<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	>
<channel>
	<title>Comments on: Using LVM for MySQL Backup and Replication Setup</title>
	<atom:link href="http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/</link>
	<description>Everything about MySQL Performance</description>
	<pubDate>Sun, 07 Sep 2008 08:07:40 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5.1</generator>
		<item>
		<title>By: peter</title>
		<link>http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-345269</link>
		<dc:creator>peter</dc:creator>
		<pubDate>Sat, 16 Aug 2008 05:15:01 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-345269</guid>
		<description>Richard,

Indeed you can use log file size. Moreover if you have Innodb only tables you do not have to do flush tables with read locks.  Unless you do DDL or modify mysql.* tables you should be fine.</description>
		<content:encoded><![CDATA[<p>Richard,</p>
<p>Indeed you can use log file size. Moreover if you have Innodb only tables you do not have to do flush tables with read locks.  Unless you do DDL or modify mysql.* tables you should be fine.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Richard Bronosky</title>
		<link>http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-345260</link>
		<dc:creator>Richard Bronosky</dc:creator>
		<pubDate>Sat, 16 Aug 2008 04:35:19 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-345260</guid>
		<description>CARP!  Wordpress ate my lines that were supposed to have:
/data/dev/mysql# mysql &#60;&#60; MYSQL
&#62; FLUSH TABLES WITH READ LOCK;</description>
		<content:encoded><![CDATA[<p>CARP!  Wordpress ate my lines that were supposed to have:<br />
/data/dev/mysql# mysql &lt;&lt; MYSQL<br />
&gt; FLUSH TABLES WITH READ LOCK;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Richard Bronosky</title>
		<link>http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-345258</link>
		<dc:creator>Richard Bronosky</dc:creator>
		<pubDate>Sat, 16 Aug 2008 04:31:43 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-345258</guid>
		<description>Peter,
It seems that the master_log_pos (Position as returned by SHOW MASTER STATUS;) is always equal to the file size of the bin-log file.  At least that's what my tests reveal on a MyISAM InnoDB hybrid system.  Is this always consistent?  If so, then you should be able to create a slave from a master snapshot without having to have the information given by SHOW MASTER STATUS.  Why does this matter?  If your nightly backup doesn't capture master status, that shouldn't prevent you from sing it to create a slave.

Here is a transcript...
/data/dev/mysql# mysql &#60; FLUSH TABLES WITH READ LOCK;
&#62; SHOW MASTER STATUS;
&#62; \! ls -lart mysql-bin*&#124;tail -n1
&#62; UNLOCK TABLES;
&#62; MYSQL
File    Position        Binlog_Do_DB    Binlog_Ignore_DB
mysql-bin.000156        956920476
-rw-rw---- 1 mysql mysql 956920476 Aug 16 00:23 mysql-bin.000156
/data/dev/mysql# mysql &#60; FLUSH TABLES WITH READ LOCK;
&#62; SHOW MASTER STATUS;
&#62; \! ls -lart mysql-bin*&#124;tail -n1
&#62; UNLOCK TABLES;
&#62; MYSQL
File    Position        Binlog_Do_DB    Binlog_Ignore_DB
mysql-bin.000156        957115478
-rw-rw---- 1 mysql mysql 957115478 Aug 16 00:23 mysql-bin.000156
/data/dev/mysql# mysql &#60; FLUSH TABLES WITH READ LOCK;
&#62; SHOW MASTER STATUS;
&#62; \! ls -lart mysql-bin*&#124;tail -n1
&#62; UNLOCK TABLES;
&#62; MYSQL
File    Position        Binlog_Do_DB    Binlog_Ignore_DB
mysql-bin.000156        957218874
-rw-rw---- 1 mysql mysql 957218874 Aug 16 00:23 mysql-bin.000156</description>
		<content:encoded><![CDATA[<p>Peter,<br />
It seems that the master_log_pos (Position as returned by SHOW MASTER STATUS;) is always equal to the file size of the bin-log file.  At least that&#8217;s what my tests reveal on a MyISAM InnoDB hybrid system.  Is this always consistent?  If so, then you should be able to create a slave from a master snapshot without having to have the information given by SHOW MASTER STATUS.  Why does this matter?  If your nightly backup doesn&#8217;t capture master status, that shouldn&#8217;t prevent you from sing it to create a slave.</p>
<p>Here is a transcript&#8230;<br />
/data/dev/mysql# mysql &lt; FLUSH TABLES WITH READ LOCK;<br />
&gt; SHOW MASTER STATUS;<br />
&gt; \! ls -lart mysql-bin*|tail -n1<br />
&gt; UNLOCK TABLES;<br />
&gt; MYSQL<br />
File    Position        Binlog_Do_DB    Binlog_Ignore_DB<br />
mysql-bin.000156        956920476<br />
-rw-rw&#8212;- 1 mysql mysql 956920476 Aug 16 00:23 mysql-bin.000156<br />
/data/dev/mysql# mysql &lt; FLUSH TABLES WITH READ LOCK;<br />
&gt; SHOW MASTER STATUS;<br />
&gt; \! ls -lart mysql-bin*|tail -n1<br />
&gt; UNLOCK TABLES;<br />
&gt; MYSQL<br />
File    Position        Binlog_Do_DB    Binlog_Ignore_DB<br />
mysql-bin.000156        957115478<br />
-rw-rw&#8212;- 1 mysql mysql 957115478 Aug 16 00:23 mysql-bin.000156<br />
/data/dev/mysql# mysql &lt; FLUSH TABLES WITH READ LOCK;<br />
&gt; SHOW MASTER STATUS;<br />
&gt; \! ls -lart mysql-bin*|tail -n1<br />
&gt; UNLOCK TABLES;<br />
&gt; MYSQL<br />
File    Position        Binlog_Do_DB    Binlog_Ignore_DB<br />
mysql-bin.000156        957218874<br />
-rw-rw&#8212;- 1 mysql mysql 957218874 Aug 16 00:23 mysql-bin.000156</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: robo</title>
		<link>http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-340639</link>
		<dc:creator>robo</dc:creator>
		<pubDate>Tue, 05 Aug 2008 19:58:36 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-340639</guid>
		<description>Crazy question w/r/t LVM snapshots on a myisam slave:  If updates via replication are not cached(further they invalidate cache?) then is it really necessary to "flush tables with read lock" or will "stop slave" be sufficient? 
Does the same apply to buffers? 
IE, we are already ok with running myisam tables, at cost of locking during the flush in the middle of a long select, is it really worth doing?</description>
		<content:encoded><![CDATA[<p>Crazy question w/r/t LVM snapshots on a myisam slave:  If updates via replication are not cached(further they invalidate cache?) then is it really necessary to &#8220;flush tables with read lock&#8221; or will &#8220;stop slave&#8221; be sufficient?<br />
Does the same apply to buffers?<br />
IE, we are already ok with running myisam tables, at cost of locking during the flush in the middle of a long select, is it really worth doing?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: battery</title>
		<link>http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-315298</link>
		<dc:creator>battery</dc:creator>
		<pubDate>Fri, 20 Jun 2008 01:37:09 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-315298</guid>
		<description>[...]tape drive[...]</description>
		<content:encoded><![CDATA[<p>[...]tape drive[...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Experiments with Running WordpressMU on EC2 &#171; Hefta-Gaub Development Blog</title>
		<link>http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-267038</link>
		<dc:creator>Experiments with Running WordpressMU on EC2 &#171; Hefta-Gaub Development Blog</dc:creator>
		<pubDate>Sun, 06 Apr 2008 00:13:42 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-267038</guid>
		<description>[...] Another article that talks about LVM, MySQL Replication, and backups. [...]</description>
		<content:encoded><![CDATA[<p>[...] Another article that talks about LVM, MySQL Replication, and backups. [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Dennis Glorioso</title>
		<link>http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-255395</link>
		<dc:creator>Dennis Glorioso</dc:creator>
		<pubDate>Sat, 22 Mar 2008 00:50:56 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-255395</guid>
		<description>We run MYSQL on Windows servers. We  use a external NAS that Snapshots an immage of the Volume on other windows servers. We mount that image on a VMware like (vurtualbox) enviromemmt as a live enviroment if the server were to crash. If we are running My SQL on a Windows server Can we use the same FLUSH TABLES WITH READ LOCK command and UNLOCK TABLES to work with our snapshot (storagecraft) SW and directly mount this image as a failover for MySQL instead of using LVM? Also what command to use to hold the lock until snapshot is finished or will it stay in read only until UNLOCK TABLES is run</description>
		<content:encoded><![CDATA[<p>We run MYSQL on Windows servers. We  use a external NAS that Snapshots an immage of the Volume on other windows servers. We mount that image on a VMware like (vurtualbox) enviromemmt as a live enviroment if the server were to crash. If we are running My SQL on a Windows server Can we use the same FLUSH TABLES WITH READ LOCK command and UNLOCK TABLES to work with our snapshot (storagecraft) SW and directly mount this image as a failover for MySQL instead of using LVM? Also what command to use to hold the lock until snapshot is finished or will it stay in read only until UNLOCK TABLES is run</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: LVM snapshots &#171; Daniel&#8217;s weblog</title>
		<link>http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-247901</link>
		<dc:creator>LVM snapshots &#171; Daniel&#8217;s weblog</dc:creator>
		<pubDate>Sun, 02 Mar 2008 20:57:19 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-247901</guid>
		<description>[...] Beware that &#8220;special&#8221; files like databases needs special care as written here or here so they stay in consistent state and as it seems even then it might be a problem when data nad db [...]</description>
		<content:encoded><![CDATA[<p>[...] Beware that &#8220;special&#8221; files like databases needs special care as written here or here so they stay in consistent state and as it seems even then it might be a problem when data nad db [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: MyLVM backup</title>
		<link>http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-246289</link>
		<dc:creator>MyLVM backup</dc:creator>
		<pubDate>Tue, 26 Feb 2008 20:05:21 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-246289</guid>
		<description>[...] Installera mylvmbackup  eller följ guiden på mysqlperformanceblog [...]</description>
		<content:encoded><![CDATA[<p>[...] Installera mylvmbackup  eller följ guiden på mysqlperformanceblog [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Marcus Herou</title>
		<link>http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-246259</link>
		<dc:creator>Marcus Herou</dc:creator>
		<pubDate>Tue, 26 Feb 2008 16:14:50 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/2006/08/21/using-lvm-for-mysql-backup-and-replication-setup/#comment-246259</guid>
		<description>Fuck, it is totally trimmed by wordpress...</description>
		<content:encoded><![CDATA[<p>Fuck, it is totally trimmed by wordpress&#8230;</p>
]]></content:encoded>
	</item>
</channel>
</rss>
