(gdb) bt #0 0x00007fd7ade49659 in io_stats_writev_cbk (frame=0x7fd7afc3f8e8, cookie=0x7fd7afc3f96c, this=0x692128, op_ret=-1, op_errno=11, prebuf=0x7fd79c02dab0, postbuf=0x7fd79c02db20) at io-stats.c:521 #1 0x00007fd7ae05cee0 in qr_writev_cbk (frame=0x7fd7afc3f96c, cookie=0x7fd7afc3ff9c, this=0x690f48, op_ret=-1, op_errno=11, prebuf=0x7fd79c02dab0, postbuf=0x7fd79c02db20) at quick-read.c:1163 #2 0x00007fd7ae26ff2f in ioc_writev_cbk (frame=0x7fd7afc3ff9c, cookie=0x7fd7afc40c80, this=0x68fe08, op_ret=-1, op_errno=11, prebuf=0x7fd79c02dab0, postbuf=0x7fd79c02db20) at io-cache.c:1186 #3 0x00007fd7ae47ec42 in ra_writev_cbk (frame=0x7fd7afc40c80, cookie=0x7fd7afc428dc, this=0x68ebe8, op_ret=-1, op_errno=11, prebuf=0x7fd79c02dab0, postbuf=0x7fd79c02db20) at read-ahead.c:628 #4 0x00007fd7ae68d09b in wb_writev_cbk (frame=0x7fd7afc428dc, cookie=0x7fd7afc42960, this=0x68d9c8, op_ret=-1, op_errno=11, prebuf=0x7fd79c02dab0, postbuf=0x7fd79c02db20) at write-behind.c:1939 #5 0x00007fd7ae8b5ca6 in dht_writev_cbk (frame=0x7fd7afc42960, cookie=0x7fd7afc3d3c8, this=0x68c7b8, op_ret=-1, op_errno=11, prebuf=0x7fd79c02dab0, postbuf=0x7fd79c02db20) at dht-common.c:2251 #6 0x00007fd7aeaeb2d5 in afr_writev_unwind (frame=0x7fd7afa13598, this=0x68aa88) at afr-inode-write.c:72 #7 0x00007fd7aeaeba9c in afr_writev_done (frame=0x7fd7afa13598, this=0x68aa88) at afr-inode-write.c:185 #8 0x00007fd7aeaf8797 in afr_post_blocking_inodelk_cbk (frame=0x7fd7afa13598, this=0x68aa88) at afr-transaction.c:907 #9 0x00007fd7aeb141c4 in afr_unlock_inodelk (frame=0x7fd7afa13598, this=0x68aa88) at afr-lk-common.c:601 #10 0x00007fd7aeb18a61 in afr_unlock (frame=0x7fd7afa13598, this=0x68aa88) at afr-lk-common.c:1666 #11 0x00007fd7aeb1583d in afr_lock_blocking (frame=0x7fd7afa13598, this=0x68aa88, child_index=2) at afr-lk-common.c:973 #12 0x00007fd7aeb14dbc in afr_lock_cbk (frame=0x7fd7afa13598, cookie=0x1, this=0x68aa88, op_ret=-1, op_errno=107) at afr-lk-common.c:756 #13 0x00007fd7aeb14e2f in afr_blocking_inodelk_cbk (frame=0x7fd7afa13598, cookie=0x1, this=0x68aa88, op_ret=-1, op_errno=107) at afr-lk-common.c:770 #14 0x00007fd7aed4eff0 in client3_1_finodelk_cbk (req=0x7fd7a91fb6e4, iov=0x7fd7af96cde0, count=1, myframe=0x7fd7afc3f444) at client3_1-fops.c:1084 #15 0x00007fd7b1062bea in saved_frames_unwind (saved_frames=0x6a1638) at rpc-clnt.c:342 #16 0x00007fd7b1062ca2 in saved_frames_destroy (frames=0x6a1638) at rpc-clnt.c:358 #17 0x00007fd7b1063207 in rpc_clnt_connection_cleanup (conn=0x6a22a8) at rpc-clnt.c:506 #18 0x00007fd7b1065657 in rpc_clnt_destroy (rpc=0x6a2278) at rpc-clnt.c:1479 ---Type <return> to continue, or q <return> to quit--- #19 0x00007fd7b106573d in rpc_clnt_unref (rpc=0x6a2278) at rpc-clnt.c:1507 #20 0x00007fd7aed490a9 in fini (this=0x687f68) at client.c:2036 #21 0x000000000040440f in cleanup_and_exit () #22 0x000000000040598f in glusterfs_sigwaiter () #23 0x00007fd7b0a1f85a in start_thread () from /lib64/libpthread.so.0 #24 0x00007fd7b078922d in clone () from /lib64/libc.so.6 #25 0x0000000000000000 in ?? () (gdb) p *frame $2 = {root = 0x7fd7afa10c18, parent = 0x7fd7afa10ca0, next = 0x0, prev = 0x7fd7afc3f96c, local = 0x0, this = 0x692128, ret = 0x7fd7adc005d6 <nfs_fop_writev_cbk>, ref_count = 0, lock = 1, cookie = 0x692128, complete = _gf_false, op = GF_FOP_NULL, begin = {tv_sec = 0, tv_usec = 0}, end = {tv_sec = 0, tv_usec = 0}} (gdb)
This happened in 'cleanup_and_exit()' path, ie, a SIGTERM was issued to the process from glusterd anyways. Hence its not considered a blocker at the moment. There are many 'cleanup_and_exit()' path crashes which are known.. we are taking it as part of 3.1.1 release. No need to panic.
(In reply to comment #1) > This happened in 'cleanup_and_exit()' path, ie, a SIGTERM was issued to the > process from glusterd anyways. Hence its not considered a blocker at the > moment. There are many 'cleanup_and_exit()' path crashes which are known.. we > are taking it as part of 3.1.1 release. No need to panic. Is that a rare case? in this scenario when we encountered Platform node was completely hosed resulting in access to mount point hung from nfs client.
*** Bug 1908 has been marked as a duplicate of this bug. ***
*** Bug 1892 has been marked as a duplicate of this bug. ***
PATCH: http://patches.gluster.com/patch/5501 in master (io-stats: handle the case of 'cleanup_and_exit()' properly)
verified with 3.1.1qa9