[MDEV-9286] XtraDB crash while replicating from 5.1.42 to 10.1.9 Created: 2015-12-14  Updated: 2020-08-25  Resolved: 2017-06-26

Status: Closed
Project: MariaDB Server
Component/s: Storage Engine - XtraDB
Affects Version/s: 10.1.9
Fix Version/s: N/A

Type: Bug Priority: Major
Reporter: Hartmut Holzgraefe Assignee: Jan Lindström (Inactive)
Resolution: Cannot Reproduce Votes: 0
Labels: need_feedback
Environment:

Linux



 Description   

This looks a bit similar to MDEV-9040 and friends as it also happens in the/a purge thread, but crashes in a different way.

The crashing system is a slave to a 5.1.42 master, a MariaDB 10.0.15 slave to the same master works without problems.

Thread backtrace shows that the actual crash happend on free(), so earlier memory corruption seems to be the cause:

#0  0x00007f6ca2ee0f8c in pthread_kill () from /lib/x86_64-linux-gnu/libpthread.so.0
#1  0x0000000000760dc1 in handle_fatal_signal (sig=11) at /home/buildbot/buildbot/build/sql/signal_handler.cc:273
#2  <signal handler called>
#3  0x00007f6ca28b6496 in ?? () from /usr/lib/libjemalloc.so.1
#4  0x00007f6ca28b77f7 in ?? () from /usr/lib/libjemalloc.so.1
#5  0x00007f6ca28b7b9b in ?? () from /usr/lib/libjemalloc.so.1
#6  0x00007f6ca28b4132 in free () from /usr/lib/libjemalloc.so.1
#7  0x0000000000955f65 in mem_area_free (ptr=0x7f6c2d81e100, pool=0x7f6ca0d40400) at /home/buildbot/buildbot/build/storage/xtradb/mem/mem0pool.cc:519
#8  0x000000000095579f in mem_heap_block_free (heap=0x7f6c2d81e100, block=0x7f6c2d81e100) at /home/buildbot/buildbot/build/storage/xtradb/mem/mem0mem.cc:521
#9  0x00000000009b11b2 in row_purge (thr=<optimized out>, undo_rec=<optimized out>, node=<optimized out>) at /home/buildbot/buildbot/build/storage/xtradb/include/mem0mem.ic:510
#10 row_purge_step (thr=0x7f6ca0c16f30) at /home/buildbot/buildbot/build/storage/xtradb/row/row0purge.cc:992
#11 0x000000000097c2d4 in que_thr_step (thr=<optimized out>) at /home/buildbot/buildbot/build/storage/xtradb/que/que0que.cc:1089
#12 que_run_threads_low (thr=<optimized out>) at /home/buildbot/buildbot/build/storage/xtradb/que/que0que.cc:1151
#13 que_run_threads (thr=0x7f6ca0c16f30) at /home/buildbot/buildbot/build/storage/xtradb/que/que0que.cc:1192
#14 0x00000000009de384 in trx_purge (n_purge_threads=12155, batch_size=300, truncate=false) at /home/buildbot/buildbot/build/storage/xtradb/trx/trx0purge.cc:1251
#15 0x00000000009c6d67 in srv_do_purge (n_threads=1, n_total_purged=0x7f6c2e7fde78) at /home/buildbot/buildbot/build/storage/xtradb/srv/srv0srv.cc:3436
#16 0x00000000009cd80a in srv_purge_coordinator_thread (arg=<optimized out>) at /home/buildbot/buildbot/build/storage/xtradb/srv/srv0srv.cc:3622
#17 0x00007f6ca2edbe9a in start_thread () from /lib/x86_64-linux-gnu/libpthread.so.0
#18 0x00007f6ca1bac38d in clone () from /lib/x86_64-linux-gnu/libc.so.6
#19 0x0000000000000000 in ?? ()



 Comments   
Comment by Jan Lindström (Inactive) [ 2015-12-15 ]

Could we have my.cnf file and full unedited error log? Does not look related to MDEV-9040 at least on surface.

Comment by Sergei Golubchik [ 2017-06-26 ]

No reply for 18 months. Still an issue?

Generated at Thu Feb 08 07:33:32 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.