Description of problem:
Two people were individually complaining that glusterd would silently end. Looking at the logs, there was no indication why.

{all normal start up lines above, including the rest of the confusion causing "Unknown key" list}
[2013-04-11 13:49:24.184786] E [glusterd-store.c:2080:glusterd_store_retrieve_volume] 0-: Unknown key: brick-25
[2013-04-11 13:49:25.152621] E [xlator.c:385:xlator_init] 0-management: Initialization of volume 'management' failed, review your volfile again
[2013-04-11 13:49:25.152738] E [graph.c:294:glusterfs_graph_init] 0-management: initializing translator failed
[2013-04-11 13:49:25.152783] E [graph.c:483:glusterfs_graph_activate] 0-graph: init failed
[2013-04-11 13:49:25.153205] W [glusterfsd.c:831:cleanup_and_exit] (-->/usr/sbin/glusterd(main+0x574) [0x407404] (-->/usr/sbin/glusterd(glusterfs_volumes_init+0x145) [0x405bb5] (-->/usr/sbin/glusterd(glusterfs_process_volfp+0x198) [0x405a68]))) 0-: received signum (0), shutting down

Version-Release number of selected component (if applicable):
3.3.1

How reproducible:
Always

Steps to Reproduce:
1. Truncate a file in /var/lib/glusterd/peers
2. start glusterd
  
Actual results:
glusterd silently fails with no indication what's wrong.

Expected results:
glusterd should throw an error explaining why it's dying.

Additional info:
I don't know how the peer files can get into an empty state. I believe this has something to do with adding new peers.