[MDEV-7362] ANALYZE TABLES crash with table-independent-statistics gathering Created: 2014-12-23  Updated: 2015-01-22  Resolved: 2015-01-22

Status: Closed
Project: MariaDB Server
Component/s: Optimizer
Affects Version/s: 10.0.15
Fix Version/s: 10.0.16

Type: Bug Priority: Major
Reporter: Daniel Black Assignee: Vicențiu Ciorbaru
Resolution: Fixed Votes: 0
Labels: None
Environment:

wheezy x86_64

Attachments: Text File bt-full.txt     Text File bt-second.txt     Text File bt.txt     File catinthehat.cnf     File mdev-7362.sql     File my.cnf    
Issue Links:
Relates
relates to MDEV-7380 engine-independent stats SEGV on ANAL... Closed

 Description   

Was running mysqlanalyze --skip-write-binlog --all-databases

Ran though a lot of tables without error. 

....
 
...
 
..
 
 
 
oq_dev_eventum.eventum_search_profile              Engine-independent statistics collected
 
status   : OK
 
oq_dev_eventum.eventum_status                      Engine-independent statistics collected
 
status   : OK
 
oq_dev_eventum.eventum_subscription                Engine-independent statistics collected
 
status   : OK
 
oq_dev_eventum.eventum_subscription_type           Engine-independent statistics collected
 
status   : OK
 
oq_dev_eventum.eventum_support_email               Engine-independent statistics collected
 
status   : OK
 
mysqlanalyze: Got error: 2013: Lost connection to MySQL server during query when executing 'ANALYZE NO_WRITE_TO_BINLOG TABLE ... '

Settings where:

| use_stat_tables | COMPLEMENTARY |
 

Dec 23 03:23:25 catinthehat mysqld: Server version: 10.0.15-MariaDB-1~wheezy-log
 
Dec 23 03:23:25 catinthehat mysqld: key_buffer_size=8388608
 
Dec 23 03:23:25 catinthehat mysqld: read_buffer_size=1048576
 
Dec 23 03:23:25 catinthehat mysqld: max_used_connections=4
 
Dec 23 03:23:25 catinthehat mysqld: max_threads=102
 
Dec 23 03:23:25 catinthehat mysqld: thread_count=4
 
Dec 23 03:23:25 catinthehat mysqld: It is possible that mysqld could use up to
 
Dec 23 03:23:25 catinthehat mysqld: key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 140842 K  bytes of memory
 
Dec 23 03:23:25 catinthehat mysqld: Hope that's ok; if not, decrease some variables in the equation.
 
Dec 23 03:23:25 catinthehat mysqld: 
Dec 23 03:23:25 catinthehat mysqld: Thread pointer: 0x0x7f53f3936008
 
Dec 23 03:23:25 catinthehat mysqld: Attempting backtrace. You can use the following information to find out
 
Dec 23 03:23:25 catinthehat mysqld: where mysqld died. If you see no messages after this, something went
 
Dec 23 03:23:25 catinthehat mysqld: terribly wrong...
 
Dec 23 03:23:25 catinthehat mysqld: stack_bottom = 0x7f5425c73e30 thread_stack 0x48000
 
Dec 23 03:23:25 catinthehat mysqld: /usr/sbin/mysqld(my_print_stacktrace+0x2b)[0x7f5428926b9b]
 
Dec 23 03:23:25 catinthehat mysqld: /usr/sbin/mysqld(handle_fatal_signal+0x422)[0x7f54284b6e92]
 
Dec 23 03:23:25 catinthehat mysqld: /lib/x86_64-linux-gnu/libpthread.so.0(+0xf8d0)[0x7f5427b078d0]
 
Dec 23 03:23:25 catinthehat mysqld: /usr/sbin/mysqld(_Z27update_statistics_for_tableP3THDP5TABLE+0x360)[0x7f54283c8850]
 
Dec 23 03:23:25 catinthehat mysqld: /usr/sbin/mysqld(+0x4f693e)[0x7f542842e93e]
 
Dec 23 03:23:25 catinthehat mysqld: /usr/sbin/mysqld(_ZN21Sql_cmd_analyze_table7executeEP3THD+0xd9)[0x7f5428430079]
 
Dec 23 03:23:25 catinthehat mysqld: /usr/sbin/mysqld(_Z21mysql_execute_commandP3THD+0x1968)[0x7f542835a7f8]
 
Dec 23 03:23:25 catinthehat mysqld: /usr/sbin/mysqld(+0x4275ba)[0x7f542835f5ba]
 
Dec 23 03:23:25 catinthehat mysqld: /usr/sbin/mysqld(_Z16dispatch_command19enum_server_commandP3THDPcj+0x14df)[0x7f5428360b7f]
 
Dec 23 03:23:25 catinthehat mysqld: /usr/sbin/mysqld(_Z24do_handle_one_connectionP3THD+0x47b)[0x7f542842273b]
 
Dec 23 03:23:25 catinthehat mysqld: /usr/sbin/mysqld(handle_one_connection+0x47)[0x7f5428422817]
 
Dec 23 03:23:25 catinthehat mysqld: /usr/sbin/mysqld(+0x6f5d5d)[0x7f542862dd5d]
 
Dec 23 03:23:25 catinthehat mysqld: /lib/x86_64-linux-gnu/libpthread.so.0(+0x80a4)[0x7f5427b000a4]
 
Dec 23 03:23:25 catinthehat mysqld: /lib/x86_64-linux-gnu/libc.so.6(clone+0x6d)[0x7f5426178ccd]
 
Dec 23 03:23:25 catinthehat mysqld: 
Dec 23 03:23:25 catinthehat mysqld: Trying to get some variables.
 
Dec 23 03:23:25 catinthehat mysqld: Some pointers may be invalid and cause the dump to abort.
 
Dec 23 03:23:25 catinthehat mysqld: Query (0x7f540a4c8020): is an invalid pointer
 
Dec 23 03:23:25 catinthehat mysqld: Connection ID (thread ID): 13
 
Dec 23 03:23:25 catinthehat mysqld: Status: NOT_KILLED
 

 mysqlcheck --skip-write-binlog --all-databases 
..
 
(all OK)
 

select * from mysql.table_stats
 
| oq_dev_eventum | eventum_round_robin_user          |           0 |
 
| oq_dev_eventum | eventum_search_profile            |          27 |
 
| oq_dev_eventum | eventum_status                    |          10 |
 
| oq_dev_eventum | eventum_subscription              |        1052 |
 
| oq_dev_eventum | eventum_subscription_type         |        4003 |
 
| oq_dev_eventum | eventum_support_email             |        4535 |
 
| oq_dev_eventum | eventum_support_email_body        |        4534 |
 
| logsleads      | clflog                            |           0 |
 
 
 
 
 
 
 
MariaDB [mysql]> show tables in oq_dev_eventum like 'eventum_%';
 
...
 
| eventum_support_email                |
 
| eventum_support_email_body           |
 
| eventum_time_tracking                |
 
| eventum_time_tracking_category       |
 
| eventum_user                         |
 
+--------------------------------------+
 
69 rows in set (0.00 sec)

After restart with:

| use_stat_tables                           | NEVER  |
 
 
 
mysqlanalyze --skip-write-binlog --all-databases 
...
 
oq_eventum.eventum_status                          OK
 
oq_eventum.eventum_subscription                    OK
 
oq_eventum.eventum_subscription_type               OK
 
oq_eventum.eventum_support_email                   OK
 
oq_eventum.eventum_support_email_body              Table is already up to date
 
oq_eventum.eventum_time_tracking                   Table is already up to date
 
oq_eventum.eventum_time_tracking_category          OK
 
oq_eventum.eventum_user                            OK
 
oq_eventum.oq_customer_contacts                    OK
 
oq_eventum.oq_customers                            OK
 
oq_ourdelta.wp_QuoteRotator                        OK
 
oq_ourdelta.wp_commentmeta                         OK
 
oq_ourdelta.wp_comments                            OK
 
oq_ourdelta.wp_links                               OK
 
...

This is repeatable on this table:

MariaDB [(none)]> set sql_log_bin=0;
 
Query OK, 0 rows affected (0.00 sec)
 
 
 
MariaDB [(none)]> analyze local table oq_dev_eventum.eventum_support_email_body;
 
+-------------------------------------------+---------+----------+-----------------------------+
 
| Table                                     | Op      | Msg_type | Msg_text                    |
 
+-------------------------------------------+---------+----------+-----------------------------+
 
| oq_dev_eventum.eventum_support_email_body | analyze | status   | Table is already up to date |
 
+-------------------------------------------+---------+----------+-----------------------------+
 
1 row in set (0.00 sec)
 
 
 
MariaDB [(none)]> set global use_stat_tables=COMPLEMENTARY;
 
Query OK, 0 rows affected (0.00 sec)
 
 
 
MariaDB [(none)]> analyze local table oq_dev_eventum.eventum_support_email_body;
 
+-------------------------------------------+---------+----------+-----------------------------+
 
| Table                                     | Op      | Msg_type | Msg_text                    |
 
+-------------------------------------------+---------+----------+-----------------------------+
 
| oq_dev_eventum.eventum_support_email_body | analyze | status   | Table is already up to date |
 
+-------------------------------------------+---------+----------+-----------------------------+
 
1 row in set (0.00 sec)
 
 
 
MariaDB [(none)]> analyze local table oq_dev_eventum.eventum_support_email_body PERSISTENT FOR ALL;
 
ERROR 2013 (HY000): Lost connection to MySQL server during query
 

| eventum_support_email_body | CREATE TABLE `eventum_support_email_body` (
 
  `seb_sup_id` int(11) unsigned NOT NULL,
 
  `seb_body` longtext NOT NULL,
 
  `seb_full_email` longtext NOT NULL,
 
  PRIMARY KEY (`seb_sup_id`),
 
  FULLTEXT KEY `ft_support_email` (`seb_body`)
 
) ENGINE=MyISAM DEFAULT CHARSET=utf8 |
 
 
 
MariaDB [(none)]> select count(*) from  oq_dev_eventum.eventum_support_email_body;
 
+----------+
 
| count(*) |
 
+----------+
 
|     4534 |
 
 
 
 
 
MariaDB [(none)]> select length(seb_body) as b from oq_dev_eventum.eventum_support_email_body order by b desc limit 1;
 
+-------+
 
| b     |
 
+-------+
 
| 44175 |
 
 
 
MariaDB [(none)]> select length(seb_full_email) as b from oq_dev_eventum.eventum_support_email_body order by b desc limit 1;
 
+---------+
 
| b       |
 
+---------+
 
| 8937583 |

Running with use_stat_tables=NEVER didn't cause a crash.

If you can repeat this without a copy of oq_dev_eventum.eventum_support_email_body that would be good. If you can't I'll see what I can do.

 Comments   
Comment by Elena Stepanova [ 2014-12-23 ]

Could you please attach your cnf file(s) – or maybe it is already attached to some other bug that you reported?

Comment by Daniel Black [ 2014-12-24 ]

use_stat_tables = COMPLEMENTARY

histogram_size=255 was set at runtime however I could repeat it with the default histogram_size=0.

Comment by Elena Stepanova [ 2014-12-24 ]

Thanks, I'll try to reproduce with your configs (had no luck with default settings).
There is, however, one confusing detail which makes me think I might be missing an important step.

In your steps to reproduce, you do the following, all in the same session:

  • run ANALYZE;
  • SET GLOBAL use_stat_tables = COMPLEMENTARY (I assume it was NEVER before?)
  • run ANALYZE;
  • run ANALYZE .. PERSISTENT FOR ALL => crash.

And you also said that with use_stat_tables=NEVER there is no crash.

But the thing is, if you set the GLOBAL value of use_stat_tables, it does not (should not, at least) have any effect on the current connection. And from the 2nd ANALYZE, we see that it's so indeed – there is nothing about statistics collection; there would have been if statistics was already activated.

Note: It seems that this description https://mariadb.com/kb/en/mariadb/documentation/optimization-and-tuning/query-optimizations/statistics-for-optimizing-queries/engine-independent-table-statistics/ is outdated or wrongly phrased; in fact, simple ANALYZE TABLE already collects statistics if use_stat_tables is anything but NEVER; and ANALYZE TABLE ... PERSISTENT collects statistics even if use_stat_tables is NEVER.

Are you absolutely sure about the point regarding use_stat_tables=NEVER preventing the crash?

Comment by Daniel Black [ 2014-12-24 ]

Sorry there was a new session created after the set global. I did the same mistake here but it does that the default use_stats_tables doesn't crash with ANALYZE TABLE and with it set in a session it does.

The following was running on a trunk developed server running

/home/dan/software_projects/mariadb-trunk/sql/mysqld --defaults-group-suffix=.2 --defaults-file=/home/dan/software_projects/mariadb-trunk/mysql-test/var/my.cnf 

MariaDB [t1]> analyze table eventum_support_email_body;
 
+-------------------------------+---------+----------+-----------------------------+
 
| Table                         | Op      | Msg_type | Msg_text                    |
 
+-------------------------------+---------+----------+-----------------------------+
 
| t1.eventum_support_email_body | analyze | status   | Table is already up to date |
 
+-------------------------------+---------+----------+-----------------------------+
 
1 row in set (0.00 sec)
 
 
 
MariaDB [t1]> SET GLOBAL use_stat_tables = COMPLEMENTARY;
 
Query OK, 0 rows affected (0.00 sec)
 
 
 
MariaDB [t1]> analyze table eventum_support_email_body;
 
+-------------------------------+---------+----------+-----------------------------+
 
| Table                         | Op      | Msg_type | Msg_text                    |
 
+-------------------------------+---------+----------+-----------------------------+
 
| t1.eventum_support_email_body | analyze | status   | Table is already up to date |
 
+-------------------------------+---------+----------+-----------------------------+
 
1 row in set (0.00 sec)
 
 
 
MariaDB [t1]> SET use_stat_tables = COMPLEMENTARY;
 
Query OK, 0 rows affected (0.00 sec)
 
 
 
MariaDB [t1]> analyze table eventum_support_email_body;
 
ERROR 2013 (HY000): Lost connection to MySQL server during query
 
MariaDB [t1]> SET use_stat_tables = COMPLEMENTARY;
 
ERROR 2006 (HY000): MySQL server has gone away
 
No connection. Trying to reconnect...
 
Connection id:    2
 
Current database: t1
 
 
 
Query OK, 0 rows affected (0.05 sec)
 
 
 
MariaDB [t1]> analyze table eventum_support_email_body;

gdb caught the process at this point which is the attached bt.txt

Documentation edits are my fault. I'll test and revert.

Comment by Daniel Black [ 2014-12-26 ]

Server running from mysql-test directory;

mysql-test $ /home/dan/software_projects/mariadb-trunk/sql/mysqld --defaults-group-suffix=.2 --defaults-file=/home/dan/software_projects/mariadb-trunk/mysql-test/var/my.cnf
 
 
 
141226 13:05:52 [Warning] You need to use --log-bin to make --log-slave-updates work.
 
141226 13:05:52 [Note] InnoDB: Using mutexes to ref count buffer pool pages
 
141226 13:05:52 [Note] InnoDB: The InnoDB memory heap is disabled
 
141226 13:05:52 [Note] InnoDB: Mutexes and rw_locks use GCC atomic builtins
 
141226 13:05:52 [Note] InnoDB: Memory barrier is not used
 
141226 13:05:52 [Note] InnoDB: Compressed tables use zlib 1.2.8
 
141226 13:05:52 [Note] InnoDB: Using Linux native AIO
 
141226 13:05:52 [Note] InnoDB: Using CPU crc32 instructions
 
141226 13:05:52 [Note] InnoDB: Initializing buffer pool, size = 8.0M
 
141226 13:05:52 [Note] InnoDB: Completed initialization of buffer pool
 
141226 13:05:52 [Note] InnoDB: Highest supported file format is Barracuda.
 
141226 13:05:52 [Note] InnoDB: 128 rollback segment(s) are active.
 
141226 13:05:52 [Note] InnoDB: Waiting for purge to start
 
141226 13:05:52 [Note] InnoDB:  Percona XtraDB (http://www.percona.com) 5.6.21-70.0 started; log sequence number 1937518
 
141226 13:05:52 [Note] Plugin 'XTRADB_READ_VIEW' is disabled.
 
141226 13:05:52 [Note] Plugin 'XTRADB_INTERNAL_HASH_TABLES' is disabled.
 
141226 13:05:52 [Note] Plugin 'XTRADB_RSEG' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_TRX' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_LOCK_WAITS' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_CMP_RESET' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_CMPMEM' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_CMPMEM_RESET' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_CMP_PER_INDEX_RESET' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_BUFFER_PAGE' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_BUFFER_PAGE_LRU' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_BUFFER_POOL_STATS' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_METRICS' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_FT_DEFAULT_STOPWORD' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_FT_DELETED' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_FT_BEING_DELETED' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_FT_CONFIG' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_FT_INDEX_CACHE' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_FT_INDEX_TABLE' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_SYS_TABLES' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_SYS_TABLESTATS' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_SYS_INDEXES' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_SYS_COLUMNS' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_SYS_FIELDS' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_SYS_FOREIGN' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_SYS_FOREIGN_COLS' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_SYS_TABLESPACES' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_SYS_DATAFILES' is disabled.
 
141226 13:05:52 [Note] Plugin 'INNODB_CHANGED_PAGES' is disabled.
 
141226 13:05:52 [Note] Plugin 'partition' is disabled.
 
141226 13:05:52 [Warning] /home/dan/software_projects/mariadb-trunk/sql/mysqld: unknown variable 'loose-feedback-user-info=mysql-test'
 
141226 13:05:52 [Note] Server socket created on IP: '::'.
 
141226 13:05:52 [Note] Event Scheduler: Loaded 0 events
 
141226 13:05:52 [Note] /home/dan/software_projects/mariadb-trunk/sql/mysqld: ready for connections.
 
Version: '10.0.16-MariaDB-log'  socket: '/home/dan/software_projects/mariadb-trunk/mysql-test/var/tmp/mysqld.2.sock'  port: 16001  Source distribution
 
 
 
141226 13:06:52 [ERROR] mysqld got signal 11 ;
 
This could be because you hit a bug. It is also possible that this binary
 
or one of the libraries it was linked against is corrupt, improperly built,
 
or misconfigured. This error can also be caused by malfunctioning hardware.
 
 
 
To report this bug, see http://kb.askmonty.org/en/reporting-bugs
 
 
 
We will try our best to scrape up some info that will hopefully help
 
diagnose the problem, but since we have already crashed, 
something is definitely wrong and this may fail.
 
 
 
Server version: 10.0.16-MariaDB-log
 
key_buffer_size=1048576
 
read_buffer_size=131072
 
max_used_connections=1
 
max_threads=153
 
thread_count=1
 
It is possible that mysqld could use up to 
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 62921 K  bytes of memory
 
Hope that's ok; if not, decrease some variables in the equation.
 
 
 
Thread pointer: 0x0x7f5d7bfea008
 
Attempting backtrace. You can use the following information to find out
 
where mysqld died. If you see no messages after this, something went
 
terribly wrong...
 
stack_bottom = 0x7f5d85b01d38 thread_stack 0x48000
 
mysys/stacktrace.c:247(my_print_stacktrace)[0xaa4889]
 
sql/signal_handler.cc:153(handle_fatal_signal)[0x6e19a8]
 
/lib64/libpthread.so.0[0x391120f6d0]
 
sql/sql_statistics.cc:2628(update_statistics_for_table(THD*, TABLE*))[0x612f80]
 
sql/sql_admin.cc:739(mysql_admin_table)[0x66c1b3]
 
sql/sql_admin.cc:1150(Sql_cmd_analyze_table::execute(THD*))[0x66c67c]
 
sql/sql_parse.cc:5078(mysql_execute_command(THD*))[0x5b085d]
 
sql/sql_parse.cc:6407(mysql_parse(THD*, char*, unsigned int, Parser_state*))[0x5b6d88]
 
sql/sql_parse.cc:1301(dispatch_command(enum_server_command, THD*, char*, unsigned int))[0x5b89b7]
 
sql/sql_connect.cc:1375(do_handle_one_connection(THD*))[0x660874]
 
sql/sql_connect.cc:1291(handle_one_connection)[0x6608d8]
 
perfschema/pfs.cc:1863(pfs_spawn_thread)[0x9bdc44]
 
/lib64/libpthread.so.0[0x3911207ee5]
 
/lib64/libc.so.6(clone+0x6d)[0x3910af4b8d]
 
 
 
Trying to get some variables.
 
Some pointers may be invalid and cause the dump to abort.
 
Query (0x7f5d7341e020): analyze table test
 
Connection ID (thread ID): 2
 
Status: NOT_KILLED
 
 
 
Optimizer switch: index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,index_merge_sort_intersection=off,engine_condition_pushdown=off,index_condition_pushdown=on,derived_merge=on,derived_with_keys=on,firstmatch=on,loosescan=on,materialization=on,in_to_exists=on,semijoin=on,partial_match_rowid_merge=on,partial_match_table_scan=on,subquery_cache=on,mrr=off,mrr_cost_based=off,mrr_sort_keys=off,outer_join_with_cache=on,semijoin_with_cache=on,join_cache_incremental=on,join_cache_hashed=on,join_cache_bka=on,optimize_join_buffer_size=off,table_elimination=on,extended_keys=on,exists_to_in=on
 
 
 
The manual page at http://dev.mysql.com/doc/mysql/en/crashing.html contains
 
information that should help you find out what is causing the crash.
 

$ mysql -S /home/dan/software_projects/mariadb-trunk/mysql-test/var/tmp/mysqld.2.sock
 
Welcome to the MariaDB monitor.  Commands end with ; or \g.
 
Your MariaDB connection id is 3
 
Server version: 10.0.16-MariaDB-log Source distribution
 
 
 
Copyright (c) 2000, 2014, Oracle, SkySQL Ab and others.
 
 
 
Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.
 
 
 
MariaDB [(none)]> use t1;
 
 
 
Database changed
 
MariaDB [t1]> source /tmp/mdev-7362.sql
 
..
 
Query OK, 1 row affected (0.00 sec)
 
 
 
MariaDB [t1]> set use_stat_tables='COMPLEMENTARY'
 
    -> ;
 
Query OK, 0 rows affected (0.00 sec)
 
 
 
MariaDB [t1]> analyze table test;
 
ERROR 2013 (HY000): Lost connection to MySQL server during query

Comment by Daniel Black [ 2014-12-26 ]

second seg fault after restarting the server

#0  free_field_buffers_larger_than (table=table@entry=0x7fe3a406a208, size=size@entry=65536) at /home/dan/software_projects/mariadb-trunk/sql/table.cc:3026
 
#1  0x00000000005736cf in close_thread_table (thd=thd@entry=0x7fe3acbea008, table_ptr=table_ptr@entry=0x7fe3acbea0c8) at /home/dan/software_projects/mariadb-trunk/sql/sql_base.cc:1045
 
#2  0x00000000005738a3 in close_open_tables (thd=<optimized out>) at /home/dan/software_projects/mariadb-trunk/sql/sql_base.cc:779
 
#3  close_thread_tables (thd=thd@entry=0x7fe3acbea008) at /home/dan/software_projects/mariadb-trunk/sql/sql_base.cc:1003
 
#4  0x000000000066b3ff in mysql_admin_table (thd=thd@entry=0x7fe3acbea008, tables=tables@entry=0x7fe3a401f0f0, check_opt=check_opt@entry=0x7fe3acbeea00,
 
    operator_name=operator_name@entry=0xafd50f "analyze", lock_type=lock_type@entry=TL_READ_NO_INSERT, open_for_modify=open_for_modify@entry=true,
 
    repair_table_use_frm=repair_table_use_frm@entry=false, extra_open_options=extra_open_options@entry=0, prepare_func=prepare_func@entry=0x0, operator_func=<optimized out>,
 
    view_operator_func=view_operator_func@entry=0x0) at /home/dan/software_projects/mariadb-trunk/sql/sql_admin.cc:1024
 
#5  0x000000000066c67c in Sql_cmd_analyze_table::execute (this=<optimized out>, thd=0x7fe3acbea008) at /home/dan/software_projects/mariadb-trunk/sql/sql_admin.cc:1148
 
#6  0x00000000005b085d in mysql_execute_command (thd=thd@entry=0x7fe3acbea008) at /home/dan/software_projects/mariadb-trunk/sql/sql_parse.cc:5078
 
#7  0x00000000005b6d88 in mysql_parse (thd=0x7fe3acbea008, rawbuf=<optimized out>, length=18, parser_state=0x7fe3b669a5b0)
 
    at /home/dan/software_projects/mariadb-trunk/sql/sql_parse.cc:6407
 
#8  0x00000000005b89b7 in dispatch_command (command=COM_QUERY, thd=0x7fe3acbea008, packet=<optimized out>, packet_length=<optimized out>)
 
    at /home/dan/software_projects/mariadb-trunk/sql/sql_parse.cc:1299
 
#9  0x00000000005b945b in do_command (thd=<optimized out>) at /home/dan/software_projects/mariadb-trunk/sql/sql_parse.cc:996
 
#10 0x0000000000660874 in do_handle_one_connection (thd_arg=thd_arg@entry=0x7fe3acbea008) at /home/dan/software_projects/mariadb-trunk/sql/sql_connect.cc:1375
 
#11 0x00000000006608d8 in handle_one_connection (arg=arg@entry=0x7fe3acbea008) at /home/dan/software_projects/mariadb-trunk/sql/sql_connect.cc:1289
 
#12 0x00000000009bdc44 in pfs_spawn_thread (arg=0x7fe3b5c2f208) at /home/dan/software_projects/mariadb-trunk/storage/perfschema/pfs.cc:1860
 
#13 0x0000003911207ee5 in start_thread (arg=0x7fe3b669b700) at pthread_create.c:309
 
#14 0x0000003910af4b8d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111

Comment by Elena Stepanova [ 2014-12-26 ]

Thanks a lot for the test case.

Comment by Elena Stepanova [ 2014-12-26 ]

Here is a shorter test case on which I'm getting the same crash as in the initial description right now, but it seems it can be volatile – different stack traces on the same test case. Below I'll give more examples of stack traces that I'm getting with variations of the test case. I hope all of them, and MDEV-7380, have the same root cause.
Stack traces are from debug 10.0 revno 4545

CREATE TABLE t1 (a longtext, FULLTEXT KEY (`a`)) ENGINE=MyISAM;
 
insert into t1 values (unhex('3E0D0A4141414142334E7A6143317963324541414141424977414141674541726D'));
 
analyze table t1 persistent for all;
 

#3  <signal handler called>
 
#4  0x0000000000712d89 in update_statistics_for_table (thd=0x7f45aeb2f070, table=0x7f45ac09e470) at 10.0/sql/sql_statistics.cc:2628
 
#5  0x00000000007a4f75 in mysql_admin_table(THD *, TABLE_LIST *, HA_CHECK_OPT *, const char *, thr_lock_type, bool, bool, uint, int (*)(THD *, TABLE_LIST *, HA_CHECK_OPT *), struct {...}, int (*)(THD *, TABLE_LIST *)) (thd=0x7f45aeb2f070, tables=0x7f45ac014180, check_opt=0x7f45aeb33ba0, operator_name=0xf4ac87 "analyze", lock_type=TL_READ_NO_INSERT, open_for_modify=true, repair_table_use_frm=false, extra_open_options=0, prepare_func=0, operator_func=(int (handler::*)(handler * const, THD *, HA_CHECK_OPT *)) 0x8712d4 <handler::ha_analyze(THD*, st_ha_check_opt*)>, view_operator_func=0) at 10.0/sql/sql_admin.cc:738
 
#6  0x00000000007a6565 in Sql_cmd_analyze_table::execute (this=0x7f45ac014760, thd=0x7f45aeb2f070) at 10.0/sql/sql_admin.cc:1148
 
#7  0x0000000000682928 in mysql_execute_command (thd=0x7f45aeb2f070) at 10.0/sql/sql_parse.cc:5078
 
#8  0x00000000006859bd in mysql_parse (thd=0x7f45aeb2f070, rawbuf=0x7f45ac014088 "analyze table t1 persistent for all", length=35, parser_state=0x7f45b5ed9610) at 10.0/sql/sql_parse.cc:6407
 
#9  0x00000000006787b2 in dispatch_command (command=COM_QUERY, thd=0x7f45aeb2f070, packet=0x7f45afbf6071 "", packet_length=35) at 10.0/sql/sql_parse.cc:1299
 
#10 0x0000000000677b57 in do_command (thd=0x7f45aeb2f070) at 10.0/sql/sql_parse.cc:996
 
#11 0x0000000000794877 in do_handle_one_connection (thd_arg=0x7f45aeb2f070) at 10.0/sql/sql_connect.cc:1375
 
#12 0x00000000007945ca in handle_one_connection (arg=0x7f45aeb2f070) at 10.0/sql/sql_connect.cc:1289
 
#13 0x0000000000ccda5e in pfs_spawn_thread (arg=0x7f45adf67bf0) at 10.0/storage/perfschema/pfs.cc:1860
 
#14 0x00007f45b5ac6b50 in start_thread (arg=<optimized out>) at pthread_create.c:304
 
#15 0x00007f45b3d7d20d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:112
 

CREATE TABLE t1 (a longtext, FULLTEXT KEY (`a`)) ENGINE=MyISAM;
 
insert into t1 values (unhex('3E0D0A4141414142334E7A614331796332454141414142497741414167454172'));
 
analyze table t1 persistent for all;
 

#6  0x00007fc9552186f1 in *__GI___assert_fail (assertion=0xf1acaa "bit < (map)->n_bits", file=<optimized out>, line=130, function=0xf1bca0 "uint bitmap_is_set(const MY_BITMAP*, uint)") at assert.c:81
 
#7  0x0000000000711237 in bitmap_is_set (map=0x7fc94d49e580, bit=42405) at 10.0/include/my_bitmap.h:130
 
#8  0x0000000000712dab in update_statistics_for_table (thd=0x7fc94ff2f070, table=0x7fc94d49e470) at 10.0/sql/sql_statistics.cc:2628
 
#9  0x00000000007a4f75 in mysql_admin_table(THD *, TABLE_LIST *, HA_CHECK_OPT *, const char *, thr_lock_type, bool, bool, uint, int (*)(THD *, TABLE_LIST *, HA_CHECK_OPT *), struct {...}, int (*)(THD *, TABLE_LIST *)) (thd=0x7fc94ff2f070, tables=0x7fc94d414180, check_opt=0x7fc94ff33ba0, operator_name=0xf4ac87 "analyze", lock_type=TL_READ_NO_INSERT, open_for_modify=true, repair_table_use_frm=false, extra_open_options=0, prepare_func=0, operator_func=(int (handler::*)(handler * const, THD *, HA_CHECK_OPT *)) 0x8712d4 <handler::ha_analyze(THD*, st_ha_check_opt*)>, view_operator_func=0) at 10.0/sql/sql_admin.cc:738
 
#10 0x00000000007a6565 in Sql_cmd_analyze_table::execute (this=0x7fc94d414760, thd=0x7fc94ff2f070) at 10.0/sql/sql_admin.cc:1148
 
#11 0x0000000000682928 in mysql_execute_command (thd=0x7fc94ff2f070) at 10.0/sql/sql_parse.cc:5078
 
#12 0x00000000006859bd in mysql_parse (thd=0x7fc94ff2f070, rawbuf=0x7fc94d414088 "analyze table t1 persistent for all", length=35, parser_state=0x7fc957425610) at 10.0/sql/sql_parse.cc:6407
 
#13 0x00000000006787b2 in dispatch_command (command=COM_QUERY, thd=0x7fc94ff2f070, packet=0x7fc950ff6071 "", packet_length=35) at 10.0/sql/sql_parse.cc:1299
 
#14 0x0000000000677b57 in do_command (thd=0x7fc94ff2f070) at 10.0/sql/sql_parse.cc:996
 
#15 0x0000000000794877 in do_handle_one_connection (thd_arg=0x7fc94ff2f070) at 10.0/sql/sql_connect.cc:1375
 
#16 0x00000000007945ca in handle_one_connection (arg=0x7fc94ff2f070) at 10.0/sql/sql_connect.cc:1289
 
#17 0x0000000000ccda5e in pfs_spawn_thread (arg=0x7fc94f367bf0) at 10.0/storage/perfschema/pfs.cc:1860
 
#18 0x00007fc957012b50 in start_thread (arg=<optimized out>) at pthread_create.c:304
 
#19 0x00007fc9552c920d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:112
 

CREATE TABLE t1 (a longtext, FULLTEXT KEY (`a`)) ENGINE=MyISAM;
 
insert into t1 values (unhex('3E0D0A4141414142334E7A6143317963324541414141424977414141674541'));
 
analyze table t1 persistent for all;
 

#3  <signal handler called>
 
#4  __strlen_sse2 () at ../sysdeps/x86_64/multiarch/../strlen.S:32
 
#5  0x0000000000716bc3 in Column_stat::set_key_fields (this=0x7f3cd8a16020, col=0x7f3cd07f4145) at 10.0/sql/sql_statistics.cc:841
 
#6  0x0000000000712df9 in update_statistics_for_table (thd=0x7f3cd132f070, table=0x7f3cce89e470) at 10.0/sql/sql_statistics.cc:2631
 
#7  0x00000000007a4f75 in mysql_admin_table(THD *, TABLE_LIST *, HA_CHECK_OPT *, const char *, thr_lock_type, bool, bool, uint, int (*)(THD *, TABLE_LIST *, HA_CHECK_OPT *), struct {...}, int (*)(THD *, TABLE_LIST *)) (thd=0x7f3cd132f070, tables=0x7f3cce814180, check_opt=0x7f3cd1333ba0, operator_name=0xf4ac87 "analyze", lock_type=TL_READ_NO_INSERT, open_for_modify=true, repair_table_use_frm=false, extra_open_options=0, prepare_func=0, operator_func=(int (handler::*)(handler * const, THD *, HA_CHECK_OPT *)) 0x8712d4 <handler::ha_analyze(THD*, st_ha_check_opt*)>, view_operator_func=0) at 10.0/sql/sql_admin.cc:738
 
#8  0x00000000007a6565 in Sql_cmd_analyze_table::execute (this=0x7f3cce814760, thd=0x7f3cd132f070) at 10.0/sql/sql_admin.cc:1148
 
#9  0x0000000000682928 in mysql_execute_command (thd=0x7f3cd132f070) at 10.0/sql/sql_parse.cc:5078
 
#10 0x00000000006859bd in mysql_parse (thd=0x7f3cd132f070, rawbuf=0x7f3cce814088 "analyze table t1 persistent for all", length=35, parser_state=0x7f3cd8a18610) at 10.0/sql/sql_parse.cc:6407
 
#11 0x00000000006787b2 in dispatch_command (command=COM_QUERY, thd=0x7f3cd132f070, packet=0x7f3cd23f6071 "", packet_length=35) at 10.0/sql/sql_parse.cc:1299
 
#12 0x0000000000677b57 in do_command (thd=0x7f3cd132f070) at 10.0/sql/sql_parse.cc:996
 
#13 0x0000000000794877 in do_handle_one_connection (thd_arg=0x7f3cd132f070) at 10.0/sql/sql_connect.cc:1375
 
#14 0x00000000007945ca in handle_one_connection (arg=0x7f3cd132f070) at 10.0/sql/sql_connect.cc:1289
 
#15 0x0000000000ccda5e in pfs_spawn_thread (arg=0x7f3cd0767bf0) at 10.0/storage/perfschema/pfs.cc:1860
 
#16 0x00007f3cd8605b50 in start_thread (arg=<optimized out>) at pthread_create.c:304
 
#17 0x00007f3cd68bc20d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:112

Generated at Thu Feb 08 07:19:00 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.