The recording and slides for my webinar on zero-downtime schema changes with MySQL are available now. Don’t miss Vadim’s webinar tomorrow!

Just as a reminder – tomorrow, May-9, 2012, at 11am PDT I will be giving a free webinar “MySQL and SSD”. This is the same talk I gave on Percona Live MySQL Conference, so if you were there – you probably will find nothing new. Otherwise, you are welcome to join!

Last month, Ronald Bradford, Giuseppe Maxia and Mark Leith spoke about how to simulate a mysqldump –ignore-database.
This mysqldump option doesn’t exist and these three guys have given us various helpful solutions.

But for those of us who use ZRM community to make backup with mysqldump, the –exclude-pattern seems to do the job :

--exclude-pattern "pattern" 
All databases or tables that match the pattern are not backed up. 
If --all-databases or --databases are specified, the exclude pattern applies
to …

Background Knowledge

You try starting, stopping or connecting to MySQL and receive the following error, “Error: Found option without preceding group in config file: /etc/mysql/conf.d/char_collation_set.cnf at line: 1″. The error message my vary but comes to the same issue. MySQL may not start or might experience connectivity issues.

This issue was experienced on Debian Squeeze v6.04 AMD64 system with MySQL v14.14 Distrib 5.1.61.

Solution

This issue is caused by a improperly formatted MySQL configuration file(s) and refers to one missing the group heading (e.g. [mysqld], [mysqld_safe], etc.).

Source: MySQL 5.1 Reference Manual :: 4.2.3.3. Using Option Files
Source: …

The MySQL Connect conference call for papers is closed and the response was amazing. Now the problem is that we have a few hundred great session and only two days and nights to get everything presented. The submissions range from the big companies to private individuals and old, familiar faces in the MySQL Community to some new folks with top notch material.

Database as a service, cloud computing, tuning, InnoDB secrests, cluster implementation, Oracle/MySQL data transfers, replication tricks, query enhancement, big data, stored routines, J2ee, server refactoring, backups, NoSQL/SQL, and other subjects are covered — and those were just the first few submissions!

So thank you if you submitted and please do not forget to register!

…

I’ve previously provided an example of using MySQL Cluster Manager to add nodes to a running MySQL Cluster deployment but I’ve since received a number of questions around how to do this in more complex circumstances (for example when ending up with more than 1 MySQL Server on a single host where each mysqld process should use a different port). The purpose of this post is to work through one of these more complex scenarios.

The starting point is an existing cluster made up of 3 hosts with the nodes (processes) as described in this MCM report:

mcm> SHOW STATUS -r mycluster;
+--------+----------+-------------------------------+---------+-----------+---------+
| NodeId | Process  | Host                          | Status …

You might have heard that Oracle made the decision not to support MySQL for IBM i any longer. This is certainly understandable. However, there are still users who want to continue running IBM i and MySQL.

That’s why we’re happy to announce that we have assisted Zend to introduce Zend DBi as a drop-in replacement for MySQL on the IBM i platform. Zend approached us to say that they want to ensure there’s a way forward for IBM i users, and asked if we’d help them. We’re delighted to do so.

The result is Zend DBi. It is basically a renamed build of MySQL for IBM i. It’s a 100% compatible drop-in replacement. Everything works on Zend DBi just as it works on MySQL, with no need to rewrite applications, management scripts, or anything else. There is no difference in query …

Portable table space part III or We can do it (with partition)

First of all, I have to say that as usual I did the three investigation, because I had fun.

MySQL is still fun for me, and also if I am not touching the magic topic of the moment like SSD,

replication and so on, I enjoy what I was doing, and I see how I can save my customer data, with this.

Second, I thanks again the people who helps me in understanding better MySQL, in this case Mattias Jonsson, who points me in right direction.

The Solution

The solution is a little bit cumbersome, but it works.

I have assume that:

• a customer have a data-set which contains a partitioned table
• we can have short period of read only data

To use the DISCARD tablespace on a partition the process is:

1. - LOCK the table with partition and the final table where I will put …

Portable table space part III or We can do it (with partition)

First of all, I have to say that as usual I did the three investigation, because I had fun.

MySQL is still fun for me, and also if I am not touching the magic topic of the moment like SSD,

replication and so on, I enjoy what I was doing, and I see how I can save my customer data, with this.

Second, I thanks again the people who helps me in understanding better MySQL, in this case Mattias Jonsson, who points me in right direction.

The Solution

The solution is a little bit cumbersome, but it works.

I have assume that:

• a customer have a data-set which contains a partitioned table
• we can have short period of read only data

To use the DISCARD tablespace on a partition the process is:

1. - LOCK the table with partition and the final table where I will put …

I have been doing some research lately on various mySQL related features and found myself distracted by mySQL 5.6. If everything that I am reading is correct, this should be the best mySQL version yet.

mySQL 5.6 will have multi thread replication, making time delayed slaves-an actual feature and not a byproduct of replicating high write volumes. Also, crash safe replication! When a master crashed and corrupted the binary log this was a pain in the ass to fix. In the past I had to write scripts to walk the primary key and do a checksum on each returned table chunk and pick which row was correct and which one was not.

The performance optimization for innodb that addresses some stalls is as exciting as multiple SQL threads for replication. One major change is in the stat layer, which was throttling throughput for in memory workloads at high thread concurrency. You can read the details …