+ +
Add in txn_idle and txn_time actions.
+
+
Version 1.2.0 (February 21, 2008)
-
- Switch vacuum and analyze date output to use 'DD', not 'D'. (Glyn Astill)
+ +
Add the 'wal_files' action, which counts the number of WAL files
+  in your pg_xlog directory.
+Fix some typos in the docs.
+Explicitly allow -v as an argument.
+Allow for a null syslog_facility in the 'logfile' action.
+
-
Version 1.1.0 (December 16, 2008)
+
Version 1.1.2 (February 5, 2008)
+
+
Fix error preventing --action=rebuild_symlinks from working.
+ +
+
Version 1.1.1 (February 3, 2008)
-
- Fixes, enhancements, and performance tracking.
- Add performance data tracking via --showperf and --perflimit
- Lots of refactoring and cleanup of how actions handle arguments.
- Do basic checks to figure out syslog file for 'logfile' action.
- Allow for exact matching of beta versions with 'version' action.
- Redo the default arguments to only populate when neither 'warning' nor 'critical' is provided.
- Allow just warning OR critical to be given for the 'timesync' action.
- Remove 'redirect_stderr' requirement from 'logfile' due to 8.3 changes.
- Actions 'last_vacuum' and 'last_analyze' are 8.2 only (Robert Treat)
+ +
Switch vacuum and analyze date output to use 'DD', not 'D'. (Glyn Astill)
+
-
Version 1.0.16 (December 7, 2007)
+
Version 1.1.0 (December 16, 2008)
+
+ +
Fixes, enhancements, and performance tracking.
+Add performance data tracking via --showperf and --perflimit
+Lots of refactoring and cleanup of how actions handle arguments.
+Do basic checks to figure out syslog file for 'logfile' action.
+Allow for exact matching of beta versions with 'version' action.
+Redo the default arguments to only populate when neither 'warning' nor 'critical' is provided.
+Allow just warning OR critical to be given for the 'timesync' action.
+Remove 'redirect_stderr' requirement from 'logfile' due to 8.3 changes.
+Actions 'last_vacuum' and 'last_analyze' are 8.2 only (Robert Treat)
+
+
Version 1.0.16 (December 7, 2007)
-
- First public release, December 2007
+ +
First public release, December 2007
+
-

-

-
-

BUGS AND LIMITATIONS

+ +

BUGS AND LIMITATIONS

+

The index bloat size optimization is rough.

+

Some actions may not work on older versions of Postgres (before 8.0).

-

Please report any problems to check_postgres@bucardo.org

-

-

-
-

AUTHOR

-

Greg Sabino Mullane <greg@endpoint.com>

-

-

-
-

NAGIOS EXAMPLES

+ +

Please report any problems to check_postgres@bucardo.org

+ +

AUTHOR

+ +

Greg Sabino Mullane <greg@turnstep.com>

+ +

NAGIOS EXAMPLES

+

Some example Nagios configuration settings using this script:

-
- define command {
-     command_name    check_postgres_size
-     command_line    $USER2$/check_postgres.pl -H $HOSTADDRESS$ -u pgsql -db postgres --action database_size -w $ARG1$ -c $ARG2$
- }
-
- define command {
-     command_name    check_postgres_locks
-     command_line    $USER2$/check_postgres.pl -H $HOSTADDRESS$ -u pgsql -db postgres --action locks -w $ARG1$ -c $ARG2$
- }
-
- define service {
-     use                    generic-other
-     host_name              dbhost.gtld
-     service_description    dbhost PostgreSQL Service Database Usage Size
-     check_command          check_postgres_size!256000000!512000000
- }
-
- define service {
-     use                    generic-other
-     host_name              dbhost.gtld
-     service_description    dbhost PostgreSQL Service Database Locks
-     check_command          check_postgres_locks!2!3
- }
-

-

-
-

LICENSE AND COPYRIGHT

-

Copyright (c) 2007-2013 Greg Sabino Mullane <greg@endpoint.com>.

-

Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met:

-
-  1. Redistributions of source code must retain the above copyright notice, 
-     this list of conditions and the following disclaimer.
-  2. Redistributions in binary form must reproduce the above copyright notice, 
-     this list of conditions and the following disclaimer in the documentation 
-     and/or other materials provided with the distribution.
-

THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND ANY EXPRESS OR IMPLIED -WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF -MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO -EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, -EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT -OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS -INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN -CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING -IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY -OF SUCH DAMAGE.

+ +
define command {
+    command_name    check_postgres_size
+    command_line    $USER2$/check_postgres.pl -H $HOSTADDRESS$ -u pgsql -db postgres --action database_size -w $ARG1$ -c $ARG2$
+}
+
+define command {
+    command_name    check_postgres_locks
+    command_line    $USER2$/check_postgres.pl -H $HOSTADDRESS$ -u pgsql -db postgres --action locks -w $ARG1$ -c $ARG2$
+}
+
+
+define service {
+    use                    generic-other
+    host_name              dbhost.gtld
+    service_description    dbhost PostgreSQL Service Database Usage Size
+    check_command          check_postgres_size!256000000!512000000
+}
+
+define service {
+    use                    generic-other
+    host_name              dbhost.gtld
+    service_description    dbhost PostgreSQL Service Database Locks
+    check_command          check_postgres_locks!2!3
+}
+ +

LICENSE AND COPYRIGHT

+ +

Copyright 2007 - 2025 Greg Sabino Mullane <greg@turnstep.com>.

+ +

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

+ +
1. Redistributions of source code must retain the above copyright notice, 
+   this list of conditions and the following disclaimer.
+2. Redistributions in binary form must reproduce the above copyright notice, 
+   this list of conditions and the following disclaimer in the documentation 
+   and/or other materials provided with the distribution.
+ +

THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

+ + + diff --git a/perlcriticrc b/perlcriticrc index ee42c309..915b182b 100644 --- a/perlcriticrc +++ b/perlcriticrc @@ -4,8 +4,9 @@ verbose = 8 profile-strictness = quiet [Documentation::PodSpelling] -stop_words = Mullane Nagios Slony Slony's nols salesrep psql dbname postgres USERNAME usernames dbuser pgpass nagios stderr showperf symlinked timesync criticals quirm lancre exabytes sami includeuser excludeuser flagg tardis WAL tablespaces tablespace perflimit burrick mallory grimm oskar ExclusiveLock garrett artemus queryname speedtest checksum checksums morpork klatch pluto faceoff slon greg watson franklin wilkins scott Sabino Seklecki dbpass autovacuum Astill refactoring NAGIOS localhost cronjob symlink symlinks backends snazzo logfile syslog parens plugin Cwd Ioannis Tambouras schemas SQL MRTG mrtg uptime datallowconn dbhost dbport ok contrib pageslots robert dylan emma fsm minvalue nextval dbstats del ret upd Bucardo noidle bucardo petabytes zettabytes tuples noobjectnames noposition nofuncbody slony noperms nolanguage battlestar pgbouncer pgBouncer datadir wget PostgreSQL commitratio idxscan idxtupread idxtupfetch idxblksread idxblkshit seqscan seqtupread hitratio xlog boxinfo pgbouncer's login maxwait noname noschema noperm filenames GSM EB executables plasmid unlinked pgAgent MERCHANTABILITY +stop_words = artemus Astill autovacuum backends battlestar boxinfo Bucardo bucardo burrick checksum checksums commitratio contrib criticals cronjob Cwd datadir datallowconn dbhost dbname dbpass dbport dbstats dbuser del dir dylan EB emma exabytes excludeuser ExclusiveLock executables faceoff filenames flagg franklin fsm garrett greg grimm GSM hitratio idxblkshit idxblksread idxscan idxtupfetch idxtupread includeuser Ioannis klatch lancre localhost logfile login lsfunc mallory maxwait MERCHANTABILITY minvalue morpork MRTG mrtg Mullane NAGIOS Nagios nagios nextval nofuncbody noidle nols noname noobjectnames noperm noperms noposition noschema ok oskar pageslots parens perflimit petabytes pgAgent pgBouncer pgbouncer pgbouncer's pgpass plasmid plugin pluto postgres PostgreSQL psql queryname quirm refactoring ret robert Sabino salesrep sami schemas scott Seklecki seqscan seqtupread showperf skipcycled slon Slony slony Slony's snazzo speedtest SQL stderr symlink symlinked symlinks syslog tablespace tablespaces Tambouras tardis timesync tuples unlinked upd uptime USERNAME usernames WAL watson wget wilkins xlog zettabytes +[-Bangs::ProhibitDebuggingModules] [-Bangs::ProhibitFlagComments] [-Bangs::ProhibitNumberedNames] [-Bangs::ProhibitVagueNames] @@ -13,7 +14,9 @@ stop_words = Mullane Nagios Slony Slony's nols salesrep psql dbname postgres USE [-BuiltinFunctions::ProhibitBooleanGrep] [-BuiltinFunctions::ProhibitComplexMappings] [-BuiltinFunctions::ProhibitReverseSortBlock] +[-BuiltinFunctions::ProhibitUselessTopic] +[-CodeLayout::ProhibitHashBarewords] [-CodeLayout::ProhibitParensWithBuiltins] [-CodeLayout::ProhibitQuotedWordLists] [-CodeLayout::RequireASCII] @@ -65,6 +68,7 @@ stop_words = Mullane Nagios Slony Slony's nols salesrep psql dbname postgres USE [-RegularExpressions::ProhibitComplexRegexes] [-RegularExpressions::ProhibitEnumeratedClasses] [-RegularExpressions::ProhibitEscapedMetacharacters] +[-RegularExpressions::ProhibitUnusedCapture] [-RegularExpressions::ProhibitUnusualDelimiters] [-RegularExpressions::RequireDotMatchAnything] [-RegularExpressions::RequireExtendedFormatting] @@ -76,6 +80,9 @@ stop_words = Mullane Nagios Slony Slony's nols salesrep psql dbname postgres USE [-Subroutines::ProhibitNestedSubs] [-Subroutines::RequireFinalReturn] + +[-TestingAndDebugging::ProhibitNoWarnings] + [-Tics::ProhibitLongLines] [-ValuesAndExpressions::ProhibitAccessOfPrivateData] @@ -86,6 +93,7 @@ stop_words = Mullane Nagios Slony Slony's nols salesrep psql dbname postgres USE [-ValuesAndExpressions::ProhibitMixedBooleanOperators] [-ValuesAndExpressions::ProhibitNoisyQuotes] [-ValuesAndExpressions::ProhibitVersionStrings] +[-ValuesAndExpressions::RequireConstantOnLeftSideOfEquality] [-ValuesAndExpressions::RequireNumberSeparators] [-ValuesAndExpressions::RequireNumericVersion] [-ValuesAndExpressions::RestrictLongStrings] diff --git a/set_translations.pl b/set_translations.pl index 1b1db000..02a810b1 100755 --- a/set_translations.pl +++ b/set_translations.pl @@ -4,11 +4,10 @@ ## This only needs to be run by developers of check_postgres, and only rarely ## Usage: $0 --pgsrc=path to top of Postgres source tree ## -## Greg Sabino Mullane -## End Point Corporation http://www.endpoint.com/ +## Greg Sabino Mullane ## BSD licensed -use 5.006001; +use 5.10.0; use strict; use warnings; use utf8; @@ -69,7 +68,7 @@ } ## A message - if ($line =~ /^(\s*)'([\w\-]+)'\s+=> (qq?)\{(.+?)}[,.](.*)/) { + if ($line =~ /^(\s*)'([\w\-]+)'\s*=> (qq?)\{(.+?)}[,.](.*)/) { my ($space,$msg,$quote,$value,$comment) = (length $1 ? 1 : 0, $2, $3, $4, $5); $msg{$lang}{$msg} = [$space,$value,$quote,$comment]; if ($lang eq 'en' and $msg =~ /\-po\d*$/) { @@ -90,7 +89,8 @@ for my $tr (sort keys %trans) { my $val = $trans{$tr}; if ($mm =~ /^$val/) { - $nn =~ s/(.+?)\s*\%.*/$1/; + $nn =~ s/(.+?)\s*\%.*/$1/; # remove everything from % on + print "$tr/$ll: $val -> $nn\n"; length $nn and $msg{$ll}{$tr} = [1,$nn,'q','']; } } @@ -106,24 +106,27 @@ last if $line =~ /^our \%msg/; } +my %all_langs = map { $_ => 1} (keys %msg, keys %po); + ## Add in the translated sections, with new info as needed for my $m (sort { ## English goes first, as the base class return -1 if $a eq 'en'; return 1 if $b eq 'en'; - ## French goes next, as the next-most-completed language + ## Then the fully translated languages + return -1 if $a eq 'es'; return 1 if $b eq 'es'; return -1 if $a eq 'fr'; return 1 if $b eq 'fr'; ## Everything else is alphabetical return $a cmp $b -} keys %po) { +} keys %all_langs) { print {$fh} qq!'$m' => {\n!; - my $size = 1; + my $size = 14; # length of "checkpoint-po" + 1 for my $msg (keys %{$msg{$m}}) { $size = length($msg) if length($msg) > $size; } for my $mm (sort keys %{$msg{$m}}) { - printf {$fh} "%s%-*s => %s{%s},%s\n", - $msg{$m}{$mm}->[0] ? "\t" : '', + printf {$fh} "%s%-*s=> %s{%s},%s\n", + $msg{$m}{$mm}->[0] ? " " : '', 2+$size, qq{'$mm'}, $msg{$m}{$mm}->[2], @@ -183,7 +186,7 @@ sub process_po_files { $isid = 0; } elsif (/^"(.*)"/) { - $isid ? ($id .= $1) : ($po{$lang}{$id} .= $1); + $isid ? ($id .= $1) : ($panda->{$lang}{$id} .= $1); } } close $fh or warn qq{Could not close "$pofile" $!\n}; diff --git a/t/00_basic.t b/t/00_basic.t index 7269ac00..fc0597e5 100644 --- a/t/00_basic.t +++ b/t/00_basic.t @@ -2,7 +2,7 @@ ## Simply test that the script compiles and gives a valid version -use 5.006; +use 5.10.0; use strict; use warnings; use Test::More tests => 2; diff --git a/t/00_release.t b/t/00_release.t index 36604b70..ca785c2c 100644 --- a/t/00_release.t +++ b/t/00_release.t @@ -4,7 +4,7 @@ ## 1. Make sure the version number is consistent in all places ## 2. Make sure we have a valid tag for this release -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -20,8 +20,8 @@ my $file = 'MANIFEST'; my @mfiles; open my $mfh, '<', $file or die qq{Could not open "$file": $!\n}; while (<$mfh>) { - next if /^#/; - push @mfiles => $1 if /(\S.+)/o; + next if /^#/; + push @mfiles => $1 if /(\S.+)/o; } close $mfh or warn qq{Could not close "$file": $!\n}; @@ -108,43 +108,43 @@ else { ## Make sure all files in the MANIFEST are "clean": no tabs, no unusual characters for my $mfile (@mfiles) { - file_is_clean($mfile); + file_is_clean($mfile); } sub file_is_clean { - my $file = shift or die; ## no critic (ProhibitReusedNames) - - if (!open $fh, '<', $file) { - fail qq{Could not open "$file": $!\n}; - return; - } - $good = 1; - my $inside_copy = 0; - while (<$fh>) { - if (/^COPY .+ FROM stdin/i) { - $inside_copy = 1; - } - if (/^\\./ and $inside_copy) { - $inside_copy = 0; - } - if (/\t/ and $file ne 'Makefile.PL' and $file !~ /\.html$/ and ! $inside_copy) { - diag "Found a tab at line $. of $file\n"; - $good = 0; - } - if (! /^[\S ]*/) { - diag "Invalid character at line $. of $file: $_\n"; - $good = 0; die; - } - } - close $fh or warn qq{Could not close "$file": $!\n}; - - if ($good) { - pass "The $file file has no tabs or unusual characters"; - } - else { - fail "The $file file did not pass inspection!"; - } + my $file = shift or die; ## no critic (ProhibitReusedNames) + + if (!open $fh, '<', $file) { + fail qq{Could not open "$file": $!\n}; + return; + } + $good = 1; + my $inside_copy = 0; + while (<$fh>) { + if (/^COPY .+ FROM stdin/i) { + $inside_copy = 1; + } + if (/^\\./ and $inside_copy) { + $inside_copy = 0; + } + if (/\t/ and $file ne 'Makefile.PL' and $file !~ /\.html$/ and ! $inside_copy) { + diag "Found a tab at line $. of $file\n"; + $good = 0; + } + if (! /^[\S ]*/) { + diag "Invalid character at line $. of $file: $_\n"; + $good = 0; die; + } + } + close $fh or warn qq{Could not close "$file": $!\n}; + + if ($good) { + pass "The $file file has no tabs or unusual characters"; + } + else { + fail "The $file file did not pass inspection!"; + } } diff --git a/t/00_signature.t b/t/00_signature.t index fc07ec78..288ae9bf 100644 --- a/t/00_signature.t +++ b/t/00_signature.t @@ -2,7 +2,7 @@ ## Test that our PGP signature file is valid -use 5.006; +use 5.10.0; use strict; use warnings; use Test::More; diff --git a/t/00_test_tester.t b/t/00_test_tester.t index 7f6f55f5..47f38168 100644 --- a/t/00_test_tester.t +++ b/t/00_test_tester.t @@ -2,7 +2,7 @@ ## Make sure we have tests for all actions -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -33,8 +33,10 @@ for my $line (split /\n/ => $info) { my $ok = 1; for my $act (sort keys %action) { ## Special known exceptions - next if $act eq 'table_size' or $act eq 'index_size'; - next if $act eq 'last_autoanalyze' or $act eq 'last_autovacuum'; + next if grep { $act eq $_ } qw( + index_size table_size indexes_size total_relation_size + last_autoanalyze last_autovacuum + ); my $file = "t/02_$act.t"; if (! -e $file) { diff --git a/t/01_validate_range.t b/t/01_validate_range.t index 849a0b30..69c52f6e 100644 --- a/t/01_validate_range.t +++ b/t/01_validate_range.t @@ -2,7 +2,7 @@ ## Test the "validate_range" function -use 5.006; +use 5.10.0; use strict; use warnings; use Test::More tests => 144; diff --git a/t/02_autovac_freeze.t b/t/02_autovac_freeze.t index 26cfbc08..e206432e 100644 --- a/t/02_autovac_freeze.t +++ b/t/02_autovac_freeze.t @@ -2,7 +2,7 @@ ## Test the "autovac_freeze" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_backends.t b/t/02_backends.t index f369756f..e11e20a9 100644 --- a/t/02_backends.t +++ b/t/02_backends.t @@ -2,7 +2,7 @@ ## Test the "backends" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -25,9 +25,11 @@ my $label = 'POSTGRES_BACKENDS'; my $ver = $dbh->{pg_server_version}; my $goodver = $ver >= 80200 ? 1 : 0; my $pg92 = $ver >= 90200 ? 1 : 0; +my $pg10 = $ver >= 100000 ? 1 : 0; ## Check current number of connections: should be 1 (for recent versions of PG) $SQL = 'SELECT count(*) FROM pg_stat_activity'; +$SQL .= q{ WHERE backend_type = 'client backend'} if $pg10; $count = $dbh->selectall_arrayref($SQL)->[0][0]; $t=q{Current number of backends is one (ourselves)}; @@ -183,4 +185,8 @@ for my $num (1..8) { $t=qq{$S returns critical when too many clients to even connect}; like ($cp->run('-w -10'), qr{^$label CRITICAL: .+too many connections}, $t); +for my $num (1..8) { + $dbh{$num}->disconnect(); +} + exit; diff --git a/t/02_bloat.t b/t/02_bloat.t index cbb9c010..d0e19602 100644 --- a/t/02_bloat.t +++ b/t/02_bloat.t @@ -2,7 +2,7 @@ ## Test the "bloat" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_checkpoint.t b/t/02_checkpoint.t index dabee60d..4420eb52 100644 --- a/t/02_checkpoint.t +++ b/t/02_checkpoint.t @@ -2,7 +2,7 @@ ## Test the "checkpoint" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -53,7 +53,7 @@ SKIP: { if ($result =~ /Date::Parse/) { - skip 'Cannot test checkpoint action unless Date::Parse module is installed', 6; + skip 'Cannot test checkpoint action unless Date::Parse module is installed', 7; } like ($cp->run(qq{-w 30 --datadir="$host"}), qr{^$label OK}, $t); diff --git a/t/02_cluster_id.t b/t/02_cluster_id.t index 6cab394c..c9a5b91c 100644 --- a/t/02_cluster_id.t +++ b/t/02_cluster_id.t @@ -2,7 +2,7 @@ ## Test the "checkpoint" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_commitratio.t b/t/02_commitratio.t index 018d33a7..4ccdc698 100644 --- a/t/02_commitratio.t +++ b/t/02_commitratio.t @@ -2,7 +2,7 @@ ## Test the "commitratio" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -80,6 +80,6 @@ $t=qq{$S returned correct performance data with include}; like ($cp->run('-w 5% --include=postgres'), qr{ \| time=\d\.\d\ds postgres=\d+}, $t); $t=qq{$S with includeuser option returns nothing}; -like ($cp->run('--includeuser postgres --includeuser mycatbeda -w 10%'), qr{No matching entries found due to user exclusion}, $t); +like ($cp->run('--includeuser mycatbeda -w 10%'), qr{No matching entries found due to user exclusion}, $t); exit; diff --git a/t/02_connection.t b/t/02_connection.t index d6fd219e..68108395 100644 --- a/t/02_connection.t +++ b/t/02_connection.t @@ -2,11 +2,11 @@ ## Test the "connection" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; -use Test::More tests => 12; +use Test::More tests => 14; use lib 't','.'; use CP_Testing; @@ -52,7 +52,14 @@ is ($cp->run('--output=MRTG'), qq{1\n0\n\n\n}, $t); $cp->fake_version('ABC'); $t=qq{$S fails if there's a fake version function}; -like ($cp->run(), qr{^$label UNKNOWN:}, $t); +like ($cp->run(), qr{^$label UNKNOWN:.*Invalid query}, $t); + +$cp->fake_version_timeout(); +$t=qq{$S fails on timeout}; +like ($cp->run('--timeout 1'), qr{^$label CRITICAL:.*(Timed out|statement timeout)}, $t); $cp->reset_path(); +$t=qq{$S fails on nonexisting socket}; +like ($cp->run('--port=1023'), qr{^$label CRITICAL:.*(could not connect to server|connection to server.*failed)}, $t); + exit; diff --git a/t/02_custom_query.t b/t/02_custom_query.t index a6b2d4fb..d4a8f006 100644 --- a/t/02_custom_query.t +++ b/t/02_custom_query.t @@ -2,7 +2,7 @@ ## Test the "custom_query" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_database_size.t b/t/02_database_size.t index 87ed11ce..a4e11174 100644 --- a/t/02_database_size.t +++ b/t/02_database_size.t @@ -2,7 +2,7 @@ ## Test the "database_size" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -114,6 +114,6 @@ $t=qq{$S returned correct performance data with include}; like ($cp->run('-w 5g --include=postgres'), qr{ \| time=\d\.\d\ds postgres=\d+}, $t); $t=qq{$S with includeuser option returns nothing}; -like ($cp->run('--includeuser postgres --includeuser mycatbeda -w 10g'), qr{No matching entries found due to user exclusion}, $t); +like ($cp->run('--includeuser mycatbeda -w 10g'), qr{No matching entries found due to user exclusion}, $t); exit; diff --git a/t/02_dbstats.t b/t/02_dbstats.t index b2f2e023..4dfa48f0 100644 --- a/t/02_dbstats.t +++ b/t/02_dbstats.t @@ -2,7 +2,7 @@ ## Test the "dbstats" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_disabled_triggers.t b/t/02_disabled_triggers.t index aac8f79c..f005b603 100644 --- a/t/02_disabled_triggers.t +++ b/t/02_disabled_triggers.t @@ -2,7 +2,7 @@ ## Test the "disabled_triggers" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_disk_space.t b/t/02_disk_space.t index 74e3a6ee..c92cc124 100644 --- a/t/02_disk_space.t +++ b/t/02_disk_space.t @@ -2,7 +2,7 @@ ## Test the "disk_space" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -39,7 +39,7 @@ $t = qq{$S identifies host}; like ($result, qr{host:$host}, $t); $t = qq{$S reports file system}; -like ($result, qr{FS /.*? mounted on /.*? is using }, $t); +like ($result, qr{FS .* mounted on /.*? is using }, $t); # in some build environments, the filesystem is reported as "-" $t = qq{$S reports usage}; like ($result, qr{ is using \d*\.\d+ [A-Z]B of \d*\.\d+ [A-Z]B}, $t); @@ -54,6 +54,6 @@ $t = qq{$S flags insufficient space}; like ($cp->run('-w "999z or 1%"'), qr{$label WARNING:}, $t); $t = qq{$S reports MRTG output}; -like ($cp->run('--output=mrtg'), qr{\A\d+\n0\n\n/.*\n}, $t); +like ($cp->run('--output=mrtg'), qr{\A\d+\n0\n\n.*\n}, $t); exit; diff --git a/t/02_fsm_pages.t b/t/02_fsm_pages.t index ce3b7161..68619716 100644 --- a/t/02_fsm_pages.t +++ b/t/02_fsm_pages.t @@ -2,7 +2,7 @@ ## Test the "fsm_pages" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_fsm_relations.t b/t/02_fsm_relations.t index 0eedc24c..0ed32f41 100644 --- a/t/02_fsm_relations.t +++ b/t/02_fsm_relations.t @@ -2,7 +2,7 @@ ## Test the "fsm_relations" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_hitratio.t b/t/02_hitratio.t index 15a73867..1022db9e 100644 --- a/t/02_hitratio.t +++ b/t/02_hitratio.t @@ -2,7 +2,7 @@ ## Test the "hitratio" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -80,6 +80,6 @@ $t=qq{$S returned correct performance data with include}; like ($cp->run('-w 5% --include=postgres'), qr{ \| time=\d\.\d\ds postgres=\d+}, $t); $t=qq{$S with includeuser option returns nothing}; -like ($cp->run('--includeuser postgres --includeuser mycatbeda -w 10%'), qr{No matching entries found due to user exclusion}, $t); +like ($cp->run('--includeuser mycatbeda -w 10%'), qr{No matching entries found due to user exclusion}, $t); exit; diff --git a/t/02_last_analyze.t b/t/02_last_analyze.t index a3c06f5e..899d0fa0 100644 --- a/t/02_last_analyze.t +++ b/t/02_last_analyze.t @@ -2,7 +2,7 @@ ## Test the "last_analyze" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_last_vacuum.t b/t/02_last_vacuum.t index 806c07a0..87a62450 100644 --- a/t/02_last_vacuum.t +++ b/t/02_last_vacuum.t @@ -2,7 +2,7 @@ ## Test the "last_vacuum" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_listener.t b/t/02_listener.t index f6dc154c..6671dc4d 100644 --- a/t/02_listener.t +++ b/t/02_listener.t @@ -2,7 +2,7 @@ ## Test the "listener" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_locks.t b/t/02_locks.t index 0765f75b..d03ddb76 100644 --- a/t/02_locks.t +++ b/t/02_locks.t @@ -2,7 +2,7 @@ ## Test the "locks" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_logfile.t b/t/02_logfile.t index 15de979d..0bfb381c 100644 --- a/t/02_logfile.t +++ b/t/02_logfile.t @@ -3,7 +3,7 @@ ## Test the "logfile" action ## this does not test $S for syslog or stderr output -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_new_version_bc.t b/t/02_new_version_bc.t index c7b8c7a8..0618bf9b 100644 --- a/t/02_new_version_bc.t +++ b/t/02_new_version_bc.t @@ -2,7 +2,7 @@ ## Test the "new_version_bc" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_new_version_box.t b/t/02_new_version_box.t index 3f6c0818..c4bd42bd 100644 --- a/t/02_new_version_box.t +++ b/t/02_new_version_box.t @@ -2,7 +2,7 @@ ## Test the "new_version_box" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_new_version_cp.t b/t/02_new_version_cp.t index 3d00afc4..3da86f75 100644 --- a/t/02_new_version_cp.t +++ b/t/02_new_version_cp.t @@ -2,7 +2,7 @@ ## Test the "new_version_cp" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -41,7 +41,7 @@ like ($cp->run(''), qr{$label OK: Version $current_version is the latest for ch $t=qq{$S returns critical for mismatched revision}; my $warncrit; -if ($crev==0) { +if (0 == $crev) { $crev = 99; $cmaj--; $warncrit = 'WARNING'; diff --git a/t/02_new_version_pg.t b/t/02_new_version_pg.t index dd4345bc..1220237f 100644 --- a/t/02_new_version_pg.t +++ b/t/02_new_version_pg.t @@ -2,7 +2,7 @@ ## Test the "new_version_pg" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_new_version_tnm.t b/t/02_new_version_tnm.t index bf106054..06c173fe 100644 --- a/t/02_new_version_tnm.t +++ b/t/02_new_version_tnm.t @@ -2,7 +2,7 @@ ## Test the "new_version_tnm" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_pgagent_jobs.t b/t/02_pgagent_jobs.t index 26f51826..bb13c44c 100644 --- a/t/02_pgagent_jobs.t +++ b/t/02_pgagent_jobs.t @@ -2,7 +2,7 @@ ## Test the "pgagent_jobs" action -use 5.006; +use 5.10.0; use strict; use warnings; use Test::More tests => 48; @@ -27,8 +27,6 @@ $dbh->{AutoCommit} = 1; $dbh->do('DROP SCHEMA IF EXISTS pgagent CASCADE'); $dbh->do(q{ - SET client_min_messages TO warning; - CREATE SCHEMA pgagent; CREATE TABLE pgagent.pga_job ( @@ -56,7 +54,6 @@ $dbh->do(q{ jslresult int4 NULL, jsloutput text ); - RESET client_min_messages; }); END { $dbh->do('DROP SCHEMA IF EXISTS pgagent CASCADE'); } @@ -124,14 +121,14 @@ like $cp->run('-c=2h'), qr{^$label OK: DB "postgres"}, "$S -c=2h returns ok with failed job before our time"; like $cp->run('-c=6h'), - qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!}, + qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF! }, "$S -c=6h returns critical with failed job within our time"; like $cp->run('-w=2h'), qr{^$label OK: DB "postgres"}, "$S -w=2h returns ok with failed job before our time"; like $cp->run('-w=6h'), - qr{^$label WARNING: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!}, + qr{^$label WARNING: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF! }, "$S -w=6h returns warninf with failed job within our time"; like $cp->run('-w=2h'), qr{^$label OK: DB "postgres"}, @@ -141,15 +138,15 @@ like $cp->run('-w=4h -c=2h'), qr{^$label OK: DB "postgres"}, "$S -w=4h =c=2h returns ok with failed job before our time"; like $cp->run('-w=5h -c=2h'), - qr{^$label WARNING: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!}, + qr{^$label WARNING: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF! }, "$S -w=5h =c=2h returns warning with failed job within our time"; like $cp->run('-w=2h -c=5h'), - qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!}, + qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF! }, "$S -w=2h =c=5h returns critical with failed job within our time"; like $cp->run('-w=5h -c=5h'), - qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!}, + qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF! }, "$S -w=5h =c=5h returns critical with failed job within our time"; # Make a second job fail, back 30 hours. @@ -165,14 +162,14 @@ like $cp->run('-c=2h'), qr{^$label OK: DB "postgres"}, "$S -c=2h returns ok with failed job before our time"; like $cp->run('-c=6h'), - qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!}, + qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF! }, "$S -c=6h returns critical with failed job within our time"; like $cp->run('-w=2h'), qr{^$label OK: DB "postgres"}, "$S -w=2h returns ok with failed job before our time"; like $cp->run('-w=6h'), - qr{^$label WARNING: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!}, + qr{^$label WARNING: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF! }, "$S -w=6h returns warninf with failed job within our time"; like $cp->run('-w=2h'), qr{^$label OK: DB "postgres"}, @@ -182,42 +179,42 @@ like $cp->run('-w=4h -c=2h'), qr{^$label OK: DB "postgres"}, "$S -w=4h =c=2h returns ok with failed job before our time"; like $cp->run('-w=5h -c=2h'), - qr{^$label WARNING: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!}, + qr{^$label WARNING: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF! }, "$S -w=5h =c=2h returns warning with failed job within our time"; like $cp->run('-w=2h -c=5h'), - qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!}, + qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF! }, "$S -w=2h =c=5h returns critical with failed job within our time"; like $cp->run('-w=5h -c=5h'), - qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!}, + qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF! }, "$S -w=5h -c=5h returns critical with failed job within our time"; # Go back further in time! like $cp->run('-w=30h -c=2h'), - qr{^$label WARNING: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!}, - "$S -w=30h -c=5h returns warning for older failed job"; + qr{^$label WARNING: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!; 64 Restore/analyze: OMGWTFLOL! }, + "$S -w=30h -c=2h returns warning for older failed job"; like $cp->run('-w=30h -c=6h'), - qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!; 64 Restore/analyze: OMGWTFLOL!}, + qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!; 64 Restore/analyze: OMGWTFLOL! }, "$S -w=30h -c=6h returns critical with both jobs, more recent critical"; like $cp->run('-c=30h'), - qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!; 64 Restore/analyze: OMGWTFLOL!}, + qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!; 64 Restore/analyze: OMGWTFLOL! }, "$S -c=30h returns critical with both failed jobs"; like $cp->run('-w=30h'), - qr{^$label WARNING: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!; 64 Restore/analyze: OMGWTFLOL!}, + qr{^$label WARNING: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!; 64 Restore/analyze: OMGWTFLOL! }, "$S -w=30h returns critical with both failed jobs"; # Try with critical recent and warning longer ago. like $cp->run('-w=30h -c=6h'), - qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!; 64 Restore/analyze: OMGWTFLOL!}, + qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!; 64 Restore/analyze: OMGWTFLOL! }, "$S -w=30h -c=6h returns critical with both failed jobs"; # Try with warning recent and critical longer ago. like $cp->run('-c=30h -w=6h'), - qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!; 64 Restore/analyze: OMGWTFLOL!}, + qr{^$label CRITICAL: DB "postgres" [()][^)]+[)] 255 Restore/analyze: WTF!; 64 Restore/analyze: OMGWTFLOL! }, "$S -c=30h -w=6h returns critical with both failed jobs"; # Undo the more recent failure. diff --git a/t/02_pgbouncer_checksum.t b/t/02_pgbouncer_checksum.t index 030ca94e..4a8ae81d 100644 --- a/t/02_pgbouncer_checksum.t +++ b/t/02_pgbouncer_checksum.t @@ -2,7 +2,7 @@ ## Test the "pgbouncer_checksum" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_prepared_txns.t b/t/02_prepared_txns.t index d588a28c..e16493d9 100644 --- a/t/02_prepared_txns.t +++ b/t/02_prepared_txns.t @@ -2,7 +2,7 @@ ## Test the "prepare_txns" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_query_runtime.t b/t/02_query_runtime.t index 4571428b..6d02ab3c 100644 --- a/t/02_query_runtime.t +++ b/t/02_query_runtime.t @@ -2,7 +2,7 @@ ## Test the "query_runtime" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_query_time.t b/t/02_query_time.t index ebd7ab6a..282a40f8 100644 --- a/t/02_query_time.t +++ b/t/02_query_time.t @@ -2,7 +2,7 @@ ## Test the "query_time" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -59,7 +59,7 @@ if ($ver < 80200) { } my $child = fork(); -if ($child == 0) { +if (0 == $child) { my $kiddbh = $cp->test_database_handle(); $cp->database_sleep($kiddbh, 3); $kiddbh->rollback(); diff --git a/t/02_relation_size.t b/t/02_relation_size.t index f13d914a..8c725747 100644 --- a/t/02_relation_size.t +++ b/t/02_relation_size.t @@ -2,11 +2,11 @@ ## Test the "relation_size" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; -use Test::More tests => 23; +use Test::More tests => 15 + 4 * 4; use lib 't','.'; use CP_Testing; @@ -97,11 +97,13 @@ $t = qq{$S includes indexes}; $dbh->do(qq{CREATE INDEX "${testtbl}_index" ON "$testtbl" (a)}); $dbh->commit; like ($cp->run(qq{-w 1 --includeuser=$user --include=${testtbl}_index}), - qr{$label WARNING.*largest relation is index "${testtbl}_index": \d+ kB}, $t); + qr{$label WARNING.*largest relation is index "\w+.${testtbl}_index": \d+ kB}, $t); -#### Switch gears, and test the related functions "check_table_size" and "check_index_size". +#### Switch gears, and test the other size actions -for $S (qw(table_size index_size)) { +for $S (qw(index_size table_size indexes_size total_relation_size)) { +SKIP: { + skip "$S not supported before 9.0", 4 if ($S eq 'indexes_size' and $ver < 90000); $result = $cp->run($S, q{-w 1}); $label = "POSTGRES_\U$S"; @@ -123,11 +125,12 @@ for $S (qw(table_size index_size)) { ($S ne 'table_size' ? '_table' : '_index'); - my $message = 'largest ' . ($S eq 'table_size' - ? 'table' - : 'index'); + my $message = 'largest ' . ($S =~ /index/ + ? 'index' + : 'table'); like ($cp->run($S, qq{-w 1 --includeuser=$user $include $exclude}), - qr|$label.*$message|, $t) + qr|$label.*$message|, $t); +} } exit; diff --git a/t/02_replicate_row.t b/t/02_replicate_row.t index 7df6354d..02a652f9 100644 --- a/t/02_replicate_row.t +++ b/t/02_replicate_row.t @@ -2,12 +2,11 @@ ## Test the "replicate_row" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; use Test::More tests => 19; -use Time::HiRes qw(usleep); use lib 't','.'; use CP_Testing; @@ -120,6 +119,7 @@ else { $dbh2->commit(); exit; } +wait; $t=qq{$S works when rows match, reports proper delay}; $dbh->commit(); @@ -128,29 +128,31 @@ if (fork) { like ($result, qr{^$label OK:.+Row was replicated}, $t); $result =~ /time=(\d+)/ or die 'No time?'; my $time = $1; - cmp_ok ($time, '>=', 3, $t); + cmp_ok ($time, '>=', 1, $t); } else { - sleep 3; + sleep 2; $SQL = q{UPDATE reptest SET foo = 'yang' WHERE id = 1}; $dbh2->do($SQL); $dbh2->commit(); exit; } +wait; $t=qq{$S works when rows match, with MRTG output}; $dbh->commit(); if (fork) { - is ($cp->run('DB2replicate-row', '-c 20 --output=MRTG -repinfo=reptest,id,1,foo,yin,yang'), - qq{1\n0\n\n\n}, $t); + like ($cp->run('DB2replicate-row', '-c 20 --output=MRTG -repinfo=reptest,id,1,foo,yin,yang'), + qr{^[1-5]\n0\n\n\n}, $t); } else { - usleep 500_000; # 0.5s + sleep 2; $SQL = q{UPDATE reptest SET foo = 'yin' WHERE id = 1}; $dbh2->do($SQL); $dbh2->commit(); exit; } +wait; $t=qq{$S works when rows match, with simple output}; $dbh->commit(); @@ -158,15 +160,16 @@ if (fork) { $result = $cp->run('DB2replicate-row', '-c 20 --output=simple -repinfo=reptest,id,1,foo,yin,yang'); $result =~ /^(\d+)/ or die 'No time?'; my $time = $1; - cmp_ok ($time, '>=', 3, $t); + cmp_ok ($time, '>=', 1, $t); } else { - sleep 3; + sleep 2; $SQL = q{UPDATE reptest SET foo = 'yang' WHERE id = 1}; $dbh2->do($SQL); $dbh2->commit(); exit; } +wait; $dbh2->disconnect(); diff --git a/t/02_replication_slots.t b/t/02_replication_slots.t new file mode 100644 index 00000000..c63c8ec7 --- /dev/null +++ b/t/02_replication_slots.t @@ -0,0 +1,135 @@ +#!perl + +## Test the "replication_slots" action + +use 5.10.0; +use strict; +use warnings; +use Data::Dumper; +use Test::More; +use lib 't','.'; +use CP_Testing; + +use vars qw/$dbh $result $t $port $host $dbname/; + +my $cp = CP_Testing->new( {default_action => 'replication_slots'} ); + +$dbh = $cp->test_database_handle(); +$dbh->{AutoCommit} = 1; +$port = $cp->get_port(); +$host = $cp->get_host(); +$dbname = $cp->get_dbname; + +diag "Connected as $port:$host:$dbname\n"; + +my $S = q{Action 'replication_slots'}; +my $label = 'POSTGRES_REPLICATION_SLOTS'; + +my $ver = $dbh->{pg_server_version}; +if ($ver < 90400) { + SKIP: { + skip 'replication slots not present before 9.4', 1; + } + done_testing(); + exit 0; +} + +$t = qq{$S self-identifies correctly}; +$result = $cp->run(q{-w 0}); +like ($result, qr{^$label}, $t); + +$t = qq{$S identifies host}; +like ($result, qr{host:$host}, $t); + +$t = qq{$S reports no replication slots}; +like ($result, qr{No replication slots found}, $t); + +$t = qq{$S accepts valid -w input}; +for my $arg ( + '1 MB', + '1 GB', + ) { + like ($cp->run(qq{-w "$arg"}), qr{^$label}, "$t ($arg)"); +} + +$t = qq{$S rejects invalid -w input}; +for my $arg ( + '-1 MB', + 'abc' + ) { + like ($cp->run(qq{-w "$arg"}), qr{^ERROR: Invalid size}, "$t ($arg)"); +} + +$dbh->do (q{SELECT * FROM pg_create_physical_replication_slot('cp_testing_slot')}); + +$t = qq{$S reports physical replication slots}; +$result = $cp->run(q{-w 0}); +like ($result, qr{cp_testing_slot.*physical}, $t); + +$t=qq{$S reports ok on physical replication slots when warning level is specified and not exceeded}; +$result = $cp->run(q{-w 1MB}); +like ($result, qr{^$label OK:}, $t); + +$t=qq{$S reports ok on physical replication slots when critical level is specified and not exceeded}; +$result = $cp->run(q{-c 1MB}); +like ($result, qr{^$label OK:}, $t); + +$dbh->do (q{SELECT pg_drop_replication_slot('cp_testing_slot')}); + +SKIP: { + + skip q{Waiting for test_decoding plugin}, 10; + +# To do more tests on physical slots we'd actually have to kick off some activity by performing a connection to them (.. use pg_receivexlog or similar??) + +$dbh->do (q{SELECT * FROM pg_create_logical_replication_slot('cp_testing_slot', 'test_decoding')}); + +$t = qq{$S reports logical replication slots}; +$result = $cp->run(q{-w 0}); +like ($result, qr{cp_testing_slot.*logical}, $t); + +$t=qq{$S reports ok on logical replication slots when warning level is specified and not exceeded}; +$result = $cp->run(q{-w 1MB}); +like ($result, qr{^$label OK:}, $t); + +$t=qq{$S reports ok on logical replication slots when critical level is specified and not exceeded}; +$result = $cp->run(q{-c 1MB}); +like ($result, qr{^$label OK:}, $t); + +$dbh->do (q{CREATE TABLE cp_testing_table (a text); INSERT INTO cp_testing_table SELECT a || repeat('A',1024) FROM generate_series(1,1024) a; DROP TABLE cp_testing_table;}); + +$t=qq{$S reports warning on logical replication slots when warning level is specified and is exceeded}; +$result = $cp->run(q{-w 1MB}); +like ($result, qr{^$label WARNING:}, $t); + +$t=qq{$S reports critical on logical replication slots when critical level is specified and is exceeded}; +$result = $cp->run(q{-c 1MB}); +like ($result, qr{^$label CRITICAL:}, $t); + +$t=qq{$S works when include has valid replication slot}; +$result = $cp->run(q{-w 1MB --include=cp_testing_slot}); +like ($result, qr{^$label WARNING:.*cp_testing_slot}, $t); + +$t=qq{$S works when include matches no replication slots}; +$result = $cp->run(q{-w 1MB --include=foobar}); +like ($result, qr{^$label UNKNOWN:.*No matching replication slots}, $t); + +$t=qq{$S returnes correct performance data with include}; +$result = $cp->run(q{-w 1MB --include=cp_testing_slot}); +like ($result, qr{ \| time=\d\.\d\ds cp_testing_slot=\d+}, $t); + +$t=qq{$S works when exclude excludes no replication slots}; +$result = $cp->run(q{-w 10MB --exclude=foobar}); +like ($result, qr{^$label OK:.*cp_testing_slot}, $t); + +$t=qq{$S works when exclude excludes all replication slots}; +$result = $cp->run(q{-w 10MB --exclude=cp_testing_slot}); +like ($result, qr{^$label UNKNOWN:.*No matching replication slots}, $t); + +$dbh->do (q{SELECT pg_drop_replication_slot('cp_testing_slot')}); + +} + +done_testing(); + +exit; diff --git a/t/02_same_schema.t b/t/02_same_schema.t index ec70f5f6..478d3d83 100644 --- a/t/02_same_schema.t +++ b/t/02_same_schema.t @@ -2,11 +2,11 @@ ## Test the "same_schema" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; -use Test::More tests => 76; +use Test::More tests => 75; use lib 't','.'; use CP_Testing; @@ -18,6 +18,7 @@ my $cp3 = CP_Testing->new({ default_action => 'same_schema', dbnum => 3}); ## Setup all database handles, and create a testing user $dbh1 = $cp1->test_database_handle(); +my $ver = $dbh1->{pg_server_version}; $dbh1->{AutoCommit} = 1; eval { $dbh1->do(q{CREATE USER alternate_owner}, { RaiseError => 0, PrintError => 0 }); }; $dbh2 = $cp2->test_database_handle(); @@ -63,6 +64,7 @@ $t = qq{$S succeeds with two empty databases}; like ($cp1->run($connect2), qr{^$label OK}, $t); + sub drop_language { my ($name, $dbhx) = @_; @@ -80,8 +82,6 @@ sub drop_language { } ## end of drop_language -#goto TRIGGER; ## ZZZ - #/////////// Languages ## Because newer versions of Postgres already have plpgsql installed, @@ -97,15 +97,12 @@ like ($cp1->run($connect3), qr{^$label OK}, $t); $t = qq{$S reports language on 3 but not 1 and 2}; $dbh3->do(q{CREATE LANGUAGE plpgsql}); like ($cp1->run($connect3), - qr{^$label CRITICAL.*Items not matched: 1 .* + qr{^$label CRITICAL.* Language "plpgsql" does not exist on all databases: \s*Exists on:\s+3 -\s+Missing on:\s+1, 2\s*$}s, +\s+Missing on:\s+1, 2}s, $t); -$t = qq{$S does not report language differences if the 'nolanguage' filter is given}; -like ($cp1->run("$connect3 --filter=nolanguage"), qr{^$label OK}, $t); - $dbh1->do(q{CREATE LANGUAGE plpgsql}); $dbh2->do(q{CREATE LANGUAGE plpgsql}); @@ -119,7 +116,6 @@ drop_language('plpgsql', $dbh3); $t = qq{$S reports on language differences}; like ($cp1->run($connect3), qr{^$label OK}, $t); - #/////////// Users $t = qq{$S reports on user differences}; @@ -249,7 +245,6 @@ $dbh2->do(q{DROP SCHEMA schema_a}); $t = qq{$S reports on schema differences}; like ($cp1->run($connect3), qr{^$label OK}, $t); - #/////////// Tables TABLE: @@ -275,16 +270,22 @@ Table "public.berri" does not exist on all databases: \s*Missing on: 1, 3\s*$}s, $t); -$t = qq{$S reports table attribute differences}; -$dbh1->do(q{CREATE TABLE berri(bfd int) WITH OIDS}); -like ($cp1->run($connect2), - qr{^$label CRITICAL.*Items not matched: 1 .* +$t = qq{$S reports table attribute (WITH OIDS) differences}; +if ($ver < 120000) { + $dbh1->do(q{CREATE TABLE berri(bfd int) WITH OIDS}); + like ($cp1->run($connect2), + qr{^$label CRITICAL.*Items not matched: 1 .* Table "public.berri": \s*"relhasoids" is different: \s*Database 1: t \s*Database 2: f\s*$}s, - $t); -$dbh1->do(q{ALTER TABLE berri SET WITHOUT OIDS}); + $t); + $dbh1->do(q{ALTER TABLE berri SET WITHOUT OIDS}); +} +else { + $dbh1->do(q{CREATE TABLE berri(bfd int)}); + pass ('No need to test relhasoids on modern databases'); +} $t = qq{$S reports simple table acl differences}; $dbh1->do(q{GRANT SELECT ON TABLE berri TO alternate_owner}); @@ -340,6 +341,9 @@ Sequence "wakko.yakko" does not exist on all databases: $t = qq{$S reports sequence differences}; $dbh2->do(q{CREATE SEQUENCE wakko.yakko MINVALUE 10 MAXVALUE 100 INCREMENT BY 3}); +$dbh1->do(q{SELECT nextval('wakko.yakko')}); +$dbh2->do(q{SELECT nextval('wakko.yakko')}); + like ($cp1->run($connect2), qr{^$label CRITICAL.*Items not matched: 1 .* \s*Sequence "wakko.yakko": @@ -487,13 +491,18 @@ $dbh1->do($SQL); $dbh2->do($SQL); $dbh3->do($SQL); $SQL = 'CREATE TRIGGER tigger BEFORE INSERT ON piglet EXECUTE PROCEDURE bouncy()'; $dbh1->do($SQL); +SKIP: { + + skip 'Cannot test trigger differences on versions of Postgres older than 9.0', 3 + if $ver < 90000; + like ($cp1->run($connect2), qr{^$label CRITICAL.*Items not matched: 2 .* \s*Table "public.piglet": \s*"relhastriggers" is different: \s*Database 1: t \s*Database 2: f -\s*Trigger "public.tigger" does not exist on all databases: +\s*Trigger "public.piglet.tigger" does not exist on all databases: \s*Exists on: 1 \s*Missing on: 2\s*$}s, $t); @@ -509,7 +518,7 @@ $dbh2->do($SQL); like ($cp1->run($connect2), qr{^$label CRITICAL.*Items not matched: 1 .* -\s*Trigger "public.tigger": +\s*Trigger "public.piglet.tigger": \s*"procname" is different: \s*Database 1: bouncy \s*Database 2: trouncy\s*}s, @@ -525,10 +534,12 @@ $dbh1->do($SQL); ## We leave out the details as the exact values are version-dependent like ($cp1->run($connect2), qr{^$label CRITICAL.*Items not matched: 1 .* -\s*Trigger "public.tigger": +\s*Trigger "public.piglet.tigger": \s*"tgenabled" is different:}s, $t); +} + ## We have to also turn off table differences $t = qq{$S does not report trigger differences if the 'notrigger' filter is given}; like ($cp1->run("$connect3 --filter=notrigger,notable"), qr{^$label OK}, $t); @@ -559,19 +570,26 @@ like ($cp1->run($connect2), \s*"relchecks" is different: \s*Database 1: 1 \s*Database 2: 0 -\s*Constraint "public.iscandar" does not exist on all databases: +\s*Constraint "public.yamato.iscandar" does not exist on all databases: \s*Exists on: 1 \s*Missing on: 2\s*$}s, $t); $t = qq{$S reports constraint with different definitions}; $dbh2->do(q{ALTER TABLE yamato ADD CONSTRAINT iscandar CHECK(nova > 256)}); -like ($cp1->run($connect2), - qr{^$label CRITICAL.*Items not matched: 1 .* -\s*Constraint "public.iscandar": + +## Version 12 removed the pg_constraint.consrc column +my $extra = $ver >= 120000 ? '' : q{ \s*"consrc" is different: \s*Database 1: \(nova > 0\) -\s*Database 2: \(nova > 256\)\s*$}s, +\s*Database 2: \(nova > 256\)}; + +like ($cp1->run($connect2), + qr{^$label CRITICAL.*Items not matched: 1 .* +\s*Constraint "public.yamato.iscandar":$extra +\s*"constraintdef" is different: +\s*Database 1: CHECK \(\(nova > 0\)\) +\s*Database 2: CHECK \(\(nova > 256\)\)\s*$}s, $t); $t = qq{$S does not report constraint differences if the 'noconstraint' filter is given}; @@ -603,7 +621,7 @@ like ($cp1->run($connect2), \s*"relhasindex" is different: \s*Database 1: t \s*Database 2: f -\s*Index "public.valen" does not exist on all databases: +\s*Index "public.valen \(gkar\)" does not exist on all databases: \s*Exists on: 1 \s*Missing on: 2\s*$}s, $t); @@ -614,8 +632,8 @@ like ($cp1->run($connect2), qr{^$label CRITICAL.*Items not matched: 1 .* \s*Index "public.valen": \s*"indexdef" is different: -\s*Database 1: CREATE INDEX valen ON gkar USING btree \(garibaldi\) -\s*Database 2: CREATE UNIQUE INDEX valen ON gkar USING btree \(garibaldi\) +\s*Database 1: CREATE INDEX valen ON (?:public\.)?gkar USING btree \(garibaldi\) +\s*Database 2: CREATE UNIQUE INDEX valen ON (?:public\.)?gkar USING btree \(garibaldi\) \s*"indisunique" is different: \s*Database 1: f \s*Database 2: t\s*$}s, @@ -738,10 +756,3 @@ $dbh1->do('DROP USER user_d'); exit; -__DATA__ - - -FINAL: -Bump version high -show number key -good key for historical diff --git a/t/02_sequence.t b/t/02_sequence.t index d42187f6..050764f2 100644 --- a/t/02_sequence.t +++ b/t/02_sequence.t @@ -2,11 +2,11 @@ ## Test the "sequence" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; -use Test::More tests => 10; +use Test::More tests => 16; use lib 't','.'; use CP_Testing; @@ -43,8 +43,12 @@ if ($ver < 80100) { my $seqname = 'cp_test_sequence'; +my $testtbl = 'sequence_test'; $cp->drop_sequence_if_exists($seqname); $cp->drop_sequence_if_exists("${seqname}2"); +$cp->drop_sequence_if_exists("${seqname}'\"evil"); +$cp->drop_sequence_if_exists("${seqname}_cycle"); +$cp->drop_table_if_exists("$testtbl"); $t=qq{$S works when no sequences exist}; like ($cp->run(''), qr{OK:.+No sequences found}, $t); @@ -78,9 +82,55 @@ like ($cp->run('--critical=10%'), qr{CRITICAL:.+public.cp_test_sequence=11% \(ca $t=qq{$S returns correct information for a sequence with custom increment}; $dbh->do("ALTER SEQUENCE $seqname INCREMENT 10 MAXVALUE 90 RESTART WITH 25"); +$dbh->do("SELECT nextval('$seqname')"); # otherwise 25 isn't visible on PG10+ +$dbh->commit(); like ($cp->run('--critical=22%'), qr{CRITICAL:.+public.cp_test_sequence=33% \(calls left=6\)}, $t); $t=qq{$S returns correct information with MRTG output}; is ($cp->run('--critical=22% --output=mrtg'), "33\n0\n\npublic.cp_test_sequence\n", $t); +# create second sequence +$dbh->do("CREATE SEQUENCE ${seqname}2"); +$dbh->commit(); +$t=qq{$S returns correct information for two sequences}; +like ($cp->run(''), qr{OK:.+public.cp_test_sequence=33% .* \| .*${seqname}=33%.*${seqname}2=0%}, $t); + +# test SQL quoting +$dbh->do(qq{CREATE SEQUENCE "${seqname}'""evil"}); +$dbh->commit(); +$t=qq{$S handles SQL quoting}; +like ($cp->run(''), qr{OK:.+'public."${seqname}''""evil"'}, $t); # extra " and ' because name is both identifier+literal quoted + +# test CYCLE sequence skipping +$dbh->do(qq{CREATE SEQUENCE ${seqname}_cycle MAXVALUE 5 CYCLE}); +$dbh->commit(); +$dbh->do("SELECT setval('${seqname}_cycle',5)"); +like ($cp->run('--skipcycled'), qr{OK:.+public.cp_test_sequence_cycle=0%;85%;95% }, $t); +like ($cp->run(''), qr{CRITICAL:.+public.cp_test_sequence_cycle=100% \(calls left=0\) }, $t); + +$dbh->do("DROP SEQUENCE ${seqname}"); +$dbh->do("DROP SEQUENCE ${seqname}2"); +$dbh->do("DROP SEQUENCE ${seqname}_cycle"); +$dbh->do(qq{DROP SEQUENCE "${seqname}'""evil"}); + +# test integer column where the datatype range is smaller than the serial range +$dbh->do("CREATE TABLE $testtbl (id serial)"); +$dbh->do("SELECT setval('${testtbl}_id_seq',2000000000)"); +$dbh->commit; +$t=qq{$S handles "serial" column}; +like ($cp->run(''), qr{WARNING:.+public.sequence_test_id_seq=93% \(calls left=147483647\)}, $t); + +if ($ver >= 90200) { + # test smallint column where the datatype range is even smaller (and while we are at it, test --exclude) + $dbh->do("ALTER TABLE $testtbl ADD COLUMN smallid smallserial"); + $dbh->do("SELECT setval('${testtbl}_smallid_seq',30000)"); + $dbh->commit; + $t=qq{$S handles "smallserial" column}; + like ($cp->run('--exclude=sequence_test_id_seq'), qr{WARNING:.+public.sequence_test_smallid_seq=92% \(calls left=2767\)}, $t); +} else { + SKIP: { + skip '"smallserial" needs PostgreSQL 9.2 or later', 1; + } +} + exit; diff --git a/t/02_settings_checksum.t b/t/02_settings_checksum.t index 21c835e4..3374aea2 100644 --- a/t/02_settings_checksum.t +++ b/t/02_settings_checksum.t @@ -2,7 +2,7 @@ ## Test the "settings_checksum" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_slony_status.t b/t/02_slony_status.t index 9165e906..31836c49 100644 --- a/t/02_slony_status.t +++ b/t/02_slony_status.t @@ -2,7 +2,7 @@ ## Test the "slony_status" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_timesync.t b/t/02_timesync.t index 9e0432a4..2df8f785 100644 --- a/t/02_timesync.t +++ b/t/02_timesync.t @@ -2,7 +2,7 @@ ## Test of the the "version" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_txn_idle.t b/t/02_txn_idle.t index 89e50f68..a0e98473 100644 --- a/t/02_txn_idle.t +++ b/t/02_txn_idle.t @@ -2,11 +2,11 @@ ## Test the "txn_idle" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; -use Test::More tests => 15; +use Test::More tests => 14; use lib 't','.'; use CP_Testing; @@ -68,7 +68,7 @@ sleep(1); like ($cp->run(q{-w 0}), qr{longest idle in txn: \d+s}, $t); $t .= ' (MRTG)'; -like ($cp->run(q{--output=mrtg -w 0}), qr{\d+\n0\n\nPID:\d+ database:$dbname username:check_postgres_testing\n}, $t); +like ($cp->run(q{--output=mrtg -w 0}), qr{\d+\n0\n\nPID:\d+ database:$dbname username:check_postgres_testing}, $t); sleep(1); @@ -80,12 +80,6 @@ like ($cp->run(q{-w +0}), qr{Total idle in transaction: \d+\b}, $t); sleep(1); $t = qq{$S identifies idle using '1 for 2s'}; like ($cp->run(q{-w '1 for 2s'}), qr{1 idle transactions longer than 2s, longest: \d+s}, $t); - -$t = qq{$S returns an unknown if running as a non-superuser}; -my $olduser = $cp->{testuser}; -$cp->{testuser} = 'powerless_pete'; -like ($cp->run('-w 0'), qr{^$label UNKNOWN: .+superuser}, $t); - $idle_dbh->commit; exit; diff --git a/t/02_txn_time.t b/t/02_txn_time.t index 8be2fa18..6b7ed00c 100644 --- a/t/02_txn_time.t +++ b/t/02_txn_time.t @@ -2,7 +2,7 @@ ## Test the "txn_time" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -73,10 +73,11 @@ $t = qq{$S identifies a one-second running txn}; my $idle_dbh = $cp->test_database_handle(); $idle_dbh->do('SELECT 1'); sleep(1); -like ($cp->run(q{-w 0}), qr{longest txn: 1s}, $t); +like ($cp->run(q{-w 0}), qr{longest txn: [1-9]s}, $t); $t .= ' (MRTG)'; -like ($cp->run(q{--output=mrtg -w 0}), qr{\d+\n0\n\nPID:\d+ database:$dbname username:\w+ query:SELECT 1\n}, $t); +my $query_patten = ($ver >= 90200) ? 'SELECT 1' : ' in transaction'; +like ($cp->run(q{--output=mrtg -w 0}), qr{\d+\n0\n\nPID:\d+ database:$dbname username:\w+ query:$query_patten\n}, $t); $idle_dbh->commit; diff --git a/t/02_txn_wraparound.t b/t/02_txn_wraparound.t index ed6305e5..eb9c6dbf 100644 --- a/t/02_txn_wraparound.t +++ b/t/02_txn_wraparound.t @@ -2,7 +2,7 @@ ## Test the "txn_wraparound" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_version.t b/t/02_version.t index 77703ac9..8ff84294 100644 --- a/t/02_version.t +++ b/t/02_version.t @@ -2,7 +2,7 @@ ## Test the "version" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/02_wal_files.t b/t/02_wal_files.t index 7241825e..988e8180 100644 --- a/t/02_wal_files.t +++ b/t/02_wal_files.t @@ -2,11 +2,11 @@ ## Test the "wal_files" action -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; -use Test::More tests => 11; +use Test::More tests => 12; use lib 't','.'; use CP_Testing; @@ -49,6 +49,10 @@ like ($cp->run('--critical=1'), qr{^$label CRITICAL}, $t); $cp->drop_schema_if_exists(); $cp->create_fake_pg_table('pg_ls_dir', 'text'); +if ($ver >= 100000) { + $dbh->do(q{CREATE OR REPLACE FUNCTION cptest.pg_ls_waldir() RETURNS table(name text) AS 'SELECT * FROM cptest.pg_ls_dir' LANGUAGE SQL}); +} +$dbh->commit(); like ($cp->run('--critical=1'), qr{^$label OK}, $t); @@ -64,6 +68,19 @@ is ($cp->run('--critical=1 --output=mrtg'), "99\n0\n\n\n", $t); $t=qq{$S returns correct MRTG information}; is ($cp->run('--critical=101 --output=mrtg'), "99\n0\n\n\n", $t); +# test --lsfunc +my $xlogdir = $ver >= 100000 ? 'pg_wal' : 'pg_xlog'; +$dbh->do(qq{CREATE OR REPLACE FUNCTION ls_xlog_dir() + RETURNS SETOF TEXT + AS \$\$ SELECT pg_ls_dir('$xlogdir') \$\$ + LANGUAGE SQL + SECURITY DEFINER}); +$cp->create_fake_pg_table('ls_xlog_dir', ' '); +$dbh->do(q{INSERT INTO cptest.ls_xlog_dir SELECT 'ABCDEF123456ABCDEF123456' FROM generate_series(1,55)}); +$dbh->commit(); +$t=qq{$S returns correct number of files}; +like ($cp->run('--critical=1 --lsfunc=ls_xlog_dir'), qr{^$label CRITICAL.+ 55 \|}, $t); + $cp->drop_schema_if_exists(); exit; diff --git a/t/03_translations.t b/t/03_translations.t index 679d6156..ca199219 100644 --- a/t/03_translations.t +++ b/t/03_translations.t @@ -2,7 +2,7 @@ ## Run some sanity checks on the translations -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -11,6 +11,7 @@ BEGIN { %complete_langs = ( 'en' => 'English', 'fr' => 'French', + 'es' => 'Spanish', ); } use Test::More; @@ -128,7 +129,7 @@ $ok and pass $t; for my $l (sort keys %complete_langs) { my $language = $complete_langs{$l}; - next if $language eq 'English'; + next if 'English' eq $language; $ok = 1; $t=qq{Language $language contains all valid message strings}; @@ -177,8 +178,19 @@ for my $l (sort keys %complete_langs) { my $val = $msg{'en'}{$msg}->[1]; my $lval = $msg{$l}{$msg}->[1]; my $indent = $msg{$l}{$msg}->[0]; - next if $language eq 'French' and ($msg eq 'PID' or $msg eq 'port' or $msg eq 'pgbouncer-pool' - or $msg eq 'index' or $msg eq 'table' or $msg eq 'transactions' or $msg eq 'mode'); + next if 'French' eq $language and ( + 'PID' eq $msg + or 'port' eq $msg + or 'pgbouncer-pool' eq $msg + or 'index' eq $msg + or 'table' eq $msg + or 'transactions' eq $msg + or 'mode' eq $msg + ); + next if 'Spanish' eq $language and ( + 'checksum-msg' eq $msg + or 'pgbouncer-pool' eq $msg + ); if ($val eq $lval and $indent) { fail qq{Message '$msg' in language $language appears to not be translated, but it not marked as such}; $ok = 0; diff --git a/t/04_timeout.t b/t/04_timeout.t index 64e6857b..435f70f1 100644 --- a/t/04_timeout.t +++ b/t/04_timeout.t @@ -2,7 +2,7 @@ ## Test the timeout functionality -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -17,11 +17,11 @@ my $cp = CP_Testing->new( {default_action => 'custom_query'} ); $dbh = $cp->test_database_handle(); $t=q{Setting the --timeout flag works as expected}; -$res = $cp->run('--query="SELECT pg_sleep(2)" -w 7 --timeout=1'); +$res = $cp->run('--query="SELECT pg_sleep(10)" -w 7 --timeout=1'); like ($res, qr{Command timed out}, $t); $t=q{Setting the --timeout flag works as expected}; -$res = $cp->run('--query="SELECT pg_sleep(1)" -w 7 --timeout=2'); +$res = $cp->run('--query="SELECT pg_sleep(1)" -w 7 --timeout=10'); like ($res, qr{Invalid format}, $t); exit; diff --git a/t/05_docs.t b/t/05_docs.t index 7c7541a3..b48c64ad 100644 --- a/t/05_docs.t +++ b/t/05_docs.t @@ -2,7 +2,7 @@ ## Some basic checks on the documentation -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; @@ -38,6 +38,7 @@ for my $action (@actions) { next if $action =~ /last_auto/; my $match = $action; + $match = 'relation_size' if $match =~ /^(index|table|indexes|total_relation)_size/; $match = 'pgb_pool' if $match =~ /pgb_pool/; if ($slurp !~ /\n\s*sub check_$match/) { diff --git a/t/99_cleanup.t b/t/99_cleanup.t index 54bb707e..4b860b58 100644 --- a/t/99_cleanup.t +++ b/t/99_cleanup.t @@ -2,7 +2,7 @@ ## Cleanup any mess we made -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; diff --git a/t/99_perlcritic.t b/t/99_perlcritic.t index 4a1e449e..77838b71 100644 --- a/t/99_perlcritic.t +++ b/t/99_perlcritic.t @@ -3,7 +3,7 @@ ## Run Perl::Critic against the source code and the tests ## This is highly customized, so take with a grain of salt -use 5.006; +use 5.10.0; use strict; use warnings; use Test::More; @@ -41,17 +41,17 @@ for my $filename (qw{Makefile.PL check_postgres.pl t/CP_Testing.pm}) { -e $filename or die qq{Could not find "$filename"!}; open my $oldstderr, '>&', \*STDERR or die 'Could not dupe STDERR'; close STDERR or die qq{Could not close STDERR: $!}; - my @vio; - my $ranok = 0; - eval { - @vio = $critic->critique($filename); - $ranok = 1; - }; - if (! $ranok) { - pass "Perl::Critic failed for file $filename. Error was: $@\n"; - $@ = undef; - next; - } + my @vio; + my $ranok = 0; + eval { + @vio = $critic->critique($filename); + $ranok = 1; + }; + if (! $ranok) { + pass "Perl::Critic failed for file $filename. Error was: $@\n"; + $@ = undef; + next; + } open STDERR, '>&', $oldstderr or die 'Could not recreate STDERR'; ## no critic close $oldstderr or die qq{Could not close STDERR copy: $!}; my $vios = 0; diff --git a/t/99_pod.t b/t/99_pod.t index cb6c3557..a9c0d7e1 100644 --- a/t/99_pod.t +++ b/t/99_pod.t @@ -2,7 +2,7 @@ ## Check our Pod, requires Test::Pod -use 5.006; +use 5.10.0; use strict; use warnings; use Test::More; diff --git a/t/99_spellcheck.t b/t/99_spellcheck.t index 37a6f801..21b054eb 100644 --- a/t/99_spellcheck.t +++ b/t/99_spellcheck.t @@ -2,10 +2,11 @@ ## Spellcheck as much as we can -use 5.006; +use 5.10.0; use strict; use warnings; use Test::More; +use utf8; my (@testfiles, $fh); @@ -79,23 +80,24 @@ SKIP: { skip 'Need Pod::Text to re-test the spelling of embedded POD', 1; } - my $parser = Pod::Text->new (quotes => 'none'); + my $parser = Pod::Text->new (quotes => 'none', width => 400, utf8 => 1); for my $file (qw{check_postgres.pl}) { if (! -e $file) { fail(qq{Could not find the file "$file"!}); } - my $string; - my $tmpfile = "$file.tmp"; + my $string; + my $tmpfile = "$file.spellcheck.tmp"; $parser->parse_from_file($file, $tmpfile); - next if ! open my $fh, '<', $tmpfile; - { local $/; $string = <$fh>; } - close $fh or warn "Could not close $tmpfile\n"; - unlink $tmpfile; - spellcheck("POD from $file" => $string, $file); + next if ! open my $fh, '<:encoding(UTF-8)', $tmpfile; + { local $/; $string = <$fh>; } + close $fh or warn "Could not close $tmpfile\n"; + unlink $tmpfile; + spellcheck("POD inside $file" => $string, $file); } } + ## Now the comments SKIP: { if (!eval { require File::Comments; 1 }) { @@ -132,22 +134,28 @@ __DATA__ ## Common: arrayref +async autovac Backends backends bc bucardo checksum +chroot commitratio +consrc cp dbh dbstats +df DBI DSN ENV +filesystem fsm goto hitratio +lsfunc Mullane Nagios ok @@ -169,14 +177,18 @@ Pre runtime Schemas selectall +skipcycled +skipobject Slony slony stderr syslog tcl +timestamp tnm txn txns +turnstep tuples wal www @@ -214,6 +226,7 @@ xmlns ## check_postgres.pl: Abrigo +Adrien Ahlgren Ahlgren Albe @@ -236,9 +249,11 @@ baz bigint Blasco Blasco +Brüssel blks Boes boxinfo +Boxinfo Bracht Bracht Bucardo @@ -248,6 +263,7 @@ checkpostgresrc checksum checksums checktype +Cédric Christoph commitratio commitratio @@ -287,6 +303,7 @@ EB Eisentraut Eloranta Eloranta +Elsasser emma endcrypt EnterpriseDB @@ -325,6 +342,8 @@ gtld Guettler Guillaume Gurjeet +Hagander +Hansper hardcode Henrik Henrik @@ -332,6 +351,7 @@ HiRes hitratio hitratio hitratio +Holger hong HOSTADDRESS html @@ -353,8 +373,10 @@ Ioannis ioguix Jacobo Jacobo +Janes Jehan Jens +Jürgen Kabalin Kirkwood klatch @@ -372,6 +394,7 @@ localtime Logfile logtime Mager +Magnus maindatabase Makefile Mallett @@ -386,6 +409,7 @@ Mika MINIPAGES MINPAGES minvalue +Moench morpork mrtg MRTG @@ -393,13 +417,14 @@ msg multi nagios NAGIOS +Nayrat +Nenciarini nextval nnx nofuncbody nofunctions noidle noindexes -nolanguage nols noname noname @@ -444,6 +469,7 @@ PGSERVICE pgsql PGUSER pid +Pirogov plasmid plugin pluto @@ -475,8 +501,8 @@ Renner repinfo RequireInterpolationOfMetachars ret -rgen ritical +rgen robert Rorthais runtime @@ -492,6 +518,7 @@ seqscan seqscan seqtupread seqtupread +SETOF showperf Sijmons Singh @@ -521,6 +548,7 @@ tablespaces Tambouras tardis Taveira +Tegeder tempdir tgisconstraint Thauvin @@ -543,11 +571,13 @@ utf valtype Villemain Vitkovsky +Vondendriesch Waisbrot Waisbrot wal WAL watson +Webber Westwood wget wiki @@ -556,4 +586,5 @@ wilkins xact xlog Yamada +Yochum Zwerschke diff --git a/t/CP_Testing.pm b/t/CP_Testing.pm index 28ff60b3..056b9572 100644 --- a/t/CP_Testing.pm +++ b/t/CP_Testing.pm @@ -1,12 +1,13 @@ -package CP_Testing; +package CP_Testing; ## -*- mode: CPerl; indent-tabs-mode: nil; cperl-indent-level: 4 -*- ## Common methods used by the other tests for check_postgres.pl -use 5.006; +use 5.10.0; use strict; use warnings; use Data::Dumper; use Time::HiRes qw/sleep/; +use File::Spec::Functions; use DBI; use Cwd; @@ -56,6 +57,9 @@ sub cleanup { if (-l $symlink) { unlink $symlink; } + my $datadir = "$dbdir$dirnum"; + system("rm -fr $datadir"); + } return; @@ -63,6 +67,18 @@ sub cleanup { } sub test_database_handle { + my $self = shift; + my $dbh; + eval { $dbh = $self->_test_database_handle(@ARGV) }; + if (!defined $dbh) { + Test::More::diag $@; + Test::More::BAIL_OUT 'Cannot continue without a test database'; + return undef; ## no critic (Subroutines::ProhibitExplicitReturnUndef) + } + return $dbh; +} + +sub _test_database_handle { ## Request for a database handle: create and startup DB as needed @@ -79,20 +95,50 @@ sub test_database_handle { -e $dbdir and die qq{Oops: I cannot create "$dbdir", there is already a file there!\n}; - Test::More::diag qq{Creating database in directory "$dbdir"\n}; - mkdir $dbdir; } + ## Find a working initdb (which also helps us find other binaries) + my $initdb = + $ENV{PGINITDB} ? $ENV{PGINITDB} + : $ENV{PGBINDIR} ? "$ENV{PGBINDIR}/initdb" + : 'initdb'; + + my ($imaj,$imin); + + my $initversion = qx{$initdb --version 2>/dev/null}; + if ($initversion =~ /(\d+)(?:\.(\d+))?/) { + ($imaj,$imin) = ($1,$2); + } + else { + ## Work harder to find initdb. First check Debian area + my $basedir = '/usr/lib/postgresql/'; + if (opendir my $dh, $basedir) { + for my $subdir (sort { $b <=> $a } grep { /^\d+[\d\.]+$/ } readdir $dh) { + $initdb = catfile($basedir, $subdir, 'bin', 'initdb'); + if (-e $initdb) { + $initversion = qx{$initdb --version 2>/dev/null}; + if ($initversion =~ /(\d+)(?:\.(\d+))?/) { + ($imaj,$imin) = ($1,$2); + last; + } + } + } + closedir $dh; + } + if (!defined $imaj) { + die qq{Could not determine the version of initdb in use! Got ($initversion) from ($initdb --version)\n}; + } + } + my $datadir = "$dbdir/data space"; if (! -e $datadir) { - my $initdb - = $ENV{PGINITDB} ? $ENV{PGINITDB} - : $ENV{PGBINDIR} ? "$ENV{PGBINDIR}/initdb" - : 'initdb'; - - $com = qq{LC_ALL=en LANG=C $initdb --locale=C -E UTF8 -D "$datadir" 2>&1}; + $com = sprintf q{LANG=C %s %s --locale C -E UTF8 -D "%s" 2>&1}, + $initdb, + # Speed up testing on 9.3+ + ($imaj > 9 or (9 == $imaj and $imin >= 3)) ? ' --nosync' : '', + $datadir; eval { $DEBUG and warn qq{About to run: $com\n}; $info = qx{$com}; @@ -108,41 +154,44 @@ sub test_database_handle { print $cfh qq{port = 5432\n}; print $cfh qq{listen_addresses = ''\n}; print $cfh qq{max_connections = 10\n}; - - ## Grab the version for finicky items - if (qx{$initdb --version} !~ /(\d+)\.(\d+)/) { - die qq{Could not determine the version of initdb in use!\n}; - } - my ($imaj,$imin) = ($1,$2); + print $cfh qq{fsync = off\n}; + print $cfh qq{log_destination = 'stderr'\n}; ## <= 8.0 - if ($imaj < 8 or ($imaj==8 and $imin <= 1)) { + if ($imaj < 8 or (8 == $imaj and $imin <= 1)) { print $cfh qq{stats_command_string = on\n}; } ## >= 8.1 - if ($imaj > 8 or ($imaj==8 and $imin >= 1)) { + if ($imaj > 8 or (8 == $imaj and $imin >= 1)) { print $cfh qq{autovacuum = off\n}; print $cfh qq{max_prepared_transactions = 5\n}; } ## >= 8.3 - if ($imaj > 8 or ($imaj==8 and $imin >= 3)) { + if ($imaj > 8 or (8 == $imaj and $imin >= 3)) { print $cfh qq{logging_collector = off\n}; } ## <= 8.2 - if ($imaj < 8 or ($imaj==8 and $imin <= 2)) { + if ($imaj < 8 or (8 == $imaj and $imin <= 2)) { print $cfh qq{redirect_stderr = off\n}; print $cfh qq{stats_block_level = on\n}; print $cfh qq{stats_row_level = on\n}; } ## <= 8.3 - if ($imaj < 8 or ($imaj==8 and $imin <= 3)) { + if ($imaj < 8 or (8 == $imaj and $imin <= 3)) { print $cfh qq{max_fsm_pages = 99999\n}; } + ## >= 9.4 + if ($imaj > 9 or (9 == $imaj and $imin >= 4)) { + print $cfh qq{max_replication_slots = 2\n}; + print $cfh qq{wal_level = logical\n}; + print $cfh qq{max_wal_senders = 2\n}; + } + print $cfh "\n"; close $cfh or die qq{Could not close "$cfile": $!\n}; @@ -163,7 +212,7 @@ sub test_database_handle { close $fh or die qq{Could not open "$pidfile": $!\n}; ## Send a signal to see if this PID is alive $count = kill 0 => $pid; - if ($count == 0) { + if (0 == $count) { Test::More::diag qq{Found a PID file, but no postmaster. Removing file "$pidfile"\n}; unlink $pidfile; $needs_startup = 1; @@ -173,9 +222,9 @@ sub test_database_handle { my $pg_ctl = $ENV{PG_CTL} ? $ENV{PG_CTL} : $ENV{PGBINDIR} ? "$ENV{PGBINDIR}/pg_ctl" - : 'pg_ctl'; + : do { ($_ = $initdb) =~ s/initdb/pg_ctl/; $_ }; - if (qx{$pg_ctl --version} !~ /(\d+)\.(\d+)/) { + if (qx{$pg_ctl --version} !~ /(\d+)(?:\.(\d+))?/) { die qq{Could not determine the version of pg_ctl in use!\n}; } my ($maj,$min) = ($1,$2); @@ -189,11 +238,11 @@ sub test_database_handle { unlink $logfile; my $sockdir = 'socket'; - if ($maj < 8 or ($maj==8 and $min < 1)) { + if ($maj < 8 or (8 == $maj and $min < 1)) { $sockdir = qq{"$dbdir/data space/socket"}; } - $com = qq{LC_ALL=en LANG=C $pg_ctl -o '-k $sockdir' -l $logfile -D "$dbdir/data space" start}; + $com = qq{LANG=C $pg_ctl -o '-k $sockdir' -l $logfile -D "$dbdir/data space" start}; eval { $DEBUG and warn qq{About to run: $com\n}; $info = qx{$com}; @@ -220,7 +269,7 @@ sub test_database_handle { } close $logfh or die qq{Could not close "$logfile": $!\n}; - if ($maj < 8 or ($maj==8 and $min < 1)) { + if ($maj < 8 or (8 == $maj and $min < 1)) { my $host = "$here/$dbdir/data space/socket"; my $COM; @@ -277,7 +326,7 @@ sub test_database_handle { $newname .= $1; } if (! -e $newname) { - warn "Creating new symlink socket at $newname\n"; + $DEBUG and warn "Creating new symlink socket at $newname\n"; (my $oldname = $dbhost) =~ s/\\//g; symlink $oldname => $newname; } @@ -293,11 +342,12 @@ sub test_database_handle { $dbh = DBI->connect(@superdsn); }; if ($@) { + my (@tempdsn, $tempdbh); if ($@ =~ /role .+ does not exist/) { ## We want the current user, not whatever this is set to: delete $ENV{PGUSER}; - my @tempdsn = ($dsn, '', '', {AutoCommit=>1,RaiseError=>1,PrintError=>0}); - my $tempdbh = DBI->connect(@tempdsn); + @tempdsn = ($dsn, '', '', {AutoCommit=>1,RaiseError=>1,PrintError=>0}); + $tempdbh = DBI->connect(@tempdsn); $tempdbh->do("CREATE USER $dbuser SUPERUSER"); $tempdbh->disconnect(); $dbh = DBI->connect(@superdsn); @@ -305,8 +355,8 @@ sub test_database_handle { elsif ($@ =~ /database "postgres" does not exist/) { ## We want the current user, not whatever this is set to: (my $tempdsn = $dsn) =~ s/postgres/template1/; - my @tempdsn = ($tempdsn, '', '', {AutoCommit=>1,RaiseError=>1,PrintError=>0}); - my $tempdbh = DBI->connect(@tempdsn); + @tempdsn = ($tempdsn, '', '', {AutoCommit=>1,RaiseError=>1,PrintError=>0}); + $tempdbh = DBI->connect(@tempdsn); $tempdbh->do('CREATE DATABASE postgres'); $tempdbh->disconnect(); $dbh = DBI->connect(@superdsn); @@ -321,7 +371,7 @@ sub test_database_handle { $dbh->{AutoCommit} = 1; $dbh->{RaiseError} = 0; - if ($maj > 8 or ($maj==8 and $min >= 1)) { + if ($maj > 8 or (8 == $maj and $min >= 1)) { $SQL = q{SELECT count(*) FROM pg_user WHERE usename = ?}; $sth = $dbh->prepare($SQL); $sth->execute($dbuser); @@ -336,11 +386,14 @@ sub test_database_handle { $dbh->do("CREATE USER $user2"); } } - $dbh->do('CREATE DATABASE beedeebeedee'); - $dbh->do('CREATE DATABASE ardala'); - $dbh->do('CREATE LANGUAGE plpgsql'); - $dbh->do('CREATE LANGUAGE plperlu'); - $dbh->do("CREATE SCHEMA $fakeschema"); + + my $databases = $dbh->selectall_hashref('SELECT datname FROM pg_database', 'datname'); + $dbh->do('CREATE DATABASE beedeebeedee') unless ($databases->{beedeebeedee}); + $dbh->do('CREATE DATABASE ardala') unless ($databases->{ardala}); + my $languages = $dbh->selectall_hashref('SELECT lanname FROM pg_language', 'lanname'); + $dbh->do('CREATE LANGUAGE plpgsql') unless ($languages->{plpgsql}); + my $schemas = $dbh->selectall_hashref('SELECT nspname FROM pg_namespace', 'nspname'); + $dbh->do("CREATE SCHEMA $fakeschema") unless ($schemas->{$fakeschema}); $dbh->{AutoCommit} = 0; $dbh->{RaiseError} = 1; @@ -367,6 +420,7 @@ sub test_database_handle { ## Sanity check $dbh->do("ALTER USER $dbuser SET search_path = public"); $dbh->do('SET search_path = public'); + $dbh->do('SET client_min_messages = WARNING'); $dbh->do('COMMIT'); return $dbh; @@ -375,7 +429,7 @@ sub test_database_handle { sub recreate_database { - ## Given a database handle, comepletely recreate the current database + ## Given a database handle, completely recreate the current database my ($self,$dbh) = @_; @@ -401,6 +455,7 @@ sub recreate_database { $dsn = "DBI:Pg:dbname=$dbname;port=$port;host=$host"; $dbh = DBI->connect($dsn, $user, '', {AutoCommit=>0, RaiseError=>1, PrintError=>0}); + $dbh->do('SET client_min_messages = WARNING'); return $dbh; @@ -437,6 +492,7 @@ sub run { $com .= qq{ --dbuser=$dbuser}; } if ($extra =~ s/--nodbname//) { + $ENV{PGDATABASE} = ''; } elsif ($extra !~ /dbname=/) { $com .= " --dbname=$dbname"; @@ -480,7 +536,7 @@ sub get_host { sub get_port { my $self = shift; - return 5432; + return 5432; } sub get_shorthost { @@ -700,7 +756,8 @@ sub drop_sequence_if_exists { $SQL = q{SELECT count(*) FROM pg_class WHERE relkind = 'S' AND relname = } . $dbh->quote($name); my $count = $dbh->selectall_arrayref($SQL)->[0][0]; if ($count) { - $dbh->do("DROP SEQUENCE $name"); + $name =~ s/"/""/g; + $dbh->do("DROP SEQUENCE \"$name\""); $dbh->commit(); } return; @@ -750,6 +807,31 @@ SELECT 'PostgreSQL $version on fakefunction for check_postgres.pl testing'::text } ## end of fake version +sub fake_version_timeout { + + my $self = shift; + my $dbh = $self->{dbh} || die; + my $dbuser = $self->{testuser} || die; + + if (! $self->schema_exists($dbh, $fakeschema)) { + $dbh->do("CREATE SCHEMA $fakeschema"); + } + + $dbh->do(qq{ +CREATE OR REPLACE FUNCTION $fakeschema.version() +RETURNS TEXT +LANGUAGE SQL +AS \$\$ +SELECT pg_sleep(10)::text; +\$\$ +}); + $dbh->do("ALTER USER $dbuser SET search_path = $fakeschema, public, pg_catalog"); + $dbh->commit(); + return; + +} ## end of fake version timeout + + sub fake_self_version { ## Look out... @@ -804,9 +886,9 @@ sub drop_all_tables { my $self = shift; my $dbh = $self->{dbh} || die; $dbh->{Warn} = 0; - my @info = $dbh->tables('','public','','TABLE'); - for my $tab (@info) { - $dbh->do("DROP TABLE $tab CASCADE"); + my $tables = $dbh->selectall_arrayref(q{SELECT tablename FROM pg_tables WHERE schemaname = 'public'}); + for my $tab (@$tables) { + $dbh->do("DROP TABLE $tab->[0] CASCADE"); } $dbh->{Warn} = 1; $dbh->commit(); @@ -821,6 +903,12 @@ sub database_sleep { my $ver = $dbh->{pg_server_version}; if ($ver < 80200) { + $dbh->{AutoCommit} = 1; + $dbh->{RaiseError} = 0; + $dbh->do('CREATE LANGUAGE plperlu'); + $dbh->{AutoCommit} = 0; + $dbh->{RaiseError} = 1; + $SQL = q{CREATE OR REPLACE FUNCTION pg_sleep(float) RETURNS VOID LANGUAGE plperlu AS 'select(undef,undef,undef,shift)'}; $dbh->do($SQL); $dbh->commit(); pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy