In my recent work I had to dive into the Internals of the MySQL protocol.

Most of the protocol is described in Chapter 8. MySQL Client/Server Protocol. "Most" as it documents the state of MySQL 4.1.x before it was stable. It is interesting for all developers who want to extend the mysql-server or want to write a native connector to MySQL which isn't using libmysqlclient.

Note: the protocol definition and the derived work at released under the terms of the GPL. See 8.1. Licensing Notice

Connection States

The MySQL protocol has two phases: auth and …

Until recently I thought that currently popular scripting languages, which mostly evolved over last 10 years or something, must allow for easier portability across different platforms compared to ye good olde C/C++.

After all, their development started a few decades after C, so its notorious caveats are all well-known and should be easy to avoid when designing a new language, right?

However, PHP just brought me a new definition of "portable" - and that was when working with... integers.

PHP is not able to handle unsigned integers, and converts values over 2^31 to signed. So if your IDs go slightly over 2 billion, and PHP decides to treat them as integers, you're in trouble.

Oh wait, no - that's on 32-bit platforms only! PHP int size is platform-dependent, and it seems to be 8 bytes on our 64-bit boxes. Yes, the very same ones where C/C++ int is 4 bytes, you know.

That was the easy part. It was …

One may think changing PHP session handler from file based to database driven is fully transparent. In many cases it is, sometimes however it may cause some unexpected problems as happened to one of our customers.

If you use file based sessions PHP will lock session file for whole script execution duration, which means all requests from the same sessions will be serialized on PHP level, which means they also will be serialized for single user on database level. If you change to store PHP sessions in MySQL instead this effect may be no more true and you may have number of requests executing for the same session at the same time. First of course means you may have your session data damaged because you will have lost session variables update from one of the script, in addition however you may run into database related issues of modifying user profile or other user/session related data in parallel, if you do not use transactions or lock tables. …

I am giving a talk at the MySQL user conference, April 23-26, Santa Clara, California. The talk is titled "Building Scalable OLAP Applications with Mondrian and MySQL" (slides).

When I gave a talk "Building OLAP Applications with Mondrian and MySQL" at the 2006 MySQL UC, it was my first chance to meet a lot of you in the mondrian and pentaho communities. It was great to finally put faces to email addresses. But it was frustrating that there wasn't enough time to have in-depth discussions and answer everyone's questions.

So this year we thought we'd solve that by hosting a Birds of a Feather session on Tuesday …

Its interesting to note some happenings in the MySQL world of late, that might be of interest to people in the database world, and those following open source software development and business models.

40% of developers say they use MySQL, according to the Evans Data Group. This is not including pilot projects, but real production use in corporate environments. A lot of MySQL’s popularity is generally attributed to the LAMP stack, though I see a change. Look at all the Ruby on Rails projects out there. They most definitely run on a MySQL backend. A good example are the products from 37signals, makers of the rather new, and cool tool, Highrise - they’re Ruby on Rails, and MySQL powered.

Is this 40% statistic …

After 29 years, InfoWorld is giving up it's print edition and going to a web-only publication model.  It's hard to believe it's been around that long.  I remember reading InfoWorld when I was in University back in the early 1980s.  Back then, the only way to know what was new in technology was through magazines like InfoWorld, ComputerWorld, PC Week, PC Magazine et al.  Over the years, InfoWorld has had some of the greatest writers and editorial staff in the industry.  And they were able to morph on occasion from tabloid to glossy magazine (and back) a couple of times, adding new columnists to focus on new techologies to stay current.  But it's been apparent for a while that InfoWorld's print edition was …

Welcome everybody!

Today we have presented the first webinar of the series.
Let me remind you the dates for the all the sessions:

• 27th March: Part 1 - High Availability and Scalability Architectures
  
• 19th April: Part 2 - Advanced Scalability Solutions
  
• 2nd May: Part 3 - MySQL Enterprise To Control Mission Critical Online Services
  
• 23rd May: Part 4 - 99.999% High Availability solutions
  
• 13th June: Part 5 - MySQL Enterprise performance and benchmarking
  
• 27th June: Part 6 - Advanced HA solutions
  

Thanks to all of you who showed up!

-ivan

Today, the time closed to apply for student positions at MySQL Summer of Code. We are very fortunate and thankful to have got 35 applicants fulfilling the formal requirements of an application fully entered and coupled with a task announced on http://forge.mysql.com/wiki/SummerOfCode — thank you!

Our next task is to narrow this group down to the number of projects which can actually start. This will be a much smaller number. Even in the theoretical case that Google would allow us to have 35 projects (and they won’t), we don’t have the internal mentoring resources to guide that many Summer of Code projects.

Colin Charles has already been quite busy replying to the incoming flood of applications, insisting on them being more specific than “I’m in, just …

… is how to get the last day of current quarter.

CREATE FUNCTION last_quarter_date (thedate DATETIME) RETURNS DATE
RETURN last_day(CONCAT(YEAR(thedate) , ‘-’, QUARTER(thedate)*3, ‘-01′));

And I can not find other solutions. It is interesting if it is the only way.

Okay, so one of the disks in a JBOD (well… single LVM) has been on the way out (hopefully can recover some stuff off it… there’s nothing completely important… but still).

I’ve now learnt and desktop has three new 320GB drives in a RAID5.

Currently installing Ubuntu 7.04 on it. I do have to say that the alternate install disk (which uses debian-installer) has a REALLY nice RAID and LVM setup now. If only it also let you pass parameters to mkfs it would be ideal.

Update: It got the bootloader horribly wrong though and I’ve gotten to piss-fart around trying to get LILO to install and boot. Current result? Blinking cursor in top left of screen. Fantastic… fucking fantastic.