testing with "f3c0281d9b4654888a5525b443f1e6090d874725" . created a dht volume - Brick1#gluster volume info Volume Name: bug Type: Distribute Status: created Number of Bricks: 2 Transport-type: tcp Bricks: Brick1: 10.1.12.170:/new1 Brick2: 10.1.12.173:/new1 Then stopped the volume and killed glusterd on both brick1 and 2. Now started glusterd only on brick1 and did volume start. Brick1 shows status as "started"- Volume Name: bug Type: Distribute Status: Started Number of Bricks: 2 Transport-type: tcp Bricks: Brick1: 10.1.12.170:/new1 Brick2: 10.1.12.173:/new1 Now started glusterd on Brick2,it shows the status as "stopped" Brick2# gluster volume info Volume Name: bug Type: Distribute Status: Stopped Number of Bricks: 2 Transport-type: tcp Bricks: Brick1: 10.1.12.170:/new1 Brick2: 10.1.12.173:/new1 You have new mail in /var/spool/mail/root
To properly resolve this bug, glusterd needs to be synced with other glusterds in the cluster whenever it is started. This causes a considerable overhead. The inconsistent status problem can be fixed by using the "force" option while starting or stopping a volume.
Let us keep this open as an enhancement
Closing as wontfix. File a new bug if this is required.