Created attachment 745607 [details] Client logs Description of problem: When one of the nodes are brought down, the client throws I/O error on reading a file but write succeeds. [root@hamm rep]# echo "Hello world" > abcd [root@hamm rep]# cat abcd cat: abcd: Input/output error [root@hamm rep]# date >> abcd [root@hamm rep]# cat abcd cat: abcd: Input/output error Version-Release number of selected component (if applicable): glusterfs 3.4.0.4rhs built on May 7 2013 13:37:36 How reproducible: Most of the times. Steps to Reproduce: 1. Do some I/O on the clients (Kernel extract, compile, rm -rf ...) 2. Bring down one of the servers. 3. I/O errors are seen. Additional info: Volume Name: bb Type: Distributed-Replicate Volume ID: 7c26fe1e-9b7e-48fa-81ed-6db4bd23b6e8 Status: Started Number of Bricks: 2 x 2 = 4 Transport-type: tcp Bricks: Brick1: tex.lab.eng.blr.redhat.com:/mnt/store/bb Brick2: wingo.lab.eng.blr.redhat.com:/mnt/store/bb Brick3: van.lab.eng.blr.redhat.com:/mnt/store/bb Brick4: mater.lab.eng.blr.redhat.com:/mnt/store/bb Happens when one of the nodes is brought down. Since we have a replica setup here the operations should be seamless. [root@tex glusterfs]# gluster volume status Status of volume: bb Gluster process Port Online Pid ------------------------------------------------------------------------------ Brick tex.lab.eng.blr.redhat.com:/mnt/store/bb 49152 Y 16177 Brick wingo.lab.eng.blr.redhat.com:/mnt/store/bb 49152 Y 11020 Brick mater.lab.eng.blr.redhat.com:/mnt/store/bb 49152 Y 9035 NFS Server on localhost 2049 Y 17601 Self-heal Daemon on localhost N/A Y 16193 NFS Server on e73a2be7-a995-4f09-a1dd-f7cfe935b1cb 2049 Y 12468 Self-heal Daemon on e73a2be7-a995-4f09-a1dd-f7cfe935b1c b N/A Y 11036 NFS Server on f41e0800-64fa-4148-ad4b-2630b1fe170a 2049 Y 10434 Self-heal Daemon on f41e0800-64fa-4148-ad4b-2630b1fe170 a N/A Y 9051 There are no active volume tasks
*** Bug 960591 has been marked as a duplicate of this bug. ***
Verified on: glusterfs 3.4.0.6rhs built on May 10 2013 14:12:00 Unable to reproduce the issue.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHBA-2013-1262.html