It is a common question: what is the lifespan of a persistent, pooled, cached, however-you-call-it PHP MySQL connection? What about CGI, FastCGI, web server module? The answer is always the same. Standard I/O file handles, including socket connections, are bound to processes. The lifespan of a persistent, pooled, cached, however-it-is-to-be-called-in-context-x connection is that of the PHP process. Depending on the web server deployment model, a PHP process handles one or multiple web requests. Read on only if puzzled.

A process

Do you have any colleguages around? Please, ask them for "APUX". In case of no reply, get a copy and study. To become a master PHP developer you will have to master todays frameworks (Zend Framework, Symfony, Dojo, …) and yesterdays. Yesterdays framework is still dominant. Yesterdays framework is your operating system. "APUX" stands for …

One of our users has published a “Quick Manual to SQLyog” to his blog.

Sometimes you do better writing about a subject as an ‘outsider’ than as an ‘insider’.  As an ‘insider’ you easily take things for granted that are actually not obvious for everyone.  This is a good example of this – and the writing is excellent. Enjoy!

Tweet

When you process MySQL slow query logs using pt-query-digest you can store samples of each query into query_review table and historical values for review trend analysis into query_review_history table. But it could be difficult to easily browse those tables without a good GUI tool.

For the visual browsing of tables created by pt-query-digest you may want to use some kind of web tools besides phpMyAdmin

Query Digest UI

This is a advanced, but easy to install, ui for pt-query-digest or mk-query-digest.

Main features:
* Dynamic filtering and searching of queries
* Colorized and normalized SQL syntax
* Explain the query …

MySQL is smarter than I thoughtSo I’m sitting at work, minding my own business, when suddenly: A wild query appears! And I do mean wild…I’m building a reporting class for a data hypercube that we have. The reporting class is basically a big query generator that takes a report object...

MySQL is smarter than I thought

So I’m sitting at work, minding my own business, when suddenly: A wild query appears! And I do mean wild…

I’m building a reporting class for a data hypercube that we have. The reporting class is basically a big query generator that takes a report object containing user-defined parameters for the report, and builds a query for the hypercube facts table, queries, post-processes the data where necesary, etc. If you’ve worked in any data-intensive company, you’ve built one of these.

I wanted to test out all the various builder functions in the class, so I configured a report that artificially included every single metric and dimension possible, and then to make sure that the query it was generating was at least syntacticly valid, I ran an EXPLAIN on it. I just wanted to make sure I didn’t mis-spell any field …

The end of September/beginning of October is a most exciting time if you dig MySQL & its diaspora. September 29-30 is MySQL Connect (register – early bird ends September 7) in San Francisco and October 1-2 is Percona Live NYC (register - early bird ends September 1) in New York City. I’m just attending MySQL Connect (thanks Dave for the ticket) and I’m speaking in New York so will take the redeye on Sunday. Four packed days of MySQL/MariaDB/Percona/tools/etc. across two coasts. Who can pass this up?

Who else …

It’s been a while since I’ve written any blog entries.  I like to think and write about basics and fundamentals, and in my work (Senior MySQL Support Engineer at Oracle), I can get caught up in particulars, so sometimes it’s hard to see the forest for the trees.

That’s why when we were offered a chance to try out Oracle’s Training on Demand (“ToD”) and take the new offering of “MySQL for Beginners”, I jumped at it.  I can go back to fundamentals, and now I can blog about the basics – in this case, the basics of Oracle’s Training-on-Demand – how it works, the experience, etc.

In my previous jobs as a DBA, I always enjoyed taking training classes (again, I like to focus on fundamentals).  I remember first (LONG ago) going to classes to learn linux.  (We felt initiated when, at the end, the instructor said “do a ps -ef and pipe it to grep for cron” and we all understood him.)  …

The servers I help manage have a lot of deadlocks, especially around a few central tables that are important to many business functions. The queries against them are complex, and they crunch a lot of data in some cases. As a result, we have long-running transactions that often deadlock against others, and there are even many short-running jobs that touch only a single row at a time that can’t get their work done sometimes.

Coming from the MySQL world, I’m used to being able to easily determine the replication delay (in seconds) via the SHOW SLAVE STATUS command: mysql> show slave status\G *************************** 1. row *************************** ... Seconds_Behind_Master: 0 ... 1 row in set (0.00 sec) Unfortunately, there is no such comparable command in PostgreSQL. The official docs propose [...]

Members of the MySQL Support Team wear a number of different hats here at Oracle.  Obviously, our top priority is to provide amazing technical support that makes customers rave.  We also have a team dedicated to processing bug reports from the MySQL Community.  Some of us are active bloggers or assist on mailing lists or forums, while others find other ways to contribute to the MySQL Community.  We help out with QA and product planning, write …