Follow up on this. We had been running glusterfs server and client in one process. Yesterday we split up server and client. The "coherency" issue caused by inode reuse is still reproducible, but glusterfs seems to be able to heal itself after a little while.
I'm going to give it another try to whittle out a shell script that can reliably reproduce the issue. I'll file bug once I have that script. Thanks, Lei
_______________________________________________ Gluster-users mailing list [email protected] http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
