As you can see from error log the relay log gets corrupted. It's either memory/disk/some_system_library problem or a bug... In my experience it happened only when server got rebooted and opened relay log file was messed up.
тогда не знаю, пинай сысадмина :) что-то там серьезно не в порядке в системе - файлы просто так не исчезают... если хочешь кучу бесплатных/бесполезных советов, спроси на sql.ru в форуме mysql ;)
Turned out the actual problem was that the binary log index file on the slave contained references to files that no longer existed on the slave.
The solution was to do
updatedb locate slave-relay-*
and then to edit the file
/var/lib/mysql/slave-relay-bin.indexand remove all lines referring to files that didn't show up in the locate results. Note find could have been used instead of updatedb and locate
( ... )
Hi, I had the same error with configuring the master. In fact, I have never been able to configure replication to start at all, can you please post a step by step note or MY.CNF files for me? would greatly appreaciate, I am using a Fedora 10 machine (mysql Ver 14.14 Distrib 5.1.47) and Mandriva 2010 (ver 14.14 Distrib 5.1.42), please help
Now i find only one solution: Drop all data, etc, and restart replication from "zero point". I even set the simple script for that like slave stop; delete ... drop .. set ... slave start;
command all from mysql manual from "replication" section.
Comments 9
In my experience it happened only when server got rebooted and opened relay log file was messed up.
Reply
Reply
если хочешь кучу бесплатных/бесполезных советов, спроси на sql.ru в форуме mysql ;)
Reply
http://blog.bit-matrix.com/2008/11/19/mysql-replication-error-1201-could-not-initialize-master-info-structure/
Reply
ERROR 1201 (HY000): Could not initialize master info structure; more error messages can be found in the MySQL error log
in response to the command:
mysql> change master to master_host='192.168.0.2', master_user='####', master_password='####', master_log_file='master-bin.002557', master_log_pos=320489201;
All that was in the error log was
100519 11:05:13 [ERROR] Error reading master configuration
which didn't exactly help!
Turned out the actual problem was that the binary log index file on the slave contained references to files that no longer existed on the slave.
The solution was to do
updatedb
locate slave-relay-*
and then to edit the file
/var/lib/mysql/slave-relay-bin.indexand remove all lines referring to files that didn't show up in the locate results. Note find could have been used instead of updatedb and locate ( ... )
Reply
Reply
slave stop;
delete ...
drop ..
set ...
slave start;
command all from mysql manual from "replication" section.
other solutions does not help for me.
Reply
change master to....
is the solution you all looked for...
Reply
Reply
Leave a comment