Hi. I'm having a problem with my Linode. Sometimes the CPU usage and Disk IO rate goes through the roof - at ~400% and 5,000+ respectively, as seen on my dashboard graphs - and the system hangs. I can't get in through either SSH or LISH, everything is just dead. The only thing I can do is schedule a reboot and wait. This works reliably, though.

This has been going for quite a while now and I have yet to find a solution. In fact, I have no clue what's going on. I've been talking with the support but everything looks fine on their end, so it must be a problem specific to my configuration.

I'm on a 720 running Fedora 8 and as far as I can tell memory isn't a problem. I'm hosting a bunch of web sites for family and friends using Apache/MySQL/PHP (blogs, forums, that kind of stuff) and a few web sites using Tomcat. I've got Postfix, Dovecot, and Mailman running as my e-mail services. There are also a couple of Subversion repositories.

The server isn't normally under much load - in fact, it must spend a lot of time "idling." But sometimes, it goes crazy like I said in the beginning.

So, has anyone had similar problems in the past? Any idea what's going on?

If there's any more information I should post, please let me know.

On an older UML build I saw this problem (the kernel had a memory leak; never did get tracked down).

For 99% of people with this problem the application is using too much memory and so the machine starts to swap. The more it swaps the slower it gets until it looks like it's frozen (it's not, it's just spending all of its time swapping pages in and out of memory).

Search around these forums and you'll find many topics on how to tune Apache and MySQL (a common memory hog) to stop it from doing this.

_________________
Rgds
Stephen
(Linux user since kernel version 0.11)

Oh, that makes sense. It hadn't occured to me that this could be what was happening.


Okay, will do. Thank you very much for your help.