According to Johans wishes I write down my concerns about MySQL (Cluster) replication. These items are things I run again and again into it with customers:

• SQL-nodes are still loosing too easy connection to cluster after data node or management node restart (which leads into gaps, see next point). Automatic fail over or reconnection is just a dream (maybe it works in about 90% of the cases at least)..
  

• Gaps: Whenever I do a cluster reconfiguration (should not be necessary too often, see loosing connection above) or a mysqld restart I get a gap (these are the planned ones, see also automatic channel fail over). Then we have the not planned ones...
  I cannot understand, why we are not able to keep at least a certain amount of traffic in a binlog-injector-buffer to avoid the majority of these gaps. This could be something like a Round-Robin buffer which stores all …

Found or rediscovered a bunch of nice software this week

• Spreed.com web-based meetings, with shared whiteboarding, application sharing, video, and soon mindmaps.  Why is it noteworthy? They explicitly support linux, even for whiteboarding and app sharing. The whiteboard is geared toward making notes on top of a presentation, and participants can easily generate a PDF of any page. The linux screen sharing client is written in python, and the mind maps (coming soon) promise to be compatible with FreeMind.
• Coccinella instant messaging client with shared whiteboards. What I really want is OmniGraffle for Linux, with the ability to share drawings with someone on the other side of the world and both edit in realtime. Inkscape’s Inkboard feature would be even better, if it worked, but it …

• PostgreSQL: Documentation: Manuals: PostgreSQL 8.3: PostgreSQL 8.3.3 Documentation
• Comparison of different SQL implementations
  How to do things in Mysql/Postgres/other databases handy cheatsheet

The book is done now, right? What’s next? Don’t tell my wife this, but a book is never done. Right now I’m proofreading the printed copy. I proofread PDF after PDF during production, but some problems will always slip through and make it to paper. I’m finding quite a few little mistakes. For example, at one point we refer to TPC as TCP three times in a row. Oops. These problems will be corrected in the next printing.

Dear reader, this is a challenge. How’s your MySQL prowess? You know about LIMIT: it cuts off the results at the specified number. mysql&gts; select actor_id from sakila.actor where actor_id % 5 = 0 limit 5; +----------+ | actor_id | +----------+ | 5 | | 10 | | 15 | | 20 | | 25 | +----------+ 5 rows in set (0.00 sec) But that query actually accessed 25 rows. What if I want to say “return up to 5 rows, but don’t read any more than 20 rows to find them?

My last post on June 10 was about when MySQL received queries from php scripts that looked like this:mysql_select_db = ('database_I_DO_want_to_replicate', $link);$sql = "INSERT INTO db_I_DO_NOT_want_to_replicate.repl_table (a,b,c) VALUES (1,2,3);";-- and thus could not pass this over to the slave unless I set Replicate_Wild_Ignore_Table value IN ADDITION to Replicate_Ignore_DB as such in my

I wanted to try using MySQL source from its new VCS, Bazaar, and I saw Daniel Fischer's extremely helpful post "Getting started with Bazaar for MySQL code". Thanks Daniel!Downloading using Bazaar is indeed slow, as Daniel warns. I am constrained to 1MBps ADSL, and the intial download took quite a bit of time. I think over an hour. But subsequent merges and other operations are

podcasting

Originally uploaded by Elliot Murphy

my podcast rig

Welcome to the 103rd edition of Log Buffer, the weekly review of database blogs.

Starting with Oracle stuff, Chen Shapira (just a simple DBA on a complex production system) is looking for great PL/SQL. Why? To become a better PL/SQL programmer. “But,” she writes, “for PL/SQL , I’m a bit stuck. I can still read my own code for bad examples, but where can I find examples for great code?  . . .  Somehow, there is simply no open-source code written in PL/SQL that I can read to get a good idea of how PL/SQL should be written.” Niall Litchfield recommends the contents of $ORACLE_HOME/rdbms/admin. Any other ideas …

With ndb_restore it is actually possible (talking MySQL Cluster 6.2 and 6.3 here, haven't looked at 5.0 for a long time) to restore a single table into a CSV file that can be loaded using LOAD DATA INFILE.

This can be useful if e.g:

1. You want to extract data from a cluster table and load it into MyISAM/Innodb. This can be useful for reporting/data-mining purposes.
2. You want to load data back into Cluster but into another table looking the same as the original table (not to spoil the original data) for some reason.

Number 2) could be used in scenarios where you want to upgrade the schema by e.g changing the column length of a char(10) to char(255) (not an online operation) and add a bunch of other stuff. Please note that add column is an online operation, changing the column length is not.

Here is an example on how you can do it generate a CSV file using …