Portal:Toolforge/Admin/Runbooks/MaintainDBUsersManyErrors
Appearance
The procedures in this runbook require admin permissions to complete.
The MaintainDBUsersManyError happens when maintain-dbusers is having many errors when trying to operate.
Debugging
Maintain-dbusers lives in one of the cloudcontrol nodes, to find out which:
02:41 PM ~/Work/wikimedia/operations-puppet (production|✔)
dcaro@acme$ git grep maintain_dbusers_primary:
hieradata/common.yaml:wmcs_maintain_dbusers_primary: cloudcontrol1007.eqiad.wmnet
It should be up and running as a systemd service and listening on port 9090:
root@cloudcontrol1007:~# systemctl status maintain-dbusers.service
● maintain-dbusers.service - Maintain wikireplica and toolsdb accounts for tools users, tools and paws users
Loaded: loaded (/lib/systemd/system/maintain-dbusers.service; static)
Active: active (running) since Wed 2025-11-12 13:01:28 UTC; 42min ago
Main PID: 3483570 (maintain-dbuser)
Tasks: 2 (limit: 617442)
Memory: 39.2M
CPU: 1min 43.333s
CGroup: /system.slice/maintain-dbusers.service
└─3483570 /usr/bin/python3 /usr/local/sbin/maintain-dbusers maintain
You can check the logs to see if there were any errors:
root@cloudcontrol1007:~# journalctl -u maintain-dbusers.service
...
You can try to restart it, or run it manually to see why it's getting stuck (might be LDAP hanging, as it has to fetch all the users from there).
Common issues
Related information
- Portal:Data_Services/Admin/Wiki_Replicas#Account_management_(maintain-dbusers)
- Portal:Toolforge/Admin#Regenerate_replica.my.cnf
Support contacts
Old incidents
- Add here any new incidents
