Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Something went wrong with my "setroubleshootd" demon on a production server. it's taking a lot of cpu and won't die no matter what!

what i've tried

sudo killall -9 setroubleshootd
sudo killall -9 <<pid>>
share|improve this question
up vote 3 down vote accepted

The best solution, unfortunately, is to reboot. It's stuck in the kernel.

You can try to unjam it. But this is inherently risky. It may cause the kernel to crash. An orderly reboot is much safer.

Basically, the way you unjam it is to figure out what it's doing (using tools like ls /proc/<pid>/fd, top, and so on) and make something happen to change whatever it's stuck on. If it's stuck on a process, try killing that process. If it's stuck on a file, try removing that file. This very rarely works, but even if it does, I'd consider the system polluted and reboot it anyway.

Messing with it can cause more processes to get stuck "behind" it.

share|improve this answer

Not the answer you want, but If the process won't die, you are probably out of luck - you probably need to reboot. (I note that the command for killing a PID is kill -9 <> rather then killall.

I suspect that there may be a very small subset of events you can get around by using the Mafic_SysRQ key if its enabled (like NFS mounts) - but I've not played with this much and it is for developers - and, I suspect, not usually enabled by default. More info here.

share|improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .