Its been a while anyone talked about Intel’s Hyper-Threading performance when it comes to databases. There were enough posts about disabling Hyper-Threading  completely when it comes to MySQL/InnoDB workloads way back when we had enough issues with scalability of InnoDB on multi-core systems. But things has changed quite a bit in the fast year or [...]

Introduction

If your WordPress comment counts got messed up, whether because of a plugin (I'm talking about you, DISQUS) or you messed with your database manually and did something wrong (yup, that's what I just did), fear not – I have a solution for you.

But first, a little background.

Comment Counts In WordPress

Here's how comment counts work in WP:

• Posts live in a table called wp_posts and each has an ID.
• Comments reside in a table called wp_comments, each referring to an ID in wp_posts.
• However, to make queries faster, the comment count is also cached in the wp_posts table, rather than getting calculated on every page load.
  If this count ever gets out of sync with the actual number of comments for some reason, WordPress, while still displaying all comments properly, will simply show the wrong count. …

Finally, it is available for orders: The MySQL Admin Cookbook is the first book I have written and it is now ready for shipping. You can find more information about it at the publisher's website, including a free sample chapter!

I have yet to receive my sample copy, however that should not keep you from going ahead early and ordering it, e. g. from amazon.de. For your convenience, here is a direct link:

On Amazon you can also have a look at the index and a few pages of the first chapter.

More international points of sale can be found on the publisher's website, …

This is a small revision and will only be available through SVN. However, it is an important one to speak about as it solves a former issue when running the application on a screen smaller than 1024px wide. While most users may not have noticed this since they have larger monitors it has been noticed [...]

I use it in Oracle and notice there are 10 days missed, for example:

ORCL> select to_date('4/10/1582','dd/mm/yyyy') SHOW_DATE from dual

SHOW_DATE -------------- 04/10/1582 ORCL> select to_date('4/10/1582','dd/mm/yyyy') + 1 SHOW_DATE from dual

SHOW_DATE -------------- 15/10/1582 Say What? the date after 4/10/1582 is 15/10/1582.

But in MySQL i try it but i didn't see this case, example:

mysql

Website health checks are a crucial service to an operations team. In addition to in-house monitoring and service state reporting it’s also important, even critical, to have an impartial third party to run checks to test your customer facing services. There are a lot of companies in this arean that would be glad to have [...]

After some testing and setup we have decided to use Eventum for our support ticketing needs. This featureful system will be in use for all of your support questions related to Kontrollbase – the MySQL analytics and performance tuning web application, as well as Kontrollkit – the collection of server automation scripts. You can read [...]

Finally: Not quite 65 million years in the making (Jurassic Park, hint, hint), but it took about 15 months to get my first book to the printer. A few days ago Udo – my co-author – and I approved the final version of the MySQL Admin Cookbook for publishing. From what I see the book has not been added consistently to the online book stores around the net, but I will most certainly put links on here, as soon as it is available.

I think, before I begin the next one, at least a few months of relaxing are in order….

During my previous post on how to improve derived tables performance, I patched the code to add indexes forcefully on internal derived table results, which made a huge difference in the performance. It was just an experiment and a thought to see if it really works without re-writing the queries, so that the logic can [...]

Last week I was working on one of the issue where the sub-query related to OLAP staging was running for about 2+ hours in the production server and finally nailed down to get the query to run in < 7 secs. It was bit interesting and kind of known issue in MySQL sub-queries world and one of the limitation from MySQL on giving more control over derived table results.

Sometimes we can re-write the sub-queries so that there is no derived tables complexity involved; but the bad part is; this particular sub-query is part of an UPDATE statement; so not all sub-queries can be re-written especially when they are part of UPDATE or DELETE statements due to its own limitations.

PROBLEM:

Here is the subset of the problem query and as you can see it runs for about 6 minutes in this small subset of data that I used for testing on Mac. All tables are InnoDB based.

-------------- …