Friday
Nov022007
How WordPress.com Tracks 300 Servers Handling 10 Million Pageviews
Friday, November 2, 2007 at 1:46AM
WordPress.com hosts 300 servers in 5 different data centers. It's always useful to learn how large installations manage all their unruly children: Currently we Nagios for server health monitoring, Munin for graphing various server metrics, and a wiki to keep track of all the server hardware specs, IPs, vendor IDs, etc. All of these tools have suited us well up until now, but there have been some scaling issues. The post covers how these different tools are working for them and the comment section has some interesting discussions too.
Reader Comments (8)
This is old news, not front page worthy in my view, but potentially useful in some searchable archive. - http://www.callum-macdonald.com/" title="Callum" target="_blank">Callum
I come to HS exactly because Todd searches out this type of info and brings it together here. Even if the original post is a few months old, I hadn't seen it before. It costs essentially nothing to link to it from the front page. It costs milliseconds to skip it the post, and only a few moments to click through and read the OP.
Michael T. Nygard
michael@michaelnygard.com
http://www.michaelnygard.com/
Author of "Release It!"
http://pragmaticprogrammer.com/titles/mnee/index.html
That's exactly the spirit of the post Michael. My perspective is always what will help normal people build scalable websites and management is a huge, under appreciated part of that process, which is why I really liked that post. It's very real. And these days it's impossible to link to content nobody else has linked to. So I just made a short intro that would help you decide if you were interested and then let the post stands on its own. I found the link via searching, so I'm not sure if it was part of the normal link parade. I try not to just post those as that wouldn't provide extra value, as Callum said.
what's the difference between nagios and Munin?
Naigos is do it yourself. Munin is plug-and-play.
> Naigos is do it yourself. Munin is plug-and-play.
That's not entirely true. Nagios is good at monitoring application state, event handling, and alerting. Munin is good at graphing lots of server performance metrics. Munin is plug-and-play for generic server resource graphing (CPU, Memory, Disk, etc.). When you write and host your own web application there is always custom work you must do to adequately monitor it. If you found a plug-and-play application that will solve all your monitoring requirements for a custom web app, you probably need to revise your requirements.
Good post. FYI, we here at New Relic just built a free monitoring and management tool that is super easy to set up - it monitors PHP apps (including wordpress)
Check it out...
http://blog.newrelic.com/2010/12/16/measuring-wordpress-performance-with-new-relic-rpm/
check it at http://www.newrelic.com
These are great WordPress resources - I actually just started digging into a really really solid book on WordPress 3.0. It's got some really nice code samples, and is written by a few pro WordPress developers (including some from Envato). I'm actually giving away 2 copies of the e-book on my site - check out the details about the e-book and the giveaway here - I think you'll dig it : http://bit.ly/lq20Ff