Three months ago, I wrote about a procedure to create MySQL FEDERATED tables.

I just added a quick fix that should make the procedure less susceptible to issues relating to bug #23856. To do that, I had to remove the ORDER BY ordinal_position bits in all the calls to GROUP_CONCAT over the rows in the information_schema.COLUMNS table.

If you include the ORDER BY ordinal_position clause, the concatenation result will mess up sometimes. The behaviour can be attenuated somewhat by decreasing value for the group_concat_max_len server variable, but so far, I have not seen a sufficiently …

What happened to the MySQL Winter of Code? Are they waiting for winter in Australia?

I live near Boston, MA and I can tell you it’s definitely winter in the northern hemisphere….

So what are we waiting for?

Well, I can say this — we’re waiting for people. The Winter of Code idea is a great one, particularly since if MySQL works with academic institutions they could help students find Master’s Projects or part of Ph.D. work. Imagine someone writing a new storage engine and having that earn them a Master’s degree. This is exactly what MySQL needs — more people who understand database internals and best theoretical practices to start coding and see where it goes. Note the “more people” — they already have staff that does this.

I’m guessing the Winter of Code is nonexistent because of other big announcements that have been happening; still, I would love to see some collaboration with …

http://www.regdeveloper.co.uk/2006/07/18/multivalued_datatypes_access/

This is an interesting read — it would be awesome if MySQL just used the “SET” or “ENUM” data types to be a placeholder for a join table, that it would create automatically for you. Of course, that’s a new level of functionality — MySQL does not implicitly create permanent tables with any commands. But it would be neat.

Thanx to Rich McIver for passing along this link:

http://www.businessintelligencelowdown.com/2007/02/top_10_largest_.html

I’m amused mostly because the article interchanges “database” with “data storage” — many of the sites have “digital documents” included in their count, and YouTube is in there completely with the amount of space their videos take up. But is all this stuff stored in databases? I do not think so. Anyone know for sure?

OK lazy web.... You have to help me out!

I try to monitor every performance metric I can within my cluster. Memcached efficiency, MySQL key buffer efficiency, etc.

One thing I can't benchmark is the efficiency of the Linux filesystem buffer (buffer cache).

From all the research I have done there's no way to to see the number of hits or total reads done by the file system cache.

Has anyone solved this problem?

The guys at Webyog have been great for producing a solid alternative to the GUI tools developed in-house by MySQL and I like meeting up with them each year at the MySQL Conference to see what they have been working on. Last year it was the release of their product as Open Source, this year it looks like they have decided to release early, and it’s an interesting offering.

According to a company blog, the guys at Webyog have released their own version of MySQL’s enterprise monitoring tool.

Their new MonYog tool is now available for early adopter download and looks like a nice offering for those who need to monitor multiple MySQL servers. If it turns out as nice …

In MySQL 5.0 mainly error control was improved, such as strict mode was added to change famous MySQL behavior of cutting too large strings, too big numbers and allowing you to use dates such as February 31st.

In one case however reverse change was done - in regards to storage engine initialization. Previously if you start MySQL and Innodb storage engine fails to initialize (ie you resized log file but forgot to delete old ones) MySQL Server simply would not start. In recent MySQL 5.0 series however it will continue loading and simply have Innodb storage engine disabled.

This can cause numerous problems especially if you got use to old behavior and do not check MySQL logs but simply check it is started. It especially hurts if you have only some tables in Innodb so you might not notice part of your application does not function. Also monitoring often monitors MySQL is up and running and will not query all tables are accessible not …

This article explains how to do subtraction in SQL over samples that wrap back to zero when they exceed a boundary.

OK lazy web…. You have to help me out!

I try to monitor every performance metric I can within my cluster. Memcached efficiency, MySQL key buffer efficiency, etc.

One thing I can’t benchmark is the efficiency of the Linux filesystem buffer (buffer cache).

From all the research I have done there’s no way to to see the number of hits or total reads done by the file system cache.

Has anyone solved this problem?

The first XAMPP release of 2007. ;) Updated in all three XAMPP versions: Apache 2.2.4, PHP 5.2.1, PHP 4.4.5, MySQL 5.0.33 and phpMyAdmin 2.9.2.

Get the downloads and more details on the specific XAMPP project page.