[root@brick5 ~]# gluster volume stop perf Stopping volume will make its data inaccessible. Do you want to Continue? (y/n) y Stopping volume perf has been successful [root@brick5 ~]# gluster volume stop perf Stopping volume will make its data inaccessible. Do you want to Continue? (y/n) y Stopping volume perf has been unsuccessful The second attempt to stop should simply say, "volume perf already stopped". Since the CLI knows the state of the volume, there is no reason not to check it before trying to stop the volume.
Vikas, a bug has already been filed for it (2010), and I have sent a patch for it. It says "volume is not in started state" if an attempt is made to stop a volume which is already stopped or just created and has not been started. But 2nd time also it asks for confirmation because the state of a volume is maintained by glusterd and not CLI. Marking this as the duplicate of it. *** This bug has been marked as a duplicate of bug 2010 ***