Hide Forgot
The setup was a 2 subvolume replicate with error-gen above replicate and also another one as a subvolume to replicate. Was running dovecot with postal benchmarking tool over the mount point. This seems like a memory corruption gdb backtrace: #0 0x0000003735070454 in malloc_consolidate () from /lib64/libc.so.6 #1 0x0000003735072a1a in _int_malloc () from /lib64/libc.so.6 #2 0x000000373507486d in calloc () from /lib64/libc.so.6 #3 0x00002af44165d0e9 in afr_writev (frame=0xeace7d0, this=0xeab7350, fd=0xeac21b0, vector=0xead3240, count=1, offset=531399, iobref=0xeaddbf0) at afr-inode-write.c:283 #4 0x00002af44143a24b in error_gen_writev (frame=0xeafff50, this=0xeab7c00, fd=0xeac21b0, vector=0xead3240, count=1, off=531399, iobref=0xeaddbf0) at error-gen.c:1130 #5 0x00002af441882e90 in ra_writev (frame=0xeac2660, this=0xeab7ec0, fd=0xeac21b0, vector=0xead3240, count=1, offset=531399, iobref=0xeaddbf0) at read-ahead.c:665 #6 0x00002af441a8b5fe in ioc_writev (frame=0xeac2d00, this=0xeab87b0, fd=0xeac21b0, vector=0xead3240, count=1, offset=531399, iobref=0xeaddbf0) at io-cache.c:1087 #7 0x00002af441c9a499 in qr_writev (frame=0xead28c0, this=0xeab91d0, fd=0xeac21b0, vector=0xead3240, count=1, off=531399, iobref=0xeaddbf0) at quick-read.c:1090 #8 0x00002af441ea1bfb in wb_sync (frame=0x2aaaac00e068, file=0xead1610, winds=0x40b7d7a0) at write-behind.c:481 #9 0x00002af441ea1d5f in wb_do_ops (frame=0x2aaaac00e068, file=0xead1610, winds=0x40b7d7a0, unwinds=<value optimized out>, other_requests=0x40b7d780) at write-behind.c:1644 #10 0x00002af441ea1e91 in wb_process_queue (frame=0x2aaaac00e068, file=0xead1610, flush_all=0 '\000') at write-behind.c:1817 #11 0x00002af441ea4d37 in wb_sync_cbk (frame=0x2aaaac00e068, cookie=<value optimized out>, this=<value optimized out>, op_ret=-1, op_errno=11, prebuf=<value optimized out>, postbuf=0x2aaaac00eea0) at write-behind.c:361 #12 0x00002af441c9524d in qr_writev_cbk (frame=0x2aaaac00e0d0, cookie=<value optimized out>, this=<value optimized out>, op_ret=-1, op_errno=11, prebuf=0x2aaaac00ee10, postbuf=0x2aaaac00eea0) at quick-read.c:996 #13 0x00002af441a8b78f in ioc_writev_cbk (frame=0x2aaaac00e290, cookie=<value optimized out>, this=<value optimized out>, op_ret=-1, op_errno=11, prebuf=0x2aaaac00ee10, postbuf=0x2aaaac00eea0) at io-cache.c:1048 #14 0x00002af441882f6c in ra_writev_cbk (frame=0x2aaaac00e8b0, cookie=<value optimized out>, this=<value optimized out>, op_ret=-1, op_errno=11, prebuf=0x2aaaac00ee10, postbuf=0x2aaaac00eea0) at read-ahead.c:635 #15 0x00002af4414359bd in error_gen_writev_cbk (frame=0x2aaaac00e910, cookie=<value optimized out>, this=<value optimized out>, op_ret=-1, op_errno=11, prebuf=0x2aaaac00ee10, postbuf=0x2aaaac00eea0) at error-gen.c:1104 #16 0x00002af44165b602 in afr_writev_unwind (frame=<value optimized out>, this=<value optimized out>) at afr-inode-write.c:75 #17 0x00002af44165d4d7 in afr_writev_done (frame=0x2aaaac010ba8, this=0xeab7350) at afr-inode-write.c:199 #18 0x00002af441661cc9 in afr_unlock (frame=0x2aaaac010ba8, this=0xeab7350) at afr-transaction.c:551 #19 0x00002af441662893 in afr_lock_rec (frame=0x2aaaac010ba8, this=0xeab7350, child_index=2) at afr-transaction.c:1267 #20 0x00002af441663182 in afr_lock_cbk (frame=0x2aaaac010ba8, cookie=<value optimized out>, this=0xeab7350, op_ret=-1, op_errno=107) at afr-transaction.c:1117 #21 0x00002af441434f1c in error_gen_finodelk_cbk (frame=0xeac37a0, cookie=<value optimized out>, this=<value optimized out>, op_ret=-1, op_errno=107) #22 0x00002af44121944b in client_finodelk_cbk (frame=0xeac99b0, hdr=<value optimized out>, hdrlen=<value optimized out>, iobuf=<value optimized out>) at client-protocol.c:5419 #23 0x00002af44121bc3e in protocol_client_xfer (frame=0xeac99b0, this=<value optimized out>, trans=0xeabc940, type=1, op=31, hdr=0xeaccc50, hdrlen=173, vector=0x0, count=0, iobref=0x0) at client-protocol.c:655 #24 0x00002af4412206fc in client_finodelk (frame=0xeac99b0, this=0xeab5800, volume=0xeab7330 "mirror-0", fd=<value optimized out>, cmd=7, flock=0x40b7de60) at client-protocol.c:3120 #25 0x00002af441438364 in error_gen_finodelk (frame=0xeac37a0, this=0xeab6a90, volume=0xeab7330 "mirror-0", fd=0xeac21b0, cmd=7, lock=0x40b7de60) at error-gen.c:1730 #26 0x00002af441662cd4 in afr_lock_rec (frame=0x2aaaac010ba8, this=0xeab7350, child_index=<value optimized out>) at afr-transaction.c:1299 #27 0x00002af441663182 in afr_lock_cbk (frame=0x2aaaac010ba8, cookie=<value optimized out>, this=0xeab7350, op_ret=-1, op_errno=107) at afr-transaction.c:1117 #28 0x00002af44121944b in client_finodelk_cbk (frame=0x2aaaac010cd0, hdr=<value optimized out>, hdrlen=<value optimized out>, iobuf=<value optimized out>) at client-protocol.c:5419 #29 0x00002af44121abab in call_bail (data=0xeabb840) at client-protocol.c:320 #30 0x00002af440760f3e in gf_timer_proc (ctx=0xeab0010) at timer.c:172 #31 0x0000003735c06617 in start_thread () from /lib64/libpthread.so.0 #32 0x00000037350d3c2d in clone () from /lib64/libc.so.6
Pranith is not able to reproduce the issue. And AFR has undergone some changes from last 8months. Should we keep this bug open?
I am not able to reproduce this. Please feel free to re-open it if you observe it again.