Description of problem: Hi, I built a disperse volume with 4+1 configuration between five nodes, NFS mounted this volume on a folder. Then did some I/O in this folder. I shutdown a node in the process of I/O, then I/O error happened. There are some xattrs in Additional Info, the version of Node 1 is different from the others. What kind of reasons would cause this result? Version-Release number of selected component (if applicable): Gluster 4.1.8 Additional info: I fetched ec xattrs on four nodes (except the node which is shutdown), the version of Node 1 is different from the other nodes. Node 1: trusted.ec.dirty=0x000000000000002b000000000000002b trusted.ec.version=0x0000000000005d210000000000005d23 Node 2: trusted.ec.dirty=0x000000000000002b000000000000002b trusted.ec.version=0x0000000000005d250000000000005d27 Node 3: trusted.ec.dirty=0x000000000000002b000000000000002b trusted.ec.version=0x0000000000005d250000000000005d27 Node 4: trusted.ec.dirty=0x000000000000002b000000000000002b trusted.ec.version=0x0000000000005d250000000000005d27 Node 5: (Shutdown)
This bug is moved to https://github.com/gluster/glusterfs/issues/849, and will be tracked there from now on. Visit GitHub issues URL for further details