Fri, 2015-01-09 13:19maria-luisaraviol

MariaDB MaxScale is now RC and together with all the MariaDB team that has been involved in the project we need to thank all the companies that agreed to become part of the MaxScale Beta Test Plan. This major step in the MaxScale life (read more here) had an important impact on the MaxScale QA process.

We have asked some companies to help us in testing MaxScale in “real” environments with different custom settings, different configurations and with traffic load as close to reality as possible.

Colt Engine, an Italian hosting company based in Turin (Italy) applied immediately. They have been a great fan of MariaDB MaxScale and have been enthusiastic early testers of MaxScale since early 2014 when it still was Alpha. Actually they have suggested some of the MaxScale filter plugins.

I started …

Covering Indexes not being chosen by optimizer I noticed this problem about a year ago when writing this blogpost. In short problem is when adding a covering index and keeping old non-covering index the optimizer opted to use old non-covering index. Only solution was to FORCE optimizer to use covering index which meant you needed to modify your DML or remove old index.
Using the same test setup as in my old blogpost but when you add new covering index do not drop the old index.
So, instead of running: mysql> ALTER TABLE big DROP INDEX CountryCode;
mysql> ALTER TABLE big ADD INDEX conPop (CountryCode, Population);
We run only statement for adding new covering index and do not remove old index: …

Join 29,000 others and follow Sean Hull on twitter @hullsean. Oh RDS, you offer such promise, but damn it if the devil isn’t always buried in the details. Diving into a recent project, I’ve been looking at upgrading RDS MySQL. Major MySQL upgrades can be a bit messy. Since the entire engine is rebuilt, queries […]

"Should vegetarians open steakhouse restaurants?"

Though someone will probably give me several examples of why they should, I'll argue that they absolutely should not. How can someone who doesn't eat steak convince others to eat at their "steak-only" restaurant?

But this is something a "professional technology benchmarker" (PTB) struggles with on a regular basis. Hello, I'm Tim Callaghan, and I'm a PTB.
professional technology benchmarker, or PTB (noun) : One who compares two technologies as part of their job. One of these technologies is usually the product of the PTB's employer, the other is almost always not. In a past experience I was tasked with comparing the performance of a fully in-memory database with Oracle and MySQL on a "TPC-C like" workload. At the time I was an Oracle expert and working for the in-memory database company, but had never started a single MySQL server in my life. At …

VividCortex now supports per-user metrics in Top Queries, so you can monitor MySQL database user activity in detail. By changing the Rank menu from Queries to Users you can now see a breakdown of user activity in the familiar format, all with 1 second resolution. Here “user” means the database user account that the application uses to access the database.

Computed columns also work as usual so you can quickly figure out which users consume more resources such as CPU or IO in addition to the standard metrics (count, time, latency, error rate, etc). Additional information such as per-host user activity and scatterplots of additional data can be found after clicking to drill into the query.

We detect the user by sniffing from the TCP traffic, as usual. There are two places we can capture it: during connection, or from a COM_CHANGE_USER packet. Sometimes we can’t detect a user for a connection in which case we show …

This is yet another story of Django web-framework with time zone support and pain dealing with python datetimes and MySQL on the backend. In other words, offset-naive vs offset-aware datetimes.

Shortly, more about the problem. After reading the official documentation about the time zones, it makes clear that in order to reflect python datetime in the necessary time zone you need to make it tz-aware first and than show in that time zone.

Here is the first issue: tz-aware in what time zone? MySQL stores timestamps in UTC and converts for storage/retrieval from/to the current time zone. By default, the current time zone is the server’s time, can be changed on MySQL globally, per connection etc. So it becomes not obvious what was tz of the value initially before stored in UTC. If you …

Twitter is not my favorite social media site. Using twitter is like driving downtown, screaming what you want to say out the window, and hoping someone hears you. There might be tens of thousands of people downtown, but your message will only be heard by a few. Your best bet is to repeat your message as often as possible.

However, twitter is free and if you want to reach as many people (theoretically) as possible, you might as well use it. But sending tweets on a scheduled basis can be a pain. There are client programs available which allow you to schedule your tweets (Hootsuite is one I have used in the past). You can load your tweets in the morning, and have the application tweet for you all day long. But you still have to load the application with your tweets – one by one.

A friend of mine asked me if there was a way to send the same 200 tweets over and over again, spaced out every 20 minutes or so. He has a consulting …

This is a follow-up post to my recent announcement of only_full_group_by improvements in 5.7, where I’d like to address some tricky corner cases where GROUP BY contains non-column expressions (functions).

In the SQL standard, GROUP BY should contain only table columns and never expressions. So this example query where we want to count how many people have the same name is illegal in the SQL2011 standard because the GROUP BY clause contains a function:

CREATE TABLE people (first_name VARCHAR(100), last_name VARCHAR(100));

SELECT CONCAT(first_name, ' ', last_name), COUNT(*)
FROM people
GROUP BY CONCAT(first_name, ' ', last_name);

But MySQL does allow expressions in a GROUP BY clause, and it’s a very convenient addition to the standard. The logic …

We want you to learn as much as possible during the three days of conference. We do that through quality and variety of both content and speakers, as well as creating a fun and friendly atmosphere.

We have presentations for any level, from beginner to advanced. You’ll learn about the backend and frontend, web and mobile, information systems and games, hard and soft skills, as well as many related topics.

We have speakers from many different countries and industries. They brings unique perspectives that you won’t find in your backyard. How often will you talk tech with someone from Taiwan or Saudi Arabia? How often will you learn from people who build robots or write their own programming language?

If your goal is to learn and meet interesting people, then ConFoo is the conference for you! Take advantage of our $190 early bird …

Imagine that two colleagues, Alice and Bob, issue a command to launch a new virtual machine at approximately the same moment in time. Both Alice’s and Bob’s virtual machines must be given an IP address within the range of IP addresses granted to their project. Let’s say that range is 192.168.20.0/28, which would allow for a total of 16 IP addresses for virtual machines [1]. At some point during the launch sequence of these instances, Nova must assign one of those addresses to each virtual machine.

How do we prevent Nova from assigning the same IP address to both virtual machines?

In this blog post, I’ll try to answer the above question and shed some light on issues that have come to light about the way in which OpenStack projects currently solve (and sometimes fail) to address this issue.

Demonstrating the …