How to log slow queries on Slave in MySQL 5.0 with pt-query-digest

Working as a Percona Support Engineer, every day we are seeing lots of issues related to MySQL replication. One very common issue is slave lagging. There are many reasons for slave lag but one common reason is that queries are taking more time on slave then master. How to check and log those long-running queries?  From MySQL 5.1, log-slow-slave-statements variable was introduced, which you can enable on slave and log slow queries. But what if you want to log slow queries on slave in earlier versions like MySQL 5.0?  There is a good solution/workaround: pt-query-digest. How? let’s take a look….

If you want to log all queries that are running on slave (including those, running by sql thread), you can use pt-query-digest with –processlist and –print (In pt-query-digest 2.1.9) OR –output (In pt-query-digest 2.2.7) options and log all queries in specific file. I have tested it in my local environment and it works.

You can start pt-query-digest like below on Slave,

[email protected]:~$ /percona-toolkit-2.1.9/bin/pt-query-digest --processlist u=msandbox,p=msandbox,S=/tmp/mysql_sandbox34498.sock --print --no-report
[email protected]:-$ /percona-toolkit-2.2.7/bin/pt-query-digest --processlist u=msandbox,p=msandbox,S=/tmp/mysql_sandbox34498.sock --no-report --output=slowlog

Run some long running queries on Master,

[email protected]:~$ mysql -umsandbox -p --socket=/tmp/mysql_sandbox34497.sock
Enter password:
mysql> use percona
Reading table information for completion of table and column names
You can turn off this feature to get a quicker startup with -A
Database changed
mysql> delete from test limit 5000000;
Query OK, 5000000 rows affected (1 min 54.33 sec)
mysql> delete from test limit 5000000;
Query OK, 5000000 rows affected (1 min 56.42 sec)


and you’ll see the output on Slave like this,

[email protected]:~/Downloads/percona-toolkit-2.1.9/bin$ ./pt-query-digest --processlist u=msandbox,p=msandbox,S=/tmp/mysql_sandbox34498.sock --print --no-report
# Time: 2014-03-18T12:10:57
# [email protected]: system user[system user] @ []
# Query_time: 114.000000 Lock_time: 0.000000 Rows_sent: 0 Rows_examined: 0
use percona;
delete from test limit 5000000;
[email protected]:~/Downloads/percona-toolkit-2.2.7/bin$ pt-query-digest --processlist u=msandbox,p=msandbox,S=/tmp/mysql_sandbox34498.sock --no-report --output=slowlog
# Time: 2014-03-18T12:21:05
# [email protected]: system user[system user] @ []
# Query_time: 117.000000 Lock_time: 0.000000 Rows_sent: 0 Rows_examined: 0
use percona;
delete from test limit 5000000;

You can also run pt-query-digest in background like a daemon and send output to the specific file like slow.log and review it.

i.e /percona-toolkit-2.1.9/bin/pt-query-digest –processlist u=msandbox,p=msandbox,S=/tmp/mysql_sandbox34498.sock –print –no-report > slow.log 2>&1


i.e /percona-toolkit-2.2.7/bin/pt-query-digest –processlist u=msandbox,p=msandbox,S=/tmp/mysql_sandbox34498.sock –no-report –output=slowlog > slow.log 2>&1

Here, the default output will be just like slow query log. If we have master-master replication where every master is slave too and we want to log only those statements that are executing by sql_thread then –filter option can be used like this:

pt-query-digest –filter ‘$event->user eq “system user”‘ –no-report –output=slowlog

Since pt-query-digest–processlist polls 10 times/second ( –interval option), it’s not reliable to use for collecting complete query logs, because quick queries could fall in between the polling intervals. And in any case, it won’t measure query time with precision any better than 1/10th of a second. But if the goal is to identify queries that are very long-running, it should be adequate.

The post How to log slow queries on Slave in MySQL 5.0 with pt-query-digest appeared first on MySQL Performance Blog.

Read more at: //

Want to archive tables? Use Percona Toolkit’s pt-archiver

pt-archiverPercona Toolkit’s pt-archiver is one of the best utilities to archive the records from large tables to another tables or files. One interesting thing is that pt-archiver is a read-write tool. It deletes data from the source by default, so after archiving you don’t need to delete it separately.

As it is done by default, you should take care before actually running it on then production server. You can test your archiving jobs with the – dry-run  OR you can use the –no-delete option if you’re not sure about. The purpose of this script is mainly to archive old data from the table without impacting OLTP queries and insert the data into another table on the same/different server OR into a file in a format which is suitable for LOAD DATA INFILE.

How does pt-archiver select records to archive? 

Pt-archiver uses the index to select records from the table. The index is used to optimize repeated accesses to the table. Pt-archiver remembers the last row it retrieves from each SELECT statement, and uses it to construct a WHERE clause. It does this using the columns in the specified index that should allow MySQL to start the next SELECT where the last one ended – rather than potentially scanning from the beginning of the table with each successive SELECT.

If you want to run pt-archiver with a specific index you can use the “-i” option in –source DSN options. The “-i” option tells pt-archiver which index it should scan to archive. This appears in a FORCE INDEX or USE INDEX hint in the SELECT statements that are used to fetch rows to archive. If you don’t specify anything, pt-archiver will auto-discover a good index, preferring a PRIMARY KEY if one exists. Most of the time, without “-i” option, pt-archiver works well.

How to run pt-archiver?

For archive records into normal file, you can run something like

pt-archiver --source h=localhost,D=nil,t=test --file '/home/nilnandan/%Y-%m-%d-tabname' --where "name='nil'" --limit-1000

From archive records from one table to another table on same server or different, you can run something like

pt-archiver --source h=localhost,D=nil,t=test --dest h=fedora.vm --where "name='nil'" --limit-1000

Please check this before you use default file option (-F) in –source  //–dest

Archiving in a replication environment:

In the replication environment it’s really important that the slave should not lag for a long time. So for that, there are two options which we can use while archiving to control the slave lag on slave server.

–check-slave-lag : Pause archiving until the specified DSN’s slave lag is less than –max-lag. In this option, you can give slave details to connect slave lag. (i.e –check-slave-lag h=localhost,S=/tmp/mysql_sandbox29784.sock)

–max-lag : Pause archiving if the slave given by –check-slave-lag lags.

This options causes pt-archiver to look at the slave every time when it’s about to fetch another row. If the slave’s lag is greater than the option’s value, or if the slave isn’t running (so its lag is NULL), pt-archiver sleeps for –check-interval seconds and then looks at the lag again. It repeats until the slave is caught up, then proceeds to fetch and archive the row.

Some useful options for pt-archiver:

–for-update/-share-lock  : Adds the FOR UPDATE/LOCK IN SHARE MODE  modifier to SELECT statements.

–no-delete : Do not delete archived rows.

–plugin : Perl module name to use as a generic plugin.

–progress : Print progress information every X rows.

–statistics : Collect and print timing statistics.

–where : WHERE clause to limit which rows to archive (required).

[email protected]:~$ pt-archiver --source h=localhost,D=nil,t=test,S=/tmp/mysql_sandbox29783.sock --file '/home/nilnandan/%Y-%m-%d-tabname' --where "name='nilnandan'" --limit=50000 --progress=50000 --txn-size=50000 --statistics --bulk-delete --max-lag=1 --check-interval=15 --check-slave-lag h=localhost,S=/tmp/mysql_sandbox29784.sock
2013-08-08T10:08:39 0 0
2013-08-08T10:09:25 46 50000
2013-08-08T10:10:32 113 100000
2013-08-08T10:11:41 182 148576
Started at 2013-08-08T10:08:39, ended at 2013-08-08T10:11:59
Source: D=nil,S=/tmp/mysql_sandbox29783.sock,h=localhost,t=test
SELECT 148576
DELETE 148576
Action Count Time Pct
print_file 148576 18.2674 9.12
bulk_deleting 3 8.9535 4.47
select 4 2.9204 1.46
commit 3 0.0005 0.00
other 0 170.0719 84.95
[email protected]:~$

Percona Toolkit’s pt-archiver works with Percona XtraDB Cluster (PXC) 5.5.28-23.7 and newer, but there are three limitations you should consider before archiving on a cluster. You can get more information here.

pt-archiver is extensible via a plugin mechanism. You can inject your own code to add advanced archiving logic that could be useful for archiving dependent data, applying complex business rules, or building a data warehouse during the archiving process. Follow this URL for more info on that.

Bugs related to pt-archiver:

More details about pt-archiver: //

The post Want to archive tables? Use Percona Toolkit’s pt-archiver appeared first on MySQL Performance Blog.

Read more at: //