Bug 802232

Description of problem:
#0  0x0000003278632885 in raise () from /lib64/libc.so.6
Missing separate debuginfos, use: debuginfo-install glibc-2.12-1.47.el6.x86_64 libgcc-4.4.6-3.el6.x86_64
(gdb) bt
#0  0x0000003278632885 in raise () from /lib64/libc.so.6
#1  0x0000003278634065 in abort () from /lib64/libc.so.6
#2  0x000000327862b9fe in __assert_fail_base () from /lib64/libc.so.6
#3  0x000000327862bac0 in __assert_fail () from /lib64/libc.so.6
#4  0x00007f398c29810f in __gf_free (free_ptr=0x1d59c20) at mem-pool.c:273
#5  0x00007f398c25d24e in data_destroy (data=0x1d59c60) at dict.c:144
#6  0x00007f398c25dff9 in data_unref (this=0x1d59c60) at dict.c:492
#7  0x00007f398c25d791 in _dict_set (this=0x1d59910, key=0x7f3988d3ee3b "features.limit-usage", value=0x1da9ad0) at dict.c:269
#8  0x00007f398c25d9b6 in dict_set (this=0x1d59910, key=0x7f3988d3ee3b "features.limit-usage", value=0x1da9ad0) at dict.c:324
#9  0x00007f398c26145e in dict_set_str (this=0x1d59910, key=0x7f3988d3ee3b "features.limit-usage", str=0x1db2490 "/:3GB")
    at dict.c:2116
#10 0x00007f3988d13024 in glusterd_quota_limit_usage (volinfo=0x1d59410, dict=0x1db2420, op_errstr=0x7fffa9315d20)
    at glusterd-quota.c:593
#11 0x00007f3988d137b2 in glusterd_op_quota (dict=0x1db2420, op_errstr=0x7fffa9315d20) at glusterd-quota.c:724
#12 0x00007f3988ce155b in glusterd_op_commit_perform (op=GD_OP_QUOTA, dict=0x1db2420, op_errstr=0x7fffa9315d20, rsp_dict=0x1db2520)
    at glusterd-op-sm.c:2486
#13 0x00007f3988ce0d4d in glusterd_op_ac_commit_op (event=0x1da9bd0, ctx=0x1d56930) at glusterd-op-sm.c:2256
#14 0x00007f3988ce4681 in glusterd_op_sm () at glusterd-op-sm.c:3864
#15 0x00007f3988cd110d in glusterd_handle_commit_op (req=0x7f3988c3e04c) at glusterd-handler.c:651
#16 0x00007f398c0370b9 in rpcsvc_handle_rpc_call (svc=0x1d44cb0, trans=0x1d48ce0, msg=0x1d45390) at rpcsvc.c:514
#17 0x00007f398c03745c in rpcsvc_notify (trans=0x1d48ce0, mydata=0x1d44cb0, event=RPC_TRANSPORT_MSG_RECEIVED, data=0x1d45390)
    at rpcsvc.c:610
#18 0x00007f398c03cdb8 in rpc_transport_notify (this=0x1d48ce0, event=RPC_TRANSPORT_MSG_RECEIVED, data=0x1d45390)
    at rpc-transport.c:498
#19 0x00007f3988a33270 in socket_event_poll_in (this=0x1d48ce0) at socket.c:1686
#20 0x00007f3988a337f4 in socket_event_handler (fd=5, idx=1, data=0x1d48ce0, poll_in=1, poll_out=0, poll_err=0) at socket.c:1801
#21 0x00007f398c29707c in event_dispatch_epoll_handler (event_pool=0x1d39290, events=0x1d480a0, i=0) at event.c:794
#22 0x00007f398c29729f in event_dispatch_epoll (event_pool=0x1d39290) at event.c:856
#23 0x00007f398c29762a in event_dispatch (event_pool=0x1d39290) at event.c:956
---Type <return> to continue, or q <return> to quit---
#24 0x0000000000407dbd in main (argc=2, argv=0x7fffa9316368) at glusterfsd.c:1611
(gdb) 


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

3.3.0qa26
How reproducible:
happened once

Steps to Reproduce:
1. create a distribute-replicate volume, the cluster is having 4 nodes and volume exists on two nodes only, lets say nodes ( nodeA, nodeB, nodeC, nodeD), and volume resides on nodeA and nodeB

2. nfs mount the volume from nodeC to a client

3. try twiking the quota limit from nodeA or nodeB
  
Actual results:

the crash seen with gluster_limit_usage  and crash seen on both nodeC and nodeD
not on nodeA and nodeB

Expected results:
creash is unexpected

Additional info: