Details
Description
http://buildbot.askmonty.org/buildbot/builders/kvm-rpm-centos74-amd64-debug/builds/4005
bb-10.2-release 985f63cce1057d7b3fd6e86e65a2a2ee 2020-05-08 19:07:46 |
main.processlist_notembedded w1 [ fail ]
|
Test ended at 2020-05-08 17:33:56
|
|
CURRENT_TEST: main.processlist_notembedded
|
mysqltest: In included file "/usr/share/mysql-test/include/wait_until_count_sessions.inc":
|
included from /usr/share/mysql-test/t/processlist_notembedded.test at line 37:
|
At line 125: Timeout in wait_until_count_sessions.inc
|
|
The result from queries just before the failure was:
|
#
|
# MDEV-20466: SHOW PROCESSLIST truncates query text on \0 bytes
|
#
|
connect con1,localhost,root,,;
|
connection con1;
|
SET DEBUG_SYNC= 'before_join_optimize SIGNAL in_sync WAIT_FOR go';
|
connection default;
|
SET DEBUG_SYNC= 'now WAIT_FOR in_sync';
|
FOUND 1 /sleep/ in MDEV-20466.text
|
SET DEBUG_SYNC= 'now SIGNAL go';
|
disconnect con1;
|
SET DEBUG_SYNC = 'RESET';
|
# Timeout in wait_until_count_sessions.inc
|
# Number of sessions expected: <= 1 found: 2
|
SHOW PROCESSLIST;
|
Id User Host db Command Time State Info Progress
|
11071 root localhost test Query 0 init SHOW PROCESSLIST 0.000
|
11072 root localhost test Query 10 debug sync point: before_join_optimize SELECT user FROM mysql.user WHERE user ='some
|
Attachments
Issue Links
- relates to
-
MDEV-20466 SHOW PROCESSLIST truncates query text on \0 bytes
-
- Closed
-
-
MDEV-21462 main.processlist_notembedded fails to clean up, causes perfschema test failures
-
- Closed
-
As reported in
MDEV-21462, this test was ruining subsequent tests by failing to ensure that all connections terminated.My change was merged by the original test author sanja upward. The bb-10.2-release branch where the test failed is a merge from bb-10.1-release. There, that test did not fail anywhere, but notably, kvm-rpm-centos74-amd64 did not run at all on bb-10.1-release.
I tried and failed to reproduce the failure locally, despite 1,200 repeats. The [ retry-fail ] in the buildbot log looks promising. I did not find anything useful in the server error log.
I am reassigning this back to the original test author, who also did the merge to bb-10.2-release. I hope that this is well repeatable on the buildbot VM.
I only fixed
MDEV-21462because the badly written test was causing a potentially large number of subsequent tests to fail. Luckily, this failure appears to be the failure of a single test for a single bug.