Fix pg_rewind regression failure after "fast promotion"

hlinnaka · hlinnaka · commit 484a848a73fc · 2015-04-30T21:59:58.000-07:00
pg_rewind looks at the control file to determine the server's timeline. If
the standby performs a "fast promotion", the timeline ID in the control
file is not updated until the next checkpoint. The startup process requests
a checkpoint immediately after promotion, so this is unlikely to be an
issue in the real world, but the regression suite ran pg_rewind so quickly
after promotion that the checkpoint had not yet completed.

Reported by Stephen Frost
diff --git a/src/bin/pg_rewind/RewindTest.pm b/src/bin/pg_rewind/RewindTest.pm
@@ -242,6 +242,14 @@ sub promote_standby
 	system_or_bail("pg_ctl -w -D $test_standby_datadir promote >>$log_path 2>&1");
 	poll_query_until("SELECT NOT pg_is_in_recovery()", $connstr_standby)
 		or die "Timed out while waiting for promotion of standby";
+
+	# Force a checkpoint after the promotion. pg_rewind looks at the control
+	# file todetermine what timeline the server is on, and that isn't updated
+	# immediately at promotion, but only at the next checkpoint. When running
+	# pg_rewind in remote mode, it's possible that we complete the test steps
+	# after promotion so quickly that when pg_rewind runs, the standby has not
+	# performed a checkpoint after promotion yet.
+	standby_psql("checkpoint");
 }
 
 sub run_pg_rewind
