Change #1147744 had a related patch set uploaded (by Marostegui; author: Marostegui):

[operations/puppet@production] db1169: Migrate to MariaDb 10.11

https://gerrit.wikimedia.org/r/1147744

Mentioned in SAL (#wikimedia-operations) [2025-05-19T11:23:57Z] <marostegui@cumin1002> dbctl commit (dc=all): 'Depool db1169 T394653', diff saved to https://phabricator.wikimedia.org/P76294 and previous config saved to /var/cache/conftool/dbconfig/20250519-112356-marostegui.json

Change #1147744 merged by Marostegui:

[operations/puppet@production] db1169: Migrate to MariaDb 10.11

https://gerrit.wikimedia.org/r/1147744

Mentioned in SAL (#wikimedia-operations) [2025-05-20T05:00:18Z] <marostegui@cumin1002> dbctl commit (dc=all): 'Depool db2236 T394653', diff saved to https://phabricator.wikimedia.org/P76319 and previous config saved to /var/cache/conftool/dbconfig/20250520-050017-marostegui.json

Mentioned in SAL (#wikimedia-operations) [2025-05-20T05:03:47Z] <marostegui> Install 10.11.13 on db2236 T394653

Installed 10.11.13

So far no issues on db2236 - no regression or performance change after migrating it from 10.11.11

Change #1148206 had a related patch set uploaded (by Marostegui; author: Marostegui):

[operations/software@master] control-mariadb-10.11-bookworm: Bump to 10.11.13

https://gerrit.wikimedia.org/r/1148206

Change #1148206 merged by jenkins-bot:

[operations/software@master] control-mariadb-10.11-bookworm: Bump to 10.11.13

https://gerrit.wikimedia.org/r/1148206

There is now a 29 hour replication lag on the cloud service replicas for databases on s4 and 25 hour replication lag for databases on s2, s6 and s7. https://replag.toolforge.org/

This is therefore now also affecting tools with only daily updates.

In T394653#10842612, @GPSLeo wrote:

There is now a 29 hour replication lag on the cloud service replicas for databases on s4 and 25 hour replication lag for databases on s2, s6 and s7. https://replag.toolforge.org/

This is therefore now also affecting tools with only daily updates.

This has nothing to do with this task, this is because of T394624

I have been testing the problematic release (10.11.12) on db1169 (s1) and so far I am not seeing any issues with it.
As I pointed out in the mariadb ticket, we do not use:

innodb_adaptive_hash_index=on
SERIALIZABLE as isolation level

10.11.12 didn't have any issues in our environment with db1169 - however, not to risk it more, after 24h, I have migrated it to 10.11.13 (version with the fix). I've reported this to MariaDB

Mmm aparently there is https://mariadb.com/kb/en/mariadb-10-11-13-release-notes/ but they never sent an email about it: https://lists.mariadb.org/hyperkitty/list/announce@lists.mariadb.org/