I’ve seen a number of solutions for finding long running transactions or sessions within InnoDB / MySQL now. Every single one of them has (in the past by necessity) been implemented as a script (other than one, more on that one later) that is either invoked manually, or via some cron job, that then connects and tries to find, and possibly diagnose or kill, transactions that break some “long …[Read more]
So it’s been a while since I’ve hit my blog, but I feel compelled to respond to Baron’s post, and many of the other (perhaps short sighted) criticisms out there against this new functionality that we’ve been working on for so long.
Everybody seems to be saying that this functionality should be implemented in the server, or that the better way to do this is to use these patches which add functionality to the logging that MySQL already provides. Well guess what people - what does that give you, other than some more details on you queries?
What’s bad on a database server?
Query Analyzer, whilst it does currently use a proxy to collect the statistics, doesn’t hit your disk at all. Everything is collected and aggregated in memory, it …[Read more]