Description of problem:
-----------------------

In a single-node cluster, when remove-brick is in progress, glusterd is killed and then brought back up. Following this, 'gluster volume status' command fails on the node - 

[root@rhs ~]# gluster v status test_dis 
Commit failed on localhost. Please check the log file for more details.

The following errors are seen in the glusterd logs - 

[2013-11-07 03:02:59.984190] I [glusterd-handler.c:3498:__glusterd_handle_status_volume] 0-management: Received status volume req for volume test_dis
[2013-11-07 03:02:59.984708] E [glusterd-op-sm.c:1973:_add_remove_bricks_to_dict] 0-management: Failed to get brick count
[2013-11-07 03:02:59.984737] E [glusterd-op-sm.c:2037:_add_task_to_dict] 0-management: Failed to add remove bricks to dict
[2013-11-07 03:02:59.984753] E [glusterd-op-sm.c:2122:glusterd_aggregate_task_status] 0-management: Failed to add task details to dict
[2013-11-07 03:02:59.984768] E [glusterd-syncop.c:993:gd_commit_op_phase] 0-management: Commit of operation 'Volume Status' failed on localhost    

Version-Release number of selected component (if applicable):
glusterfs 3.4.0.35.1u2rhs

How reproducible:
Always

Steps to Reproduce:
1. Create a distribute volume with two bricks, start it, fuse mount it and create some data on the mount point.
2. Start remove-brick of one of the bricks.
3. While remove-brick is in progress, kill glusterd and start it again.
4. Check volume status - 
# gluster volume status

Actual results:
The command fails with the following message - 

Commit failed on localhost. Please check the log file for more details.

Expected results:
The command should not fail.

Additional info: