This post was thought of as an attempt to make some performance test with new multi-threaded replication in 5.6, at least that was my initial intention. Based on Luis Soarez – Replication Team Leader in Oracle – post. I wanted to play with new set of variables and features in order to understand how new multi-threaded replication works and make some performance tests.

First impressions:
=================
I will start with my negative impressions about how to setup new replication features:

– Lack of documentation: I know this is a labs release and most of new features are not documented yet, I’ve just found the Luis blog which has some definitions but is mostly an overview than a manual.

– Variables not present: mts% variables are no longer present in last release.

Good …

As those of you with an interest in MySQL may have seen, I’ve recently surged into the list of “Most Active Authors” on Planet MySQL, alas I’m not as prolific a blogger as it may seem, and this is purely down to an error with the feed I submitted to Planet MySQL. Having recently joined [...]

I’ve been following the development of Tungsten Replicator for quiet some time now, and recently was fortunate enough to find the time to take a look at the product in more detail. If you haven’t heard of Tungsten Replicator yet, it’s an open source database replication engine that can be used to complement or completely [...]

MySQL Replication is a powerful tool and it’s hard to find a production system not using it. On the other hand debugging replication issues can be very hard and time consuming. Especially if your replication setup is not straightforward and you are using filtering of some kind.

Recently we got an alert from our monitoring system that replication stopped on production slave with the following error:

Can't find record in 'some_table', Error_code: 1032;
handler error HA_ERR_KEY_NOT_FOUND;
the event's master log binlog.000011, end_log_pos 735766642

This means that a ROW-based replication event was going to be applied on slave, but could not find the row it was supposed to be applied to. This is something I like about ROW format — it allows you to catch such data synchronization issues right away. In this particular case MIXED format was used, but if this event was written in STATEMENT format, slave would just apply it …

If you support Oracle RDBMS 11.2.0.2 and want to zero downtime applying
patches to databases then it is time to have a look at new possibility
of Online Patching delivered with 11.2.0.2 version and described here
RDBMS Online Patching Aka Hot Patching [ID 761111.1]

The syntax is “opatch apply online -connectString … ” but I could not find description
of the online option (only -connectString) using opatch -help (version 11.2.0.1.6)
although based on the note it is available since 11.1.0.6 version of opatch

And while going through README for the patch I found the blog
about online patching …

A few days ago, we faced an interesting problem on one of our customer’s slave mysqld servers.  An Alter for adding a new column was run on master server took 542 seconds where as it took few hours on the slave server to complete due to a SELECT blocking the Alter was not allowed to complete.

Here is the the select on the master server and it’s execution time from the binary logs.

# at 825737566
#110720 19:55:21 server id 31415  end_log_pos 825737730         Query   thread_id=83250629      exec_time=542   error_code=0
use collect/*!*/;
SET TIMESTAMP=1311216921/*!*/;
/*!\C latin1 *//*!*/;
SET @@session.character_set_client=8,@@session.collation_connection=8,@@session.collation_server=8/*!*/;
ALTER TABLE `track` ADD `source_track_id` int(10) UNSIGNED NULL DEFAULT NULL  AFTER `vendor_track_id`
/*!*/;

Alter statement completed well on the master and it got blocked by a SELECT on the slave where as the time frames of the …

Yesterday I was discussing with a fellow DBA about ways to check the status of existing and/or past RMAN jobs. Good backup scripts usually write their output to some sort of log file so, checking the output is usually a straight-forward task. However, backup jobs can be scheduled in many different ways (crontab, Grid Control, Scheduled Tasks, etc) and finding the log file may be tricky if you don’t know the environment well.
Furthermore, log files may also have already been overwritten by the next backup or simply just deleted. An alternative way of accessing that information, thus, may come handy.

Fortunately, RMAN keeps the backup metadata around for some time and it can be accessed through the database’s V$ views. Obviously, if you need this information because your database just crashed and needs to be restored, the method described here is useless.

Backup jobs’ status and metadata

A lot of metadata about …

Pythian’s Oracle ACE, Fahd Mirza appears in this month’s Community: Peer-to-Peer review “In With the New“, as published in the September/October 2011 issue of Oracle Magazine.

Fahd states that “Oracle Exadata Database Machine” has most changed his life – changing the game, and setting very high standards of performance, support, scalability, reliability and unification.

Shout out to Fahd from your peers at Pythian!

I guess there might be just a little truth to Pythian’s growing reputation as an “Oracle ACE Factory” ;), as recently mentioned by Justin Kestelyn in the May 11, 2011 OPN PartnerCast:

Please join me in congratulating Fahd by adding a …

Pythian is pleased to announce our speaking schedule at this year’s Oracle OpenWorld 2011, October 2-6, 2011 in San Francisco, CA.

We’re excited to be joined by our customers Western Union, and Worldwide Technologies (WWT) as we present real-world experiences and project success. If you’re attending, don’t miss the chance to hear our team of experts. Bring your toughest questions to be answered as they relate to any of the subjects below.

• Follow Pythian on LinkedIn for up to date OOW info
• Schedule a meeting with Pythian at OOW to discuss your database …

On many of our clients, we have a need to run XtraBackup as a regular OS user. Aside from running into the issue where tar4ibd was not provided with Percona’s xtrabackup-1.6.2.tar.gz package, our main issues have been with permissions when attempting a streaming backup.

I have found the following:

1. The user needs permissions for a temp directory to stream to/from. The my.cnf of the target database cannot be used because the user does not have permission to write to /tmp/mysql-stdout, so we set a tmpdir in a separate defaults-file.
2. A backup target directory must be used that the user has read/write permissions to. It seems to me a target directory should not be needed for a streaming backup, …