Maatkit version 2725 releasedMon, Dec 29, 2008 in Databases Open Source
This release of Maatkit is basically all about mk-log-parser, our new log analysis tool. Daniel and I have been working hard on it all month, adding killer new features like –review, fixing tons of bugs, and making it twice as fast. I believe it’s far and away the best slow-log analysis tool in existence, both in terms of functionality and quality. (Do any of the other tools even have test suites…?) But you should take it for a spin and see yourself. If you run it into something it won’t parse, or find anything else wrong, please use the Google Code project and the mailing list (not the comments on this blog!) for support.
* Added --analyze option to combine analysis and --review (issue 162). * Added --fingerprints (-f) option (issue 160). * Trying to take log of 0 caused a crash (issue 141). * Improved performance of log parsing and fingerprinting about 2x (issue 137). * Slow log parsing was buggy (issue 136 and many smaller issues discovered). * Converted a lot of hardcoded things into dynamically built functions. * Added more information to the default output and reformatted it. * Incompatible changes to fingerprint; old reviews will lose their history. * Much enhanced --review functionality. * Default --top to 95% to analyze the top 95% of the load (issue 171).
That’s just the major stuff. Note that it’s in very active development, and that will probably continue. However, the documentation is now up-to-date and pretty complete.
Big thanks to Pythian, who together with Percona have co-sponsored some of the –review improvements. In fact, the entire –review functionality was born from a combination of Sheeri’s ideas, ideas from Percona consultants, and perhaps a few others who don’t get credit where it is due ;).
About The Author
Baron is the founder and CEO of VividCortex. He is the author of High Performance MySQL and many open-source tools for performance analysis, monitoring, and system administration. Baron contributes to various database communities such as Oracle, PostgreSQL, Redis and MongoDB.