Bug 849136

+++ This bug was initially created as a clone of Bug #824533 +++

Description of problem: While deleting files from fuse mount, I got this crash.

Volume Name: rdma
Type: Distributed-Replicate
Volume ID: ac8032f9-d169-4992-9288-393a99122227
Status: Started
Number of Bricks: 2 x 2 = 4
Transport-type: rdma
Bricks:
Brick1: 10.16.157.95:/home/s0
Brick2: 10.16.157.97:/home/s0
Brick3: 10.16.157.99:/home/s0
Brick4: 10.16.157.101:/home/s0


Version-Release number of selected component (if applicable): 3.3.0qa42


How reproducible: Consistently


Additional info:

 gdb glusterfs -c /core.11262

(gdb) bt
#0  0x0000003fe0c32885 in raise (sig=6) at ../nptl/sysdeps/unix/sysv/linux/raise.c:64
#1  0x0000003fe0c34065 in abort () at abort.c:92
#2  0x0000003fe0c2b9fe in __assert_fail_base (fmt=<value optimized out>, assertion=0x7fd050f77628 "0", file=0x7fd050f77477 "inode.c", 
    line=<value optimized out>, function=<value optimized out>) at assert.c:96
#3  0x0000003fe0c2bac0 in __assert_fail (assertion=0x7fd050f77628 "0", file=0x7fd050f77477 "inode.c", line=1090, 
    function=0x7fd050f777a0 "__inode_path") at assert.c:105
#4  0x00007fd050f33d6e in __inode_path (inode=0x7fd040991bdc, name=0x0, bufp=0x7fd041d80808) at inode.c:1090
#5  0x00007fd050f34156 in inode_path (inode=0x7fd040991bdc, name=0x0, bufp=0x7fd041d80808) at inode.c:1191
#6  0x00007fd04c774aef in protocol_client_reopendir (this=0x130e700, fdctx=0x7fd03c029940) at client-handshake.c:1096
#7  0x00007fd04c77532a in client_post_handshake (frame=0x7fd04fb4f84c, this=0x130e700) at client-handshake.c:1281
#8  0x00007fd04c775b6a in client_setvolume_cbk (req=0x7fd04126a1c4, iov=0x7fd04126a204, count=1, myframe=0x7fd04fb4f84c)
    at client-handshake.c:1439
#9  0x00007fd050cf6a38 in rpc_clnt_handle_reply (clnt=0x2997e80, pollin=0x7fd03c026550) at rpc-clnt.c:788
#10 0x00007fd050cf6dd5 in rpc_clnt_notify (trans=0x2993cc0, mydata=0x2997eb0, event=RPC_TRANSPORT_MSG_RECEIVED, data=0x7fd03c026550)
    at rpc-clnt.c:907
#11 0x00007fd050cf2eb8 in rpc_transport_notify (this=0x2993cc0, event=RPC_TRANSPORT_MSG_RECEIVED, data=0x7fd03c026550)
    at rpc-transport.c:489
#12 0x00007fd04531c760 in gf_rdma_pollin_notify (peer=0x299b198, post=0x220b2c0) at rdma.c:3100
#13 0x00007fd04531cadc in gf_rdma_recv_reply (peer=0x299b198, post=0x220b2c0) at rdma.c:3187
#14 0x00007fd04531ce1f in gf_rdma_process_recv (peer=0x299b198, wc=0x7fd041d80e40) at rdma.c:3277
#15 0x00007fd04531d0f0 in gf_rdma_recv_completion_proc (data=0x13058f0) at rdma.c:3362
#16 0x0000003fe10077f1 in start_thread (arg=0x7fd041d81700) at pthread_create.c:301
#17 0x0000003fe0ce570d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:115

--- Additional comment from amarts on 2012-07-11 07:07:55 EDT ---

mostly happened because of RDMA disconnects (and also the 'lock-self-heal' was on).