MySQL Bugs: #61275: Error in err file of slave, when stopping IO thread

Bug #61275	Error in err file of slave, when stopping IO thread
Submitted:	24 May 2011 11:59	Modified:	15 Jan 2013 7:12
Reporter:	Jérôme FONTAINE	Email Updates:
Status:	Duplicate	Impact on me:	None
Category:	MySQL Server: Replication	Severity:	S3 (Non-critical)
Version:	5.5.11, 5.5.21, 5.6.5	OS:	Linux
Assigned to:		CPU Architecture:	Any
Tags:	regression, server_errno=2013

Description:
Hi,

When i'm stopping replication on the database SLAVE, an error message is occuring in .err log file :

110524 11:50:14 [Note] Error reading relay log event: slave SQL thread was killed
110524 11:50:14 [ERROR] Error reading packet from server: Lost connection to MySQL server during query ( server_errno=2013)
110524 11:50:14 [Note] Slave I/O thread killed while reading event
110524 11:50:14 [Note] Slave I/O thread exiting, read up to log 'dppc3.000003', position 107

How to repeat:

Each time, whith the following command,

> STOP SLAVE IO_THREAD;
or
> STOP SLAVE
or 
stopping process mysqld of slave database

Suggested fix:
Don't show the error line ([ERROR]...), when it is a normal stop. Only [Note] lines must be appear.

It's very awkward when ".err" file is monitoring by software like Nagios or Patrol, because it occur a notification of technical problem.

B.R.

Do you really use version 5.1.11? It's ages old, current one is 5.1.57.

Please, check also if there is any error/anything unusual in master's error log for that moment.

Sorry, it's a 5.5.11 !

BR

There is no message on master log when stop slave, only when replication restart :

Log on Master : 
110524 11:50:34 [Note] Start binlog_dump to slave_server(10), pos(dppc3.000003, 107)

BR

Hi,

I'm using MySQL 5.1.14 on production environment.
5.1.14 version have same bug.

Please tell me current status.

sorry,mt

I'm using MySQL 5.5.14 on production environment.

Hi,

New release tested : 5.5.17

> stop slave;

111026 11:51:02 [Note] Error reading relay log event: slave SQL thread was killed
111026 11:51:02 [ERROR] Error reading packet from server: Lost connection to MySQL server during query ( server_errno=2013)
111026 11:51:02 [Note] Slave I/O thread killed while reading event
111026 11:51:02 [Note] Slave I/O thread exiting, read up to log 'dppc3.000018', position 107

> start slave;

111026 11:51:29 [Note] Slave SQL thread initialized, starting replication in log 'dppc3.000018' at position 107, relay log '/db-pcn/sgbd/dppc3sla/arch/dppc3sla-relay-bin.000511' position: 249
111026 11:51:29 [Note] Slave I/O thread: connected to master 'xxxxxxx@xxxxxxx:xxxx',replication started in log 'dppc3.000018' at position 107

The strings "[ERROR]" and "( server_errno=2013)" are interfering with a good monitoring of err file. It's not well-justify for a "normal stop" of slave process.

BR

Thank you for the report.

Verified as described.

Not repeatable with 5.1

This is a duplicate of an internally filed bug that was fixed in 5.7.0.