Description of problem: glusterfs core dumps upon rebalance operation Version-Release number of selected component (if applicable): Mainline How reproducible: often Steps to Reproduce: 1. create a distribute volume 2. fill up with some files 3. add another brick and initiate rebalance Actual results: rebalance fails dumping the core Expected results: Additional info: #0 __pthread_mutex_lock (mutex=0x38) at pthread_mutex_lock.c:50 #1 0x00007f24e04a24b1 in client3_1_fremovexattr (frame=0x7f24e3f65b0c, this=0x12ba660, data=0x7f24d70c5970) at client3_1-fops.c:4642 #2 0x00007f24e0488208 in client_fremovexattr (frame=0x7f24e3f65b0c, this=0x12ba660, fd=0x7f24d8b86088, name=0x7f24e026f15a "trusted.glusterfs.dht.linkto") at client.c:1448 #3 0x00007f24e59bf5fd in syncop_fremovexattr (subvol=0x12ba660, fd=0x7f24d8b86088, name=0x7f24e026f15a "trusted.glusterfs.dht.linkto") at syncop.c:498 #4 0x00007f24e02319c9 in dht_migrate_file (this=0x12bbea0, loc=0x12c96b8, from=0x12b9380, to=0x12ba660, flag=1) at dht-rebalance.c:692 #5 0x00007f24e0231cb7 in rebalance_task (data=0x7f24e3f64aec) at dht-rebalance.c:739 #6 0x00007f24e59bd11e in synctask_wrap (old_task=0x12c9d40) at syncop.c:109 #7 0x00007f24e4dbb1a0 in ?? () from /lib/x86_64-linux-gnu/libc.so.6 #8 0x0000000000000000 in ?? () ============================================================================ [2012-02-03 13:29:52.785134] E [stat-prefetch.c:691:sp_remove_caches_from_all_fds_opened] (-->/usr/local/lib/glusterfs/3git/xlator/mount/fuse.so(fuse_setxattr_resume+0x892) [0x7f24e2bd7aa3] (-->/usr/local/lib/glusterfs/3git/xlator/debug/io-stats.so(io_stats_setxattr+0x2cf) [0x7f24db593526] (-->/usr/local/lib/glusterfs/3git/xlator/performance/stat-prefetch.so(sp_setxattr+0x138) [0x7f24db7b4d17]))) 0-test-stat-prefetch: invalid argument: inode > [2012-02-03 13:29:52.785202] I [dht-common.c:2237:dht_setxattr] 0-test-dht: fixing the layout of / > [2012-02-03 13:29:52.802398] I [dht-rebalance.c:510:dht_migrate_file] 0-test-dht: /here: attempting to move from test-client-0 to test-client-1 > pending frames: > > frame : type(1) op(SETXATTR) > frame : type(1) op(SETXATTR) > > patchset: git://git.gluster.com/glusterfs.git > signal received: 11 > time of crash: 2012-02-03 13:29:53 > configuration details: > argp 1 > backtrace 1 > dlfcn 1 > fdatasync 1 > libpthread 1 > llistxattr 1 > setfsid 1 > spinlock 1 > epoll.h 1 > xattr.h 1
gdb) f 1 #1 0x00007f24e04a24b1 in client3_1_fremovexattr (frame=0x7f24e3f65b0c, this=0x12ba660, data=0x7f24d70c5970) at client3_1-fops.c:4644 4644 CLIENT_GET_REMOTE_FD(conf, args->fd, remote_fd, op_errno, unwind); (gdb) p conf $1 = (clnt_conf_t *) 0x0 conf is NULL
*** Bug 787182 has been marked as a duplicate of this bug. ***
CHANGE: http://review.gluster.com/2712 (protocol/client: assign the right value to 'conf' before de-refing it) merged in master by Vijay Bellur (vijay)
works fine