ma_pagecache.c:1341: unlink_block: Assertion `block->next_used != ((void *)0)' failed.

[Philip Stoev]

When executing the SMF RQG scenario, Aria asserted as follows:

mysqld: ma_pagecache.c:1341: unlink_block: Assertion `block->next_used != ((void *)0)' failed.

backtrace:

#6 0x000000307b62b945 in __assert_fail (assertion=0xd56b58 "block->next_used != ((void *)0)", file=<value optimized out>, line=1341,
    function=<value optimized out>) at assert.c:81
#7 0x000000000094716b in unlink_block (pagecache=0x1a091a0, block=0x7f5642acbcd8) at ma_pagecache.c:1341
#8 0x000000000094e760 in free_block (pagecache=0x1a091a0, block=0x7f5642acbcd8) at ma_pagecache.c:4120
#9 0x000000000094d20d in pagecache_delete_internal (pagecache=0x1a091a0, block=0x7f5642acbcd8, page_link=0x7f5642fb98d0, flush=1 '\001')
    at ma_pagecache.c:3499
#10 0x000000000094d898 in pagecache_delete (pagecache=0x1a091a0, file=0x7f55a8288978, pageno=1393, lock=PAGECACHE_LOCK_WRITE, flush=1 '\001')
    at ma_pagecache.c:3702
#11 0x000000000094d967 in pagecache_delete_pages (pagecache=0x1a091a0, file=0x7f55a8288978, pageno=1393, page_count=11, lock=PAGECACHE_LOCK_WRITE,
    flush=1 '\001') at ma_pagecache.c:3725
#12 0x000000000099a67a in _ma_bitmap_free_full_pages (info=0x7f55a8288578, extents=0x7f55a805aae8 "q\005", count=1) at ma_bitmap.c:2385
#13 0x0000000000988533 in free_full_pages (info=0x571, row=0xb00000004) at ma_blockrec.c:2471
#14 0x000000000098d465 in _ma_delete_block_record (info=0x7f55a8288578, record=0x7f55a8012790 "\002") at ma_blockrec.c:4256
#15 0x00000000009a3dde in maria_delete (info=0x7f55a8288578, record=0x7f55a8012790 "\002") at ma_delete.c:109
#16 0x0000000000956e5f in ha_maria::delete_row (this=0x7f55a803a640, buf=0x7f55a8012790 "\002") at ha_maria.cc:2090
#17 0x00000000007c2651 in handler::ha_delete_row (this=0x7f55a803a640, buf=0x7f55a8012790 "\002") at handler.cc:4690
#18 0x00000000007405ef in mysql_delete (thd=0x343cf08, table_list=0x7f55bc014a38, conds=0x7f55bc015058, order=0x343f1c8, limit=18446744073709551595,
    options=0, reset_auto_increment=false) at sql_delete.cc:320
#19 0x0000000000684484 in mysql_execute_command (thd=0x343cf08) at sql_parse.cc:3417
#20 0x000000000068c729 in mysql_parse (thd=0x343cf08, rawbuf=0x7f55bc014910 "DELETE FROM smf_themes WHERE id_member IN (2 , 8 , 13)", length=54,
    found_semicolon=0x7f56402d8c70) at sql_parse.cc:6092
#21 0x000000000067e930 in dispatch_command (command=COM_QUERY, thd=0x343cf08, packet=0x343fb09 "DELETE FROM smf_themes WHERE id_member IN (2 , 8 , 13)",
    packet_length=54) at sql_parse.cc:1269
#22 0x000000000067d93b in do_command (thd=0x343cf08) at sql_parse.cc:895
#23 0x000000000067a959 in handle_one_connection (arg=0x343cf08) at sql_connect.cc:1138
#24 0x000000307ba07761 in start_thread (arg=0x7f56402d9710) at pthread_create.c:301
#25 0x000000307b6e150d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:115

bzr version-info:

revision-id: <email address hidden>
date: 2010-09-15 16:01:03 +0300
build-date: 2010-09-20 12:05:51 +0300
revno: 2938
branch-nick: 5.1

rqg command line:

/home/buildbot/randgen/runall.pl \
--queries=1M \
--engine=Maria \
--mysqld=--default-storage-engine=Maria \
--mysqld=--safe-mode \
--mysqld=--sync-sys=0 \
--mysqld=--log-output=file \
--mysqld=--maria_log_purge_type=at_flush \
--reporters=ErrorLog,Backtrace,Recovery,Shutdown \
--duration=240 \
--threads=5 \
--rows=1000 \
--mask-level=0 \
--mysqld=--maria-repair-threads=2 \
--mysqld=--loose-maria-group-commit=hard \
--mysqld=--loose-maria_group_commit_interval=10 \
--mysqld=--maria-checkpoint-interval=1 \
--mysqld=--maria-block-size=4K \
--mask=9312 \
--queries=100000000 \
--mask=9312 \
--seed=1285005403 \
--basedir=. \
--mysqld=--init-file=/home/buildbot/randgen/conf/smf/smf2.sql \
--grammar=conf/smf/smf2.yy

[Michael Widenius (monty)]

This case should be fixed in 5.2 as part of other fixes for pagecache on 2011-06-24

I did run the test in a loop for several hours without any problems., so this bug should be gone.