Fixes stream load memory leak in WifiSecureClient for SSL CACert, Certificate, and Private Key #6387

booptech · 2022-03-08T02:00:57Z

No description provided.

…tificate, and Private Key. Issue presented during any subsequent invocation of loadCACert, loadCertificate, and loadPrivateKey, respectively, after the first invocation.

CLAassistant · 2022-03-08T02:01:03Z

All committers have signed the CLA.

mrengineer7777 · 2022-03-08T18:07:19Z

Possibly related:
#6062
#6055

timr49 · 2022-03-21T05:06:03Z

Hmmm it's working well for initial load, and if same cert is reloaded again, but if there is a different cert loaded it will panic with:

assert failed: heap_caps_realloc heap_caps.c:345 (heap != NULL && "free() pointer is outside heap areas")

i will investigate a bit more and push another commit with fixes.

Anyone have any ideas on this?

Feedback on proposed fix:

In each of these three cases, if _streamLoad() returns null (e.g. because the stream has insufficient data available or malloc() fails) then _CA_cert/_cert/_private_key etc. are left pointing to space that has now been free()d. Suggest that between each call to free() and the following _streamLoad(), the member just free()d is set to null.2.
_CA_cert/_cert/_private_key may have been passed by the calling application via setCACert/setCertificate/setPrivateKey in which case the calling application would (should?) be free()ing them, leading to the same space being free()d twice.
There are still opportunities for memory leaks. For example ...

(a) the destructor for WiFiClientSecure does not appear to free any of these things. For example, if the example was changed to instantiate a new WiFiClientSecure object each time, it would get the some amount of leakage.

(b) a call to setInsecure() will leak any of these things that have previously been allocated;

note that the solution is not as simple as free()ing them in ~WiFiClientSecure() etc. because they may have been allocated by the calling application and passed via calls to setCACert/setCertificate/setPrivateKey, in which case the calling application will be expecting to free() them.
therefore, a solution needs allocator/reallocator/deallocator methods which take account of how they were previously allocated (or not), deallocates if appropriate before allocating/reallocating, and which deallocates when ~WiFiClientSecure() (or Insecure() or anything else that currently just assigns null) are called.

github-actions · 2022-04-26T11:26:24Z

Unit Test Results

0 files 0 suites 0s ⏱️
0 tests 0 ✔️ 0 💤 0 ❌

Results for commit 46bf03c.

me-no-dev · 2022-04-26T11:44:28Z

Hi @fillybrese ! It is always good to make sure that we do not leak any memory. Thanks for the contribution!

TD-er · 2022-04-26T11:56:08Z

I only now noticed this PR...
I think there are multiple other issues here, especially with the mbedtls data structure.
This one isn't initialized correct, so the internal members may still contain random data resuling in pointers not being null and thus cause crashes when destructing the structure.
I did mention this several times in related issues.
See also the 4 "Helpers/ESPEasy_xxx" files in this PR: https://github.com/letscontrolit/ESPEasy/pull/3788/files#diff-e1489f0e20f34ec2d70451ffec1baaeeac1c789eb4c2dc54219e75e5ea6c10d5
These are copy/paste versions of the files in this repository, but including changes.

What happens is if a connection is failed and retried, then you may run into these issues. Partly fixed in this PR, but not in the other files (as far as I know)

* LittleFs is working with C3 * Delete .skip.esp32c3 * Add support for extra flash images (espressif#6625) This PR adds support for uploading additional flash images (e.g. Adafruit Tiny UF2 bootloader) specified in board manifests. Additionally, the PR switches the PlatformIO CI script to the upstream version of the ESP32 dev-platform (basically reverts changes introduced in espressif#5387 as they are no longer required). * publish.yml: Remove a leftover parenthesis that was making the workflow (espressif#6620) Description of Change Remove a leftover parenthesis that was making the workflow that was making the workflow invalid. Tests scenarios Github Workflow. Related links https://github.com/espressif/arduino-esp32/actions/runs/2213167501 Signed-off-by: Abdelatif Guettouche <abdelatif.guettouche@espressif.com> * Enable LittleFs sketches for C3 (espressif#6618) * LittleFs is working with C3 * Delete .skip.esp32c3 * Update LittleFS PlatformIO example (espressif#6617) * Added RainMaker support on Arduino IDE for ESP32-C3/S2/S3 (espressif#6598) * Added RainMaker support on Arduino IDE for ESP32-C3/S2/S3 Closes espressif#6573 Note related to the issue espressif#6435 * Touch change to init only selected GPIO. (espressif#6609) * Separated init for touch / channel called by touchRead() * compile error * Fixed touch_V2 + ISR * Allow BluetoothSerial::connect() with specified channel and more options (espressif#6380) * BTAddress const, add bool() * BTAdvertisedDevice: const functions * BluetoothSerial: add: getChannels, add isClosed, add read/peek timeout, add connect with channel# * BluetoothSerial: add sec_mask, role in ::connect * BluetoothSerial add discover and connect with channel number example * DiscoverConnect: add SPP_ENABLED check * DiscoverConnect: disable on esp32s3 * Fixes stream load memory leak in WifiSecureClient for SSL CACert, Certificate, and (espressif#6387) Private Key. Issue presented during any subsequent invocation of loadCACert, loadCertificate, and loadPrivateKey, respectively, after the first invocation. * Call i2c_set_timeout in i2cSetClock (espressif#6537) * Uniform behaviour of WiFiClientSecure and WiFiClient setTimeout() (espressif#6562) * Uniform timeout WiFiClient-WiFiClientSecure * Added missing prototype * Add socket check on setTimeout * enh(log) salvage TAG from ESP_IDF log-statements > (espressif#6567) by converting result log-rows from the 1st line to the 2nd (`NET` is the tag): ``` [ 73419][D][telelogger.cpp:915] telemetry(): state: 33C [ 73419][D][telelogger.cpp:915] telemetry(): [NET] state: 33C ``` Co-authored-by: Me No Dev <me-no-dev@users.noreply.github.com> * add AirM2M_CORE_ESP32C3 board (espressif#6613) * add AirM2M_CORE_ESP32C3 board * Added Unexpected Maker Reflow Master Pro (UM RMP) board (espressif#6630) Fixed wrong SCK and MISO pins for TinyS2 * Tasmota changes (#24) * Update install-arduino-core-esp32.sh * Update CMakeLists.txt * Update Esp.cpp * Update Updater.cpp Co-authored-by: Valerii Koval <valeros@users.noreply.github.com> Co-authored-by: Abdelatif Guettouche <abdelatif.guettouche@espressif.com> Co-authored-by: Pedro Minatel <pedro.minatel@espressif.com> Co-authored-by: Jan Procházka <90197375+P-R-O-C-H-Y@users.noreply.github.com> Co-authored-by: Christian Ferbar <5595808+ferbar@users.noreply.github.com> Co-authored-by: Billy <billyfrese@gmail.com> Co-authored-by: Paul <pc235603@gmail.com> Co-authored-by: Gonzalo Brusco <gonzabrusco@gmail.com> Co-authored-by: Kostis Anagnostopoulos <ankostis@gmail.com> Co-authored-by: Me No Dev <me-no-dev@users.noreply.github.com> Co-authored-by: Darren Cheng <1912544842@qq.com> Co-authored-by: Unexpected Maker <seon@unexpectedmaker.com>

mrengineer7777 · 2022-05-09T22:49:56Z

I will review the feedback and implement the changes as soon as i can. Just have been short on time recently, will do my best.

There were the issues discovered by myself and others in the original PR code. I would have recommended merging to master be delayed until these raised points are addressed.

Understood. Your PR was helpful, so can't complain about the merge. Create a new PR to address the remaining issues when you have a chance. Thanks in advance for your contribution!

Fixes stream load memory leak in WifiSecureClient for SSL CACert, Cer…

cec0701

…tificate, and Private Key. Issue presented during any subsequent invocation of loadCACert, loadCertificate, and loadPrivateKey, respectively, after the first invocation.

Merge branch 'master' into issue-5826-fix

46bf03c

me-no-dev merged commit d7ffd57 into espressif:master Apr 26, 2022

mrengineer7777 mentioned this pull request Mar 3, 2023

Fix memory leak on stream load when updating CA Certificate #6062

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Fixes stream load memory leak in WifiSecureClient for SSL CACert, Certificate, and Private Key #6387

Fixes stream load memory leak in WifiSecureClient for SSL CACert, Certificate, and Private Key #6387

Uh oh!

booptech commented Mar 8, 2022 •

edited

Loading

Uh oh!

CLAassistant commented Mar 8, 2022 •

edited

Loading

Uh oh!

mrengineer7777 commented Mar 8, 2022

Uh oh!

timr49 commented Mar 21, 2022

Uh oh!

github-actions bot commented Apr 26, 2022

Uh oh!

me-no-dev commented Apr 26, 2022

Uh oh!

TD-er commented Apr 26, 2022

Uh oh!

mrengineer7777 commented May 9, 2022

Uh oh!

Uh oh!

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

Fixes stream load memory leak in WifiSecureClient for SSL CACert, Certificate, and Private Key #6387

Fixes stream load memory leak in WifiSecureClient for SSL CACert, Certificate, and Private Key #6387

Uh oh!

Conversation

booptech commented Mar 8, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

CLAassistant commented Mar 8, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

mrengineer7777 commented Mar 8, 2022

Uh oh!

timr49 commented Mar 21, 2022

Uh oh!

github-actions bot commented Apr 26, 2022

Unit Test Results

Uh oh!

me-no-dev commented Apr 26, 2022

Uh oh!

TD-er commented Apr 26, 2022

Uh oh!

mrengineer7777 commented May 9, 2022

Uh oh!

Uh oh!

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

booptech commented Mar 8, 2022 •

edited

Loading

CLAassistant commented Mar 8, 2022 •

edited

Loading