Comments on: How SHOW SLAVE STATUS relates to CHANGE MASTER TO

By: Robin

Robin — Sun, 27 Nov 2011 04:48:25 +0000

Hello,everybody.
Does mysql have a feature to customize the slave port? I see my slave port is random.
Thank you!

By: Imran Chaudhry

Imran Chaudhry — Mon, 12 Sep 2011 11:58:20 +0000

Hi Peter, are you sure you are correct here on Relay_Master_Log_File:Exec_Master_Log_Pos being the values to use?

I have a master-master replication scenario (2 nodes). If replication breaks on one node (lets say due to duplicate key insert) and I want to kickstart replication then I stop slaves on both, refresh mysql data (dump with –master-data) on passive node from active node, “change master to” on passive node using the values in the dump. Then to deal with the active node I also use “change master to” but I use the values fetched from “show master status” on the passive slave. It would be incorrect to use the Relay_Master_Log_File:Exec_Master_Log_Pos from the passive slave as you seemed to imply. I’m sure it’s just a misunderstanding and you didn’t mean this

Thanks,
Imran

By: Alexis Guajardo

Alexis Guajardo — Thu, 07 Apr 2011 17:57:36 +0000

Getting this error:
Got fatal error 1236 from master when reading data from binary log: ‘Could not find first log file name in binary log index file’

tried changing binlon file to position to full path, but no luck, any suggestions?

By: Peter Zaitsev

Peter Zaitsev — Tue, 22 Feb 2011 19:13:55 +0000

Kyle,

It looks like you either have wrong position (for whatever reason) or binary log is corrupted for some reason. It is unlikely you really have 1.5GB packet.

By: Kyle

Kyle — Tue, 22 Feb 2011 17:18:05 +0000

Hi,

We are trying to add an additional server to our replication using the Relay_Master_Log_File/Exec_Master_Log_Pos of an existing slave and copying the data. When we try and restart the new slave with those values we get the following error.

Got fatal error 1236 from master when reading data from binary log: ‘error reading log entry’

Lots of posts have recommended increasing the max_allowed_packet and keeping them the same on the master/slave. We have done this with no success. When I run mysqlbinlog on the Exec_Master_Log_Pos the position is invalid and I get:

ERROR: Error in Log_event::read_log_event(): ‘Event too big’, data_len: 1415072078, event_type: 82

Looking at the binlog the Exec position is not valid in the binlog. Why would that be?

By: Suresh Prajapati

Suresh Prajapati — Thu, 07 Oct 2010 23:18:40 +0000

Hi Peter .
Thanx for such a quality information.
Can u please tell me where i can learn mysql administration concepts like replication and things like this.
Because the organizxation in which i am working using mysql for various tasks nad there are a lots of slave and circular replication.
I will appreciate if you please help in getting the concepts related to memcache.

thanx.

By: Mat

Mat — Mon, 14 Dec 2009 11:11:49 +0000

Am Using Multi master Replication, and everything works great
The problem is when a failure happens or connection is lost , replication does not recover automatic
I think its because the slave Position changes
and i have to stop slaves and show master positions again then retype the change master into and insert there new values again to make it up again
how can i avoid that ?! isnt there a way to make the slave know the position automatic ?! or even fix the position and not let it change ?

Thanks in advance

By: martir

martir — Fri, 04 Sep 2009 06:50:28 +0000

I have problem and I don’t know how to solve this problem. When I do the command “SHOW MASTER STATUS;’ I get resoult “Empty set (0.00 sec)”. I don’t know what to do. Please if someone can halp me. Thanks a lot

By: Darko Bunic | www.redips.net

Darko Bunic | www.redips.net — Thu, 05 Feb 2009 10:52:34 +0000

Just to be sure. In case when slave stops to work (create table statement on master specified InnoDB storage but slave is MyISAM only), what options do I have to fix replication? – master executed SQL without any problem but slave didn’t.
1) I suppose I can read the position from the master and write it to the slave with “change master to master_log_file= …” statement
2) I suppose I can skip problematic statement with SQL_SLAVE_SKIP_COUNTER on slave only
3) is there any other option like:
slave stop;
change master to master_log_file=’(binlog name in relay_master_log_file)’, master_log_pos=(exec_master_log_pos number);
slave start;
… on slave of course, or I missed the point.

By: Jure Pecar

Jure Pecar — Tue, 28 Oct 2008 21:14:34 +0000

Looks like mysql 6 is introducing MASTER_IGNORE_SERVER_ID to fix this problem. See http://bugs.mysql.com/bug.php?id=25998