[MDEV-3497] LP:887051 - Aria mysqld: recovered pages: 0% 20%111105 21:05:20 [ERROR] mysqld got signal 11 ; Created: 2011-11-07  Updated: 2015-02-02  Resolved: 2012-10-04

Status: Closed
Project: MariaDB Server
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Critical
Reporter: Dreas van Donselaar (Inactive) Assignee: Michael Widenius
Resolution: Fixed Votes: 0
Labels: Launchpad

Attachments: XML File LPexportBug887051.xml     File LPexportBug887051_test.pl    

 Description   

MySQL failed to startup on a server with:
=====
Nov 5 21:05:20 server1 mysqld_safe: Starting mysqld daemon with databases from /var/lib/mysql
Nov 5 21:05:20 server1 mysqld: 111105 21:05:20 [Warning] 'master-host' is deprecated and will be removed in a future release. Please use ''CHANGE MASTER'' instead.
Nov 5 21:05:20 server1 mysqld: 111105 21:05:20 [Note] Plugin 'BLACKHOLE' is disabled.
Nov 5 21:05:20 server1 mysqld: 111105 21:05:20 [Note] Plugin 'FEDERATED' is disabled.
Nov 5 21:05:20 server1 mysqld: 111105 21:05:20 [Note] mysqld: Aria engine: starting recovery
Nov 5 21:05:20 server1 mysqld: recovered pages: 0% 20%111105 21:05:20 [ERROR] mysqld got signal 11 ;
Nov 5 21:05:20 server1 mysqld: This could be because you hit a bug. It is also possible that this binary
Nov 5 21:05:20 server1 mysqld: or one of the libraries it was linked against is corrupt, improperly built,
Nov 5 21:05:20 server1 mysqld: or misconfigured. This error can also be caused by malfunctioning hardware.
Nov 5 21:05:20 server1 mysqld: We will try our best to scrape up some info that will hopefully help diagnose
Nov 5 21:05:20 server1 mysqld: the problem, but since we have already crashed, something is definitely wrong
Nov 5 21:05:20 server1 mysqld: and this may fail.
Nov 5 21:05:20 server1 mysqld:
Nov 5 21:05:20 server1 mysqld: key_buffer_size=0
Nov 5 21:05:20 server1 mysqld: read_buffer_size=131072
Nov 5 21:05:20 server1 mysqld: max_used_connections=0
Nov 5 21:05:20 server1 mysqld: max_threads=502
Nov 5 21:05:20 server1 mysqld: threads_connected=0
Nov 5 21:05:20 server1 mysqld: It is possible that mysqld could use up to
Nov 5 21:05:20 server1 mysqld: key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 1098540 K
Nov 5 21:05:20 server1 mysqld: bytes of memory
Nov 5 21:05:20 server1 mysqld: Hope that's ok; if not, decrease some variables in the equation.
Nov 5 21:05:20 server1 mysqld:
Nov 5 21:05:20 server1 mysqld: Thread pointer: 0x0
Nov 5 21:05:20 server1 mysqld: Attempting backtrace. You can use the following information to find out
Nov 5 21:05:20 server1 mysqld: where mysqld died. If you see no messages after this, something went
Nov 5 21:05:20 server1 mysqld: terribly wrong...
Nov 5 21:05:20 server1 mysqld: stack_bottom = (nil) thread_stack 0x40000
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld(my_print_stacktrace+0x2e) [0x7f6dd2ef195e]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld(handle_segfault+0x3eb) [0x7f6dd2b27b2b]
Nov 5 21:05:20 server1 mysqld: /lib/libpthread.so.0 [0x7f6dd1124a80]
Nov 5 21:05:20 server1 mysqld: /lib/libc.so.6 [0x7f6dd06a8561]
Nov 5 21:05:20 server1 mysqld: /lib/libc.so.6(memmove+0x7c) [0x7f6dd06a738c]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld(_ma_apply_redo_index+0x65c) [0x7f6dd2d2fc9c]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld [0x7f6dd2d4f2fd]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld [0x7f6dd2d4b441]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld [0x7f6dd2d4e90d]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld(maria_apply_log+0x19c) [0x7f6dd2d5025c]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld(maria_recovery_from_log+0x50) [0x7f6dd2d50bc0]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld [0x7f6dd2d09066]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld(ha_initialize_handlerton(st_plugin_int*)+0x38) [0x7f6dd2c1a768]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld [0x7f6dd2ca232c]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld(plugin_init(int*, char**, int)+0x518) [0x7f6dd2ca4578]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld [0x7f6dd2b29bbb]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld(main+0x1f0) [0x7f6dd2b2a740]
Nov 5 21:05:20 server1 mysqld: /lib/libc.so.6(__libc_start_main+0xe6) [0x7f6dd06491a6]
Nov 5 21:05:20 server1 mysqld: /usr/sbin/mysqld [0x7f6dd2a59759]
Nov 5 21:05:20 server1 mysqld: The manual page at http://dev.mysql.com/doc/mysql/en/crashing.html contains
Nov 5 21:05:20 server1 mysqld: information that should help you find out what is causing the crash.
Nov 5 21:05:20 server1 mysqld: Writing a core file
Nov 5 21:05:20 server1 mysqld_safe: mysqld from pid file /var/run/mysqld/mysqld.pid ended
Nov 5 21:05:34 server1 /etc/init.d/mysql[32456]: 0 processes alive and '/usr/bin/mysqladmin --defaults-file=/etc/mysql/debian.cnf ping' resulted in
Nov 5 21:05:34 server1 /etc/init.d/mysql[32456]: #007/usr/bin/mysqladmin: connect to server at 'localhost' failed
Nov 5 21:05:34 server1 /etc/init.d/mysql[32456]: error: 'Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2)'
Nov 5 21:05:34 server1 /etc/init.d/mysql[32456]: Check that mysqld is running and that the socket: '/var/run/mysqld/mysqld.sock' exists!
Nov 5 21:05:34 server1 /etc/init.d/mysql[32456]:
=====

Moving out aria_log.00000003 and aria_log_control allowed MySQL to start again. Running with 5.2.9. Uploaded a copy of those files to FTP.

 Comments   
Comment by Elena Stepanova [ 2011-12-06 ]

Re: Aria mysqld: recovered pages: 0% 20%111105 21:05:20 [ERROR] mysqld got signal 11 ;

  1. Test case
  2. Put it in the same mysql-test folder where mysql-test-run.pl resides.
  3. Run as
  4. perl ./<test name>
  5. The same test case will also be attached

$ENV

{MTR_VERSION}

=2;

print "#### Creating test file and data...####\n";

open( TEST, '>t/test.test' )

die "Could not open t/test.test for writing\n";
print TEST "
USE test; \
DROP TABLE IF EXISTS t1; \
CREATE TABLE t1 \
( word VARCHAR(255) PRIMARY KEY ) ENGINE=Aria; \
LOAD DATA LOCAL INFILE 'update.in' INTO TABLE t1; \
SET AUTOCOMMIT=0; \
DELETE FROM t1; \
LOAD DATA LOCAL INFILE 'update.in' \
INTO TABLE t1 IGNORE 1 LINES; \
COMMIT; \
";
close( TEST );
open( DATA, '>update.in' )
die "Could not open update.in for writing\n";
print DATA "saved1\nsaved2";
close( DATA );

print "#### Starting server... ####\n";
system( "perl ./mysql-test-run.pl --start-and-exit test" );

print "#### Running SQL... ####\n";
system( "perl mysql-test-run.pl "
. "--extern socket=var/tmp/mysqld.1.sock test" );

print "#### Reading PID file... ####\n";
open( PID, 'var/run/mysqld.1.pid' )

die "Could not open var/run/mysqld.1.pid\n";
my $pid = <PID>;
close( PID );
chomp $pid;

print "#### Killing server.... ####\n";
system( "kill -9 $pid" );

print "#### Trying to restart the server.... ####\n";
system( "perl mysql-test-run.pl --start-dirty test" );
system( "cat var/log/mysqld.1.err" );

  1. EOF
Comment by Elena Stepanova [ 2011-12-06 ]

Re: Aria mysqld: recovered pages: 0% 20%111105 21:05:20 [ERROR] mysqld got signal 11 ;
To reproduce the problem manually:

  • Create a text file update.in with two lines in it, e.g.
    saved1
    saved2
  • Start a server on a fresh data folder, default server settings are enough
  • Execute SQL

USE test;
DROP TABLE IF EXISTS t1;
CREATE TABLE t1
( word VARCHAR(255) PRIMARY KEY ) ENGINE=Aria;
LOAD DATA LOCAL INFILE 'update.in' INTO TABLE t1;
SET AUTOCOMMIT=0;
DELETE FROM t1;
LOAD DATA LOCAL INFILE 'update.in'
INTO TABLE t1 IGNORE 1 LINES;
COMMIT;

(make sure you have the right path to update.in)

  • kill server with SIGKILL or alike
  • try to start server on the same datadir again
Comment by Elena Stepanova [ 2011-12-06 ]

Re: Aria mysqld: recovered pages: 0% 20%111105 21:05:20 [ERROR] mysqld got signal 11 ;
Same perl test as an attachment, for convenience

Comment by Elena Stepanova [ 2011-12-06 ]

Same perl test as an attachment, for convenience
Test script
LPexportBug887051_test.pl

Comment by Rasmus Johansson (Inactive) [ 2012-02-27 ]

Launchpad bug id: 887051

Generated at Thu Feb 08 06:49:04 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.