First reboot of omero* instances after updates may fail
Opened this issue · 2 comments
manics commented
As part of the standard deployment all servers are patched the first time and automatically rebooted if there as a kernel upgrade. If omeroreadwrite and omeroreadonly-N are rebooted simultaneously the omeroreadonly-N may hang since they mount omeroreadwrite with NFS, and when omeroreadwrite goes down the mounts are left hanging and never seem to recover, even after the server comes back up. The current solution is to force a hard reboot openstack server reboot --hard omeroreadonly-N
This is related to #68
joshmoore commented
Also seen in the (rather strewn out) deployment of prod47.