[Gluster-devel] crash with xen

[Nathan Stratton]

glusterfs 2.0.4

6 bricks with replicate and distribute

http://share.robotics.net/glusterfs.vol
http://share.robotics.net/glusterfsd.vol
http://share.robotics.net/glusterfs.log

Everything starts fine, then I reboot the 3rd node at 18:54:35. Things 
continue to run (as they should!) and the bad brick is disconnected. The 
server comes back up at 18:56:16 and gluster completes the handshake then 
gets a "polling received on tcp socket (peer: 172.16.0.202:6997) after 
handshake is complete" error and disconnects a few times and then the 
whole share goes offline and all the xen hosts crash. :(

> <>
Nathan Stratton                                CTO, BlinkMind, Inc.
nathan at robotics.net                         nathan at blinkmind.com
http://www.robotics.net                        http://www.blinkmind.com