I came up with this analogy today comparing MySQL and Oracle and it seemed to fit. Only real Unix geeks will get this, but here goes:

MySQL is to Oracle as Vi is to Emacs

Maybe this has been said before, but think about it.

Here's what they all have in common:

• 
  
• All are cool technologies and worthy of respect in their own right
  
• All have zealots vigorously defend their technology to the death
  
• Each set (Oracle/Emacs and MySQL/Vi) represents a different way of looking at and solving a problem
  

For the Oracle/Emacs world things work like this:

The technology is the be-all/end-all, make it do everything we want.

In many ways this is a huge advantage, things are tightly integrated, you don't need to worry so much about system conventions and …

MySQL Table Checksum 1.1.5 adds a some useful features and fixes a couple of bugs. Now you can checksum tables in chunks, and there is an option to pause between chunks as well. This has already helped me recover a very large table that got out of sync on the slaves, and many of you have also requested this feature.

One thing that has been a curious topic among some of us here at work has been if raw InnoDB partitions are better than using files on-top of a file-system. Well, I was able to find a test server that, while it has modest hardwawre, it did have a spare drive to use for this purpose.

First things first, the tests were run on a system with an Athlon XP 2000+, 512MB of RAM, and 2 identical 80GB EIDE drives with 2MB of disk cache. The OS was Red-Hat ES4 and I used MySQL 5.0.41 for the tests. The InnoDB data was housed on a single partition on the second drive for all tests. The InnoDB logs, however, remained in /var/lib/mysql. MySQL itself was configured using default settings (with the exception of having to change the innodb_data_file_path of course).

The file-system test was done by formatting the partition with EXT3, mounting it under /mnt/innodb and having MySQL use it for the ibdata files. For the raw partition, though …

Von Kriesse.

Nutzt Du Social Bookmarking/Networking Seiten wie Digg, Yigg, Mister Wong oder Del.Icio.Us? Wenn ja, welche?

Digg fuer News und del.icio.us zum Bookmarks ablegen und finden.

Schaust Du Dir Videos bei Videoportalen wie YouTube, Sevenload oder DailyMotion an? Wenn ja, was für Videos guckst Du Dir meistens an?

Ab und zu YouTube, entweder albernes Zeug oder irgendwelche neue Musik, die es noch nicht zu kaufen gibt.

Kennst bzw. nutzt Du Flickr?

Ja

Was hältst Du von Diensten wie Twitter oder Frazr?

Twitter ist nett um Events zu verfolgen, an denen man nicht teilnehmen kann. Sonst ist das eher uninteressant.

Wie viele RSS-Feeds hast Du abonniert?

138

Liest Du die alle …

Last week, I wrote about my subjective Top Ten Reasons To Work For MySQL. Let me be a bit more specific this week. We’re looking for a CIO!

This leadership opportunity is about leading our IT department as CIO/Vice President of IT. Personally, I had a stab at these responsibilities before becoming VP Community Relations, and I can tell you it is a challenge.

MySQL?s technology is helping power the internet and we are enjoying explosive growth. Our company unites the power of open-source computing with web-based application providers and the CIO is leading this evolution from the front. Ideally, the candidate comes from like-minded enterprises and thrives in the excitement of being part of an international, virtual company providing disruptive technology.

The …

Microsoft upped the FUD (Fear, Uncertainty, Doubt) in the open source world with a carefully planned story given to Fortune magazine Monday in which their legal counsel revealed that Microsoft believes Linux and other open source software violate more than 200 of Microsoft's software patents. 

As the article points out, the whole idea of software patents is a relative new one and there are likely thousands of bogus patents that will not stand up in court.  In fact it's not that many years ago that the Bill Gates was talking about how software patents could be abused: 

"If people had understood how patents would be granted when most of today's ideas were invented and had taken out patents, …

Since 5.0.34 of mySQL, something changed that is not compatible with my generic all static build of a mySQL RPM I can get it to work if I use mysqlmanager, but executing mysqld at the command line blocks waiting for a signal. The steps below shows how to track down problems with mysql binaries.

/usr/sbin/mysqld-debug --basedir=/ --datadir=/var/lib/mysql --user=mysql --pid-file=/var/lib/mysql/db.pid --skip-external-locking --console --debug --log --one-thread

This will create a mysql.trace file in /tmp

the output of the trace file suspends just like the mysqld process, this is post innodb setup phase.

T@16384: | | | >_mymalloc
T@16384: | | | | enter: Size: 24
T@16384: | | | | exit: ptr: 0x1d1cc5c8
T@16384: | | | …

that says something along the lines of “contrary to popular beleif, I don’t speak Swedish”. I find that some people will switch to Swedish when speaking to me…  Although I do pick up a bit…. not quite a local though :)

New alpha release 1.0-pre4 of the MySQL Master-Master Replication Manager. This release has lots of major fixes and I’m glad to announce first sponsored port of mmm to non-linux platform - it has been ported to Solaris 10. So, here are our changes in this version:

• Real checks timeouts - I’ve found and fixed lots of problems in checks timeout code and now if you specified in your mmm_mon.conf, that some check should timeout in 5 sec, it would timeout correctly on all supported platforms.
• External third-party tools using - On all supported non-linux platforms mmm will use system binaries for fping and arp_ping so porting to another platforms would be much easier.
• Agent …

UPDATE:
* s/Using join cache/Using join buffer/, changed to show the final variants of EXPLAIN output as described here
* s/join_buff_size/join_buffer_size/

Starting from 5.1.18, EXPLAIN output may show "Using join cache", like in this example:

mysql> explain select * from t1, t2 where t1.col < 10 and t2.col < 'bar';
+----+-------------+-------+-------+-...-+--------------------------------+
| id | select_type | table | type  |     | Extra                          |
+----+-------------+-------+-------+-...-+--------------------------------+
|  1 | SIMPLE      | t1    | range |     | Using where                    |
|  1 | SIMPLE      | t2    | range |     | Using where; Using join buffer |
+----+-------------+-------+-------+-...-+--------------------------------+

The join cache is actually not a new feature. It has been …