Have you tried turning off the gluster to see if problem goes away? Can you look at sar reports during the time you were having issues with CPU, memory, processes? What else runs on that box other than gluster? How many processes you have? What do you see in /var/log/messages? It's very difficult to guess what might be wrong with very little detail you have given 2011/5/31 ???(yongjoon kong)/Cloud Computing ???? <andrew.kong at sk.com>: > > > In my case, one of the brick( I use distributed replication) is gone in > zombie status, > > > > The file system is just like freezing > > > > For ping test with bricks, it looks like it's up and running ( Using gluster > brick status, it looks like connected ) > > > > But when I try to log into that brick thru sshd, it failed with time out. > > > > > > In that situation, Gluster is freezing. > > > > Every volume is going to be in read-only mode. > > > > I tried everything to turn it back to normal. But nothing works except > rebooting the zombie brick it. > > > > I really doubt there's some huge issue on brick membership management. > > > > > > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://gluster.org/cgi-bin/mailman/listinfo/gluster-users > >
