main.subselect_notembedded fails on Ubuntu Oneiric in Jenkins
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
MySQL Server |
Unknown
|
Unknown
|
|||
Percona Server moved to https://jira.percona.com/projects/PS |
Invalid
|
High
|
Unassigned | ||
5.1 |
Fix Released
|
High
|
Laurynas Biveinis | ||
5.5 |
Invalid
|
High
|
Unassigned |
Bug Description
main.subselect_
_StringException: Text attachment: traceback
------------
Comment: Server [mysqld.1 - pid: 14587, winpid: 14587, exit: 256] failed during test run
Server log from this test:
----------SERVER LOG START-----------
111209 20:03:37 [Warning] '--default-
111209 20:03:37 [Note] Flashcache bypass: disabled
111209 20:03:37 [Note] Flashcache setup error is : open flash device failed
111209 20:03:37 [Note] Plugin 'FEDERATED' is disabled.
111209 20:03:37 [Note] Plugin 'InnoDB' is disabled.
111209 20:03:37 [Warning] /home/jenkins/
111209 20:03:37 [Note] Event Scheduler: Loaded 0 events
111209 20:03:37 [Note] /home/jenkins/
Version: '5.1.60-log' socket: '/tmp/Vy4Y0ESdV
----------SERVER LOG END-------------
Logfile:
CURRENT_TEST: main.not_
mysqltest: At line 51: query 'SELECT '1' RLIKE RPAD('1', 10000, '(')' failed with wrong errno 2013: 'Lost connection to MySQL server during query', instead of 1436...
The result from queries just before the failure was:
select 1;
1
1
SHOW VARIABLES like 'slave_
Variable_name Value
slave_skip_errors OFF
#
# Bug#58026: massive recursion and crash in regular expression handling
#
------------
Related branches
- Stewart Smith (community): Approve
-
Diff: 42 lines (+16/-16)1 file modifiedPercona-Server/config/ac-macros/misc.m4 (+16/-16)
Changed in percona-server: | |
status: | New → Confirmed |
Changed in percona-server: | |
importance: | Undecided → High |
status: | Confirmed → Triaged |
The tests not_embedded_ server sp_notembedded subselect_ notembedded execution_ constants
main.
main.
main.
main.
always fail or pass together and this happens whenever the stack growth direction is determined wrong (i.e. 1 instead of -1 on intel). See bug 902471. Only applies to 5.1, not 5.5.