Steps to replicate the issue (include links if applicable):
My Bagmala2 tool was used as an experiment to move a toolforge DB to trove. The host name I received was:
oxztsldqokc.svc.trove.eqiad1.wikimedia.cloud
What happens?:
Since a few days ago, I can no longer connect to the host. Neither can the (PHP) web-tool.
What should have happened instead?:
I should be able to connect to the database.
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Open | BUG REPORT | None | T329949 [Cloud VPS] Trove dbs do not restart after a hypervisor restart | ||
| Resolved | BUG REPORT | taavi | T329853 Trove database vanished |
I suspect it had got shut down during the Ceph outage a few days back (T329535). I've restarted it from Horizon (select the baglama2 project, then Database → Instances from the sidebar and on the instance Actions → Restart instance) and it seems to be back up. Next time please send these kinds of support questions to a proper support venue.