Why? Highbase is currently comprised of several shell scripts  and some C code. It’s actually a good project (talk about self promotion) that hasn’t reached a stable release yet just because It hasn’t been tested enough in production environments I’ve been amazingly busy during the last years. Lots of work, and lots of parenting in … Continue reading Rewriting Highbase in Erlang →

No related posts.

This post gives some brief tips for debugging Mysql.

First of all you can run Mysql on many different platforms. So start by choosing the platform that will make your particular debugging task easier. For memory leaks I would run the server on linux so that I can take advantage of the Valgrind tool, for memory leak checking, which is only available on linux. For bugs where you want to use a debugger to step into the code, Windows is a surprisingly great option due the ease of use of the debugger that comes with Visual C++.

Now that you have your debugger of choice up and running, you want to see what is Mysql doing in some scenario. Mysql code is quite complex so finding the right place to put the breakpoint can be a challenge. If I don't know what function calls mysql is making, normally I would start with a breakpoint in sql_parse.cc in the mysql_execute_command function at the point where the code says:
…

I’ve had about enough of Microsoft and its approach to help small businesses develop application. The .NET platform is FAR FAR from ‘working everywhere’ and being compatible to everything. Let me write down some points to what I mean.

Windows Applications that are developed in Visual Studio.NET by SMBs usually (almost always):

• Do not work on other machines, even when its .NET. Don’t assume a .NET application will work on every machine. You need an installation of visual studio for each machine you want your application to work on.
• Do not work with Crystal Reports. It might work “ok” on your computer, but if/when it doesn’t print on someone else’s machine, try to find some support.
  
• Stop working when you update …

SQL is great. There are millions of people who know it, it’s semantics are standardized (well, as much as anything is ever I suppose), and it’s a great language for interacting with data. The folks at SQLstream have taken a core piece of the technology world, and turbocharged it for a whole new application: querying streams of data. You can think of what SQLstream is doing as the ‘calculus’ of SQL - not just query results of static data in databases but query results of a stream of records flowing by.

Let’s consider some of the possibilities. Remember the results aren’t a one time poll - the values will continue to be updated as the query continues.

// Select the stocks Average price over the last 10 ticks
select stream stock, avg(price) over (previous 10) from stock_ticker group by stock …

One very helpful use of the technique Sheeri described in Remote connections without leaving the mysql shell is making sure that replication is working properly.

According to the MySQL Reference Manual’s section on SHOW SLAVE STATUS Syntax, it shows information corresponding to the slave thread in the slave server. When replication is broken, however, or not working properly due to network issues between master and slave, this information may not be accurate. This has improved over recent releases, but it’s still not perfect.

The question, then, is: how to be 100% sure (or as close as you can get to 100%) that replication is running fine? The answer, as offered by Sheeri: use CONNECT.

Example …

MySQL has a stored procedure language. People have told me how bad it is, how inefficient, how it had to be there to make MySQL appear more professional. Yes, it’s far from perfect, it’s not pre-compiled so not necessarily any faster than running single statements by hand. However, it works and you can do many useful things with it. The complaints however miss the point of stored procedure languages as used inside a database. They are generally used for 3 things:

1. To hide the implementation from the database user, so that the implementation can change but the interface remains stable.
2. To simplify the DBAs day to day tasks of maintaining the database.
3. To speed up complex queries.

While MySQL doesn’t excel at point 3, it can be used in the other two cases. I’ve used Sybase for quite a while and many maintenance tasks could be fully programmed inside the database, just as you would write a …

Another project that I am doing introduces a new shard type. First, think of a Shard as RAID-10 on your database layer. The data is striped across N servers and mirrored for failure recovery.

Shard Types are Shards that serve specific purposes. For instance, I have an archive shard. This shard keeps data the is rarely ever read on even cheaper hardware-with slower disks and the data is naively compressed using the innodb_file_format=Barracuda
option. If and/or when a person requests data that is archived, my software layer detects this case and migrates the archive data from the super crappy RAID-5 servers to the less crappy RAID-10 servers with 32 GB of RAM, all within 5-10 seconds.


Another shard type is the sub-shard. I use this terminology to define a different way of federating data for a predefined global object_id. A predefined global object_id is what you are federating by, for instance userid. …

I’d like to take a step back from technical issues to distill some of my thoughts on the challenges of data warehousing in the 21st century.

Having worked on a number of warehouse projects in different industries over the years, I’ve encountered many challenges, some failures, some successes. One thing is certain: all organizations that have a reasonable amount of data should be building a data warehouse if they don’t already have one. In 2009, given the economic atmosphere, no one wants to wait as long, or pay as much, as they did in 1999 to get one.

While this is a huge opportunity for open-source competitors like MySQL, it comes with big challenges for an organization that thinks it will get a $10MM warehouse (in 1999 dollars) for $300,000 (2009 dollars).

My contention is that in a web-connected, high-traffic and high-speed world, a monolithic approach with a rigid set of requirements, and a project team isolated …

Yesterday there was a short discussion on our public IRC channel (#workbench on freenode). It was about how Workbench handles Foreign-Key-Names.

Especially if you intend to use Synchronization feature (available in SE edition only) its good to have names for FKs (and their coresponding Indices). Workbench does that by default if you’re using the Relationship-tools (Hotkeys 1 to 6 when canvas is focused). But if you prefer to create your FKs in a not-so-visual way (entering the values on the FK-Tab in the table-Editor) you are on your own when it comes to naming your FKs. Though Workbench initially only accepts creating keys after you typed a name into the name-field theres still the possibility to rename the FK later to even empty names. If you try to create empty named FKs on a MySQL Server the server will generate a name for the key.

If you want to change FK names later, when your model has already grown to a size, where …

This tutorial explains the technique to generate unique identifiers that are extremely difficult to predict.