Lighthouse Database issues

rick

Posted by rick at February 17th, 2009

I woke up this morning to some crazy database issues on Lighthouse. It’s hard to tell what’s going on after the fact, it already seemed like it was starting to settle down. I did some looking around, and noticed that there some projects that were deleted. Active projects tend to have a lot of dependent objects that need to be cleaned up, such as Events, Tickets, Messages, Comments. I just pushed a change that will offload this to a background job.

database response time chart

This particular issue was a bit difficult to track down, because the end result was that everything was slower. New Relic just showed the usual suspects at the top of the most active queries: Ticket.find, Spam.find, etc.

top database consumers

Things seem like they’re doing much better now, but I’ll be keeping a closer eye on thing and double checking that we have the proper database indexes in place.

1 Comment

  1. Derek Haynes Derek Haynes said on February 19th, 2009

    We’ve run into this issue before – one MySQL command we used to debug is SHOW PROCESSLIST. It returns a list of all running processes along with the elapsed time for each.

    Also useful because DB issues could be occuring outside the request cycle (ie – background jobs).

Make your voice heard

Sorry, but comments are closed for this item.