the internal CEPH storage in use by most internal management services now restarted. Jobs starting since half an hour. Several jobs still running since prior the outage.
Some jobs likely were experiencing issues during outage.
(GPUs still off, security audit pushed forward in time.)