At 17:00 UTC, a hypervisor (hv-mtl2-012) stopped responding. The on-call team got an alert and starting the investigation. It seems that the hypervisor just rebooted itself.
We are trying to find the reason and making sure that all the services on that server restarted correctly.
UPDATE 17:24 UTC: the team just finished checking all the services: they are now up and running.
update: OVHCloud’s status confirms what we saw (server rebooting for no reason):
The problem impacts other servers (not ours) as well. Fortunately for us, we made sure to avoid choosing our OVH servers in the same racks. We’ll wait for the result of their investigation.
UPDATE 2024-09-05 08:55 UTC: The incident has been resolved on OVH side.