Error / Incident
All the ceph monitor nodes are down in the system.
Ceph monitor nodes are the ones acting as entry points for the system, that means that the service is not working and the whole ceph cluster is unusable.
When they go down all the VMs stop having disk access.
This might be the result of an ongoing intervention gone wrong, so first try to communicate with the rest of the team and look for any ongoing actions:
- Ask on irc (#wikimedia-cloud-admin) if someone is taking any action and forgot about it
- Check SAL (or sal.toolforge.org) for host being rebooted/taken down
- Check Netbox to make sure the host is not being decommed or network changes were not applied
- Check puppet for recent changes that might have affected the node
Then you can continue trying to figure out why the hosts are going down and not coming back up:
- Try to ssh to the host
- Try to ssh to the management port (see SRE/Dc-operations/Hardware_Troubleshooting_Runbook)
If you got ssh access, you can continue by checking the logs for the monitor service:
root@cloudcephmon1001:~# systemctl status systemctl status email@example.com
If the ceph cli is working, you can also query the cluster status:
ceph health detail
See also Portal:Cloud_VPS/Admin/Runbooks/CephClusterInError in case that the problem is at the application level.
Add here any issues you encounter.
- SAL (or sal.toolforge.org).
- Puppet merged changes
Communication and support
Support and administration of the WMCS resources is provided by the Wikimedia Foundation Cloud Services team and Wikimedia movement volunteers. Please reach out with questions and join the conversation:
Provide a list of phabricator tasks of previous incidents to be able to more review previous instances of the issue.