500 Error on site after noting 100% CPU usage
I began noticing that my CPU is running at 100% capacity even though I have a simple Wordpress site with minimal traffic. Since yesterday my site is also down and throwing a 500 error. In looking at my latest console logs I'm also seeing "proc/sys/kernel/hung_task_timeout_secs". Can someone help me understand what would cause a spike in usage and resulting 500 error?
1 Reply
Whitney from Linode here! I'll do my best to point you in the right direction.
The hung task timeout you are seeing in your console logs might be evidence of a possible kernel panic, and if so it is possible this is what is causing your CPU to spike.
I would recommend rebooting this Linode to see if it solves the issue. If not, you may have to run a file system check to repair your Linode's disk. To do this, you should boot your Linode into Rescue Mode and then follow along in the Performing a File System Check section of our Rescue and Rebuild Guide.
If this doesn't help your CPU usage, I would suggest taking a look at the following post on our Community Questions site that a member of our team wrote up a while back:
https://www.linode.com/community/questions/17820/why-is-my-linode-suddenly-using-100-of-its-cpu
It has some great tips on tracking down the source of CPU usage, like using the top command to check for running processes. Coincidentally, it also mentions some tricks to look out for a potential compromised if you do think your Linode might have seen an attack. In particular, it mentions taking a look at your auth log, which might give you an idea of where that increased traffic came from. You can usually find that log in its default location:/var/log/auth.log
Hopefully this information helps you track down and troubleshoot these CPU issues!