Posts by Sheeri Cabral
Architecture of Maria: A New Storage Engine with a Transactional Design
Why the sandbox? To be able to set up 1 server in under 10 seconds. And to be able to set up multiple MySQL instances very quickly, and to use them quickly. The sandbox untars in seconds, for installing alternative servers, not main instance, it creates a separated environment (datadir, port, sockets) — for groups of related or unrelated servers. Really good for testing out new server versions. Save yourself some time and read this post to find out how to automate MySQL Sandbox!
Yesterday I presented “Best Practices for Database Administrators” at the MySQL User Conference and Expo. There are links to the slides and video in this post.
Jonathan Schwartz’s Keynote at the 2008 MySQL Conference can be played from the embedded video at the bottom of this post or directly at YouTube (follow the link in this post)
If you can’t attend the MySQL Conference, you can still virtually attend. 2-3 pm PDT on Tuesday, April 15th for the “Best Practices for Database Administrators” session, and, 11:55 am -12:40 pm PDT on Thursday, April 17th for the “Database Security Using White-Hat Google Hacking” session. You need nothing but your web browser; unlike some other live meetings, audio streams from the website too, so there’s no phone number to call or participant code or anything. There’s also a live chat so you can participate and ask questions while the session is going on.
There’s no video for Jacob Nikom’s December 2007 Boston MySQL User Group meeting, but the slides for “Measuring MySQL Server Performance” can be downloaded here.
One of my colleagues made a typo in a query today that led to me discovering this issue — I think it’s a bug, though it may just be how the language is defined. But certainly a subquery that cannot run should not act as if it returns TRUE. Part of me thinks there’s an implicit join going on or something, but I’m not sure how that’s working. Have a look.
I spent way too long figuring out code that would simply load balance based on, always go to one server, go to another server only when the first server is down., again running into the problem where the manual hasn’t been updated. I have made a Forge snippet of this code, but it does not hurt to post it here.
While doing a standard audit for a new client, I recommended a few changes to get better performance. Because I had several changes, I used the documentation and found that innodb_flush_log_at_trx_commit is a dynamic variable. I was surprised, because most operations dealing with file sizes and operations are not dynamic. So I searched for others who may have had the same error, and ended up getting an unintentional googlewhack.
Calling a group of people with common interests “community” is just as meaningless as saying I live in a “neighborhood”. There has to be a bond there. I am proud to be a part of the MySQL Community, which actually has forged bonds. If you are new to the MySQL community, feel free to come up and talk to me (or anyone, really) — during the conference, or otherwise. Even if you feel you have nothing to say, just say hello. And I must end with a disclaimer: I won last year’s “Community Advocate” award from MySQL, so I guess all in all, I’m still a community advocate.