crash_case_error_log.txt
webdb1_crash17.log
[MDEV-32834] Galera buffer cache gets corrupted and mariadb crashes with signal 6 Created: 2023-11-18 Updated: 2024-01-17 |
|
| Status: | Needs Feedback |
| Project: | MariaDB Server |
| Component/s: | Galera |
| Affects Version/s: | 10.5.6 |
| Fix Version/s: | 10.4, 10.5, 10.6 |
| Type: | Bug | Priority: | Critical |
| Reporter: | Michael Amadi | Assignee: | Michael Amadi |
| Resolution: | Unresolved | Votes: | 1 |
| Labels: | None | ||
| Environment: |
CentOS Linux 7 (Core) |
||
| Attachments: |
crash_case_error_log.txt
webdb1_crash17.log
|
| Description |
|
During normal node operation (not startup or IST/SST), Galera buffer cache gets corrupted and mariadb crashes with signal 6. From the logs, this causes a signal 6 abort, attempts a backtrsace and hits a Fatal signal 11. However system does an IST and rejoins the cluster. |
| Comments |
| Comment by Jan Lindström [ 2023-11-21 ] |
|
michael.amadi Unfortunately, provided log does not really contain any such information that would help to give even educated guess why server is crashing. It could be that gcache gets corrupted as error log suggest. However, used MariaDB server release and Galera library are very old. Therefore, I suggest to use more recent version of MariaDB server and Galera library and if issue reproduces please provide full error log from all nodes, opened stack trace from core dump (for some reason it was not successful on report, you need to open core dump in same environment as where server is running) and node configuration. |