Description of problem: 3-way Distributed-Replicate volume... [root@node1 ~]# gluster volume status vol1 Status of volume: vol1 Gluster process TCP Port RDMA Port Online Pid ------------------------------------------------------------------------------ node1:brick1 49165 0 Y 3665 node2:brick2 49164 0 Y 17094 node3:brick3 49154 0 Y 19888 node1:brick4 49172 0 Y 3670 node2:brick5 49171 0 Y 17099 node3:brick6 49155 0 Y 19907 NFS Server on localhost 2049 0 Y 3679 Self-heal Daemon on localhost N/A N/A Y 3689 NFS Server on node3 2049 0 Y 19927 Self-heal Daemon on node3 N/A N/A Y 19935 NFS Server on node2 2049 0 Y 20374 Self-heal Daemon on node2 N/A N/A Y 20382 Task Status of Volume vol1 ------------------------------------------------------------------------------ There are no active volume tasks [root@node1 ~]# If the first node in the cluster is shut down, the one where the volume was created and started from, when I try and display on the two remaining nodes information about the clients connected to the volume I don't get any information back. [root@node2 glusterfs]# gluster volume status vol1 client Client connections for volume vol1 ---------------------------------------------- ---------------------------------------------- [root@node3 glusterfs]# gluster volume status vol1 clients Client connections for volume vol1 ---------------------------------------------- ---------------------------------------------- As soon as the node (node1) is powered back up and everything is up and running again, the clients are visible. ========== Version-Release number of selected component (if applicable): All nodes: gluster-nagios-addons-0.2.5-1.el7rhgs.x86_64 gluster-nagios-common-0.2.3-1.el7rhgs.noarch glusterfs-3.7.5-19.el7rhgs.x86_64 glusterfs-api-3.7.5-19.el7rhgs.x86_64 glusterfs-cli-3.7.5-19.el7rhgs.x86_64 glusterfs-client-xlators-3.7.5-19.el7rhgs.x86_64 glusterfs-fuse-3.7.5-19.el7rhgs.x86_64 glusterfs-geo-replication-3.7.5-19.el7rhgs.x86_64 glusterfs-libs-3.7.5-19.el7rhgs.x86_64 glusterfs-rdma-3.7.5-19.el7rhgs.x86_64 glusterfs-server-3.7.5-19.el7rhgs.x86_64 python-gluster-3.7.5-19.el7rhgs.noarch samba-vfs-glusterfs-4.2.4-13.el7rhgs.x86_64 vdsm-gluster-4.16.30-1.3.el7rhgs.noarch ========== How reproducible: Completely ========== Steps to Reproduce: See description
RCA: In CLI side the response dictionary is parsed assuming all the bricks to be up. Since in this case one of the node was brought down client details for the bricks hosted by the same node were not available in the dictionary resulting into a blank output for 'gluster volume status <volname> clients'
Upstream patch http://review.gluster.org/#/c/14842 posted for review
Verified this bug using the the build - glusterfs-3.8.4-2. I am able to see the command "gluster volume status clients" result on the active cluster nodes when one of the cluster node is down. Moving to verified state.
The reason is explained in the patch http://review.gluster.org/#/c/14842 "In cli the response dictionary is parsed assuming all the bricks to be up. If in a given cluster one of the node is down client details for the bricks hosted by the same node are not available in the dictionary resulting into a blank output for 'gluster volume status <volname> clients'"
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. https://rhn.redhat.com/errata/RHSA-2017-0486.html