Update: As some people have pointed out, if this is on all the time in a production environment, it could be bad. We only enable it when debugging. But, in the interest of keeping the PHP world safe, I have edited the code.

Preface: I would not use this in a production environment on a full time basis.  When we do use it, we turn it on, gather some logs and turn it off.  Also, it has been pointed out to me that this will cause MySQL to not use the query cache in versions previous to 5.0.  That is another reason to not use this in a production environment full time.

So, on dealnews.com we often end up with a lot of queries that look kind of the same. It usually because we have a complex query that simply selects article ids from one or two tables and then a second query to select the data from the article table using those ids. If you have used MySQL …

As I explained in this blog, it’s always a difficult decision for me to keep postponing my own project in order to work on the patent policy front. But once again, like so many times before, I have decided to do so for some more time. Given what happened at the July 12 hearing, there are some important things to do right away. I’ll talk a little more about my personal future on some other occasion, but suffice it to say that I’m still actively involved in the European patent policy debate!

This spring I have added some new modules to our system. The old ones are all based on Microsoft SQL Server, but for the new ones we decided to go with MySQL as a part of a strategy of moving to MySQL with the whole system. We decided to use InnoBase as a storage engine and Hibernate for accessing the database from our Java application. No strange decisions, I believe.

The development phase is now over, with no major problems, and the system is up and running. During this autumn we will see how it works, but I'm nit very worried, it seems stable.

Unfortunately the customers strategy is today unclear, so we're stuck in a situation with two databases. Personally, I feel that using MS SQL Server in a Java environment is a questionable approach, since Microsoft is positioning the product as a part of the .net platform. And the 2000 version, which we are currently running, is getting old...

The wonder of film scanners. I have found a whole lot of old pictures on scratched up black and white film and dusty slides and with the scanner they are coming back. I'm considering doing a "picture every other day/week/..." thing. Hmmn..

Here's a picture from New York in the summer of 19925. Since this is my geeky weblog, then let me put it in those terms: Before I did Perl.Several years before I used OS/2. Four yearsA year before my first Linux box. Years before I used MySQL.

Oh, and before 2001 - obviously.

(See also large size, or the Flickr Entry if you want to mark it a favorite).

update: maybe it was 1995 and not 1992...

Wrapped up a few final touches this evening on the memcache engine:
http://tangent.org/index.pl?lnode_id=506

What works:
SELECT, UPDATE, DELETE, INSERT
INSERT into foo SELECT ...

What doesn't work?
Probably ORDER BY operations.
REPLACE (I think)
IN ()
NULL
multiple memcache servers (this would be cake though to add)
table namespace, right now it treats the entire server as one big namespace

There is probably a lot more I haven't thought up that is now working. Right now the version I put up only allows you to have a primary key and one attribute. Doing multiple attributes means coming up with a way to store MySQL row format, aka UNIREG, in something which would be easy for anyone to take apart. Mark suggested XML, but XML would be damn slow to parse.

If …

I've been writing a lot of articles about locks, deadlocks, and transactions recently, and it occurs to me I've neglected to mention the single most important thing to know. If you only learn one thing about transactions in database systems, you should learn this, and learn it thoroughly -- burn it into your brain permanently, if possible.

Red Herring magazine (yes, it still exists) has several articles in the August 21st issue commemorating the 15 year anniversary of Linux.  Linus Torvalds first posted information about his open source operating system on usenet just 15 years ago, the same year that the world-wide web started.  Perhaps it is no coincidence that Linux and the LAMP stack (Linux, Apache, MySQL, PHP / Perl / Python) have become the dominant platform for Web applications. 

Linux has continued to be the fastest growing server operating system in recent years.  According to Gartrer, it grew at 35% in 2005, to about $7 billion, outstripping the growth rate of Windows by a wide margin.  While Linux has not established itself on the desktop in any signficant volume, there are signs that even that could change as …

MySQL AB announced today that its internal benchmark team has won an international contest sponsored by the renowned IT industry magazine c´t for the fastest e-commerce database application. The magazine's editors held the contest to evaluate database performance in real-world business use by creating a standard online inventory system.

Peter Zaitsev over at the excellent MySQL Performance Blog recently wrote an article on duplicated and redundant indexes – any indexes which cover exactly the same columns as another index, or cover a leftmost prefix of another index. While there are subtleties, such as FULLTEXT indexes not being the same as non-FULLTEXT, for the most part this is sufficient criteria to raise possible duplicates to a DBA’s attention. I opened my big mouth in the comments and said I could write a quick Perl script to discover possible offenders in just a few lines of code.

Well, it’s Sunday night so I will put this down to being the weekend. The background to being caught out is a request I made to my local Users Group mailing list for some information on people’s environments because I wanted to some empirical data analysis without having any more knowledge of the systems.

In summary (without the surrounding fan-fare, I was seeking):

SELECT VERSION();
SHOW STATUS;
SHOW VARIABLES;  // Optional

I was however perplexed why my first data point analysis (Read/Write ratio) using the Status values Com_insert, Com_update, Com_delete and Com_select was not always giving me expected results. In particular, a number of server results showed 0 for values while I knew the results came from working MySQL environments.

So, sanity check with good friend Morgan and I get the response to answer the dilemma SHOW STATUS defaults to session …