Bug 1042795

Description of problem:
======================
glusterd crashed on few nodes while creating 2 different CG's simultaneusly on different volumes


Version-Release number of selected component (if applicable):
=============================================================
glusterfs 3.4.0.snap.dec03.2013git

How reproducible:


Steps to Reproduce:
===================
1.Create 4 distribute replicate volumes and start it 

2.Mount the volume and create files 

3.Create a CG with first 2 volumes from the first node and simultaneously create another CG with 3rd and 4th volumes from a different session 

first node:
----------
 gluster snapshot create volume3 volume4 -n CG1 -d "this is new CG"
snapshot create: CG1: consistency group created successfully

[first node : different session ]
----------------------------------
gluster snapshot create volume5 volume6 -n CG2 -d "this is new CG"
snapshot create: CG2: consistency group created successfully

Both the CG s were created successfully , but glusterd crashed on few nodes 

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~`
ESC[?1034h(gdb) bt
#0  0x0000003f7944812c in vfprintf () from /lib64/libc.so.6
#1  0x0000003f795000eb in __fprintf_chk () from /lib64/libc.so.6
#2  0x0000003ad20488e0 in fprintf (fd=60, key=0x7f2bd65b0134 "username", value=0x6c6f760032656d61 <Address 0x6c6f760032656d61 out of bounds>)
    at /usr/include/bits/stdio2.h:98
#3  gf_store_save_value (fd=60, key=0x7f2bd65b0134 "username", value=0x6c6f760032656d61 <Address 0x6c6f760032656d61 out of bounds>) at store.c:325
#4  0x00007f2bd655cf9c in glusterd_volume_exclude_options_write (fd=60, volinfo=0x7f2bcc1ba720) at glusterd-store.c:666
#5  0x00007f2bd655d1aa in glusterd_store_volinfo_write (fd=60, volinfo=0x7f2bcc1ba720) at glusterd-store.c:840
#6  0x00007f2bd655f797 in glusterd_store_perform_snap_volume_store (volinfo=0x7f2bcc00d3e0, snap_volinfo=0x7f2bcc1ba720) at glusterd-store.c:1356
#7  0x00007f2bd655f83f in glusterd_store_snap_volume (volinfo=0x7f2bcc00d3e0, snap=0x7f2bcc1bb770) at glusterd-store.c:1401
#8  0x00007f2bd655fb13 in glusterd_store_perform_snap_store (volinfo=0x7f2bcc00d3e0) at glusterd-store.c:1502
#9  0x00007f2bd659dc19 in glusterd_do_snap (volinfo=0x7f2bcc00d3e0, snapname=0x7f2bc00304e0 "CG3_volume5_snap", dict=0x7f2bd87cd650, 
    cg=<value optimized out>, cg_id=0x7f2bc000b340, volcount=1, snap_volid=0x7f2bc000b2c0 "$B\277\311\033GI\275\201\346\243QX'\376\257 ")
    at glusterd-snapshot.c:3132
#10 0x00007f2bd659f5a5 in glusterd_snapshot_create_commit (dict=<value optimized out>, op_errstr=0x1d94b70, rsp_dict=<value optimized out>)
    at glusterd-snapshot.c:3370
#11 0x00007f2bd659f92e in glusterd_snapshot (dict=0x7f2bd87cd650, op_errstr=0x1d94b70, rsp_dict=0x7f2bd87cdce0) at glusterd-snapshot.c:3588
#12 0x00007f2bd65a358e in gd_mgmt_v3_commit_fn (op=GD_OP_SNAP, dict=0x7f2bd87cd650, op_errstr=0x1d94b70, rsp_dict=0x7f2bd87cdce0)
    at glusterd-mgmt.c:174
#13 0x00007f2bd65a0813 in glusterd_handle_commit_fn (req=0x7f2bd6492818) at glusterd-mgmt-handler.c:546
#14 0x00007f2bd6519d4f in glusterd_big_locked_handler (req=0x7f2bd6492818, actor_fn=0x7f2bd65a05d0 <glusterd_handle_commit_fn>)
    at glusterd-handler.c:78
#15 0x0000003ad204cdd2 in synctask_wrap (old_task=<value optimized out>) at syncop.c:293
#16 0x0000003f79443bf0 in ?? () from /lib64/libc.so.6
#17 0x0000000000000000 in ?? ()
(gdb) f 3
#3  gf_store_save_value (fd=60, key=0x7f2bd65b0134 "username", value=0x6c6f760032656d61 <Address 0x6c6f760032656d61 out of bounds>) at store.c:325
325             ret = fprintf (fp, "%s=%s\n", key, value);
(gdb) p key
$1 = 0x7f2bd65b0134 "username"
(gdb) p value
$2 = 0x6c6f760032656d61 <Address 0x6c6f760032656d61 out of bounds>
(gdb) quit

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Actual results:
===============
glusterd crashed while creating 2 different CG's simultaneusly on different volumes  


Expected results:



Additional info: