Page MenuHomePhabricator

db2175 replication breakage with: Error 'Index for table 'recentchanges' is corrupt; try to repair it' on query. Default database: 'nlwiki'.
Closed, ResolvedPublic

Description

Was depooled and downtimed until Monday.

CRITICAL slave_sql_state Slave_SQL_Running: No, Errno: 1034, Errmsg: Error 'Index for table 'recentchanges' is corrupt: try to repair it' on query. Default database: 'nlwiki'. [Query snipped]

Event Timeline

Started on a screen on db2175:

mysql:root@localhost [nlwiki]> ALTER TABLE recentchanges ENGINE=InnoDB, FORCE;
jcrespo triaged this task as High priority.

The repair finished in 10 seconds, faster than I expected without issues (table & index rebuild succesfully, and I restarted replication).

@ABran-WMF When you are back, please check everything is fine (the server, the table, replication, no other strange logs), the right procedure is followed, and specially the host is removed the long downtime and repooled, in that order. If happy, then please resolve the ticket.

Not needed to close the ticket- but I found something really weird- no megacli cmd. Do you know what's the binary for manual checking on newer hosts? Also, learning cycles are disabled on those hosts, too, right?

It's probably the mariadb bug again? We had a spreadsheet.

Please add it there, I am happy to help with the mysql side, but I am not familiar with the DBA's day to day procedures.

spreadsheet is filled, host is back in sync, I'll repool it with the cluster

Mentioned in SAL (#wikimedia-operations) [2024-10-11T12:13:26Z] <arnaudb@cumin1002> dbctl commit (dc=all): 'db2175 (re)pooling @ 1%: T376988', diff saved to https://phabricator.wikimedia.org/P69687 and previous config saved to /var/cache/conftool/dbconfig/20241011-121325-arnaudb.json

repooling is in progress: P69687

Mentioned in SAL (#wikimedia-operations) [2024-10-11T12:28:31Z] <arnaudb@cumin1002> dbctl commit (dc=all): 'db2175 (re)pooling @ 2%: T376988', diff saved to https://phabricator.wikimedia.org/P69688 and previous config saved to /var/cache/conftool/dbconfig/20241011-122830-arnaudb.json

Mentioned in SAL (#wikimedia-operations) [2024-10-11T12:43:36Z] <arnaudb@cumin1002> dbctl commit (dc=all): 'db2175 (re)pooling @ 5%: T376988', diff saved to https://phabricator.wikimedia.org/P69690 and previous config saved to /var/cache/conftool/dbconfig/20241011-124336-arnaudb.json

Mentioned in SAL (#wikimedia-operations) [2024-10-11T12:58:42Z] <arnaudb@cumin1002> dbctl commit (dc=all): 'db2175 (re)pooling @ 10%: T376988', diff saved to https://phabricator.wikimedia.org/P69691 and previous config saved to /var/cache/conftool/dbconfig/20241011-125841-arnaudb.json

Mentioned in SAL (#wikimedia-operations) [2024-10-11T13:13:47Z] <arnaudb@cumin1002> dbctl commit (dc=all): 'db2175 (re)pooling @ 25%: T376988', diff saved to https://phabricator.wikimedia.org/P69692 and previous config saved to /var/cache/conftool/dbconfig/20241011-131347-arnaudb.json

Mentioned in SAL (#wikimedia-operations) [2024-10-11T13:28:53Z] <arnaudb@cumin1002> dbctl commit (dc=all): 'db2175 (re)pooling @ 50%: T376988', diff saved to https://phabricator.wikimedia.org/P69693 and previous config saved to /var/cache/conftool/dbconfig/20241011-132852-arnaudb.json

Mentioned in SAL (#wikimedia-operations) [2024-10-11T13:43:59Z] <arnaudb@cumin1002> dbctl commit (dc=all): 'db2175 (re)pooling @ 75%: T376988', diff saved to https://phabricator.wikimedia.org/P69694 and previous config saved to /var/cache/conftool/dbconfig/20241011-134357-arnaudb.json

Mentioned in SAL (#wikimedia-operations) [2024-10-11T13:59:04Z] <arnaudb@cumin1002> dbctl commit (dc=all): 'db2175 (re)pooling @ 100%: T376988', diff saved to https://phabricator.wikimedia.org/P69695 and previous config saved to /var/cache/conftool/dbconfig/20241011-135903-arnaudb.json