Taskhost.exe running cpu at 50%

open the task manager and right click on the taskhost.exe, go to properties and see what this is hooked to, you may need to download Malwarebytes and run a scan, this is not normal or good

I will try this after work. I run Bitdefender and Avira for anti virus, so its not a virus problem. this has had me worried the last few weeks and has had me thinking. I was a computer tech a few years ago but still build PC's, so i know a bit about computers but this has me stumped. I will try your idea tonight and get back to you.

i cant find anything that it is hooked to. i did a print screen but cannot attach the photo.

The photo can be found here or i can email it.


http://www.facebook.com/photo.php?pid=3714827&id=629802...

http://social.answers.microsoft.com/Forums/en-US/w7perf...

Anyway - my cpu was being consumed by the Windows Reliability Analysis Task - the way that I narrowed it down to this was by installing process explorer and letting it show me what was actually running under taskhost.exe

I then went into scheduled tasks and stopped the reliability analysis scheduled activity - and presto the taskhost.exe went away and processor utilization shot down to essentially nothing.

Anyway - that seemed to help me... maybe it will help you



Not sure why it is trying to burn up so much cpu...

Sorry to ressurect a dead thread, but I ran into this today as well, and found that it was related to a service perpetually failing and restarting.

bernimansell's post pointing to the Reliability Analysis task was spot on (it's located under:Task Scheudler > Task Schedule Library > Microsoft > Windows > RAC > RacTask) but instead of just killing it, I decided to look at the triggers.

One of the triggers runs this task each time the Application log sees event ID 1007 from Customer Experience Improvement Program. So I dug into the application log, and found several events per minute related to Intel's Wireless ZeroConfig Service crashing.

Sure enough, Intel(R) PROSet/Wireless ZeroConfig Service's recovery options were set to restart the service on first, second, and subsequent failures. It appears that this service was in a fail/restart/fail loop, and was triggering Reliabilty Analysis faster than it could finish.

I don't use the Intel ZeroConfig tools, so I set the "Subsequent Failures" recovery option to "Take No Action" and killed the taskhost.exe process (could have also just waited for it to work through its backlog, but why bother?), and my problem was resolved.

A more complete solution would have been to troubleshoot the ZeroConfig service, or whatever service was causing the trouble, but since I know I don't need that one, just preventing it from restarting perpetually was good enough for me.

Hope this helps