[Gluster-users] one brick one volume process dies?

[lejeczek]

hi everyone

I have 3-peer cluster with all vols in replica mode, 9 vols.
What I see, unfortunately, is one brick fails in one vol, 
when it happens it's always the same vol on the same brick.
Command: gluster vol status $vol - would show brick not online.
Restarting glusterd with systemclt does not help, only 
system reboot seem to help, until it happens, next time.

How to troubleshoot this weird misbehaviour?
many thanks, L.

.