I get the same error... did you find a solution to this? I have log level set to WARN and this message appears... 

I think this error is caused when the scheduler tries to run the job publishing versions (don´t know the exact translation atm), that is by default run every minute. If this job not could start due to some reason, the error occurs. Try to check your database for jobs that is set to true, but started a long time ago. Set them to false. This will only make them reevaluate themselves when triggered again. Then, check that you don´t have any scheduled job that takes too much time to complete. Best practice in scheduled job is to run them in a separate thread to avoid those kinds of errors.

Hope you got some help of my post. :)

How do you run a job in a seperate thread? Are you talking about custom jobs or all the standard jobs as well?

I have this error too. The job was running once an hour and takes seconds to complete, yet the job is still running it says.

John, check if there is another job preventing your current job to complete. "Running" doesn´t always mean running :)

Check the table in your Episerver database for more information.

This problem did not go away after applying the hotfix for CMS 6R2. The issue of the jobs "still running" does. Due to it being a bug fixed in the hotfix: http://world.episerver.com/Blogs/Shahid-Nawaz/Dates/2012/1/General-Hotfix-CMS-6-R2/

So it seems that the error in the log may be unrelated to my problem - the question is, what does the error mean then?

I am getting this error in EPiServer 7.1 now. Does anybody know what it is? Can it be safely ignored? It shows as a warning in the log so maybe it's not a huge deal?

Hi!

I´m also experiencing this problem in EPiServer 7.1.2 with Patch4 installed. Does anybody know what´s wrong?

Im also having this problem with CMS 6 r2.. 

I think this problem is related to another error i have, timeout error from TMG "Error Code: 408"

it fits well with the itmestamps in "Terminating client connection: Error reading from pipe"

We are experienceing this too and it is consistently accompanied by a significant performance hit within a particular page load process, so we need to determine the cause.  Did anyone make any progress on figuring this out?

I'm having this problem too with CMS 6 r2 with some customers. It always comes in one minute intervals.

It is not Publishing versions job. If I change the publishing versions job interval to 5 minutes the error still repeats in every minute.

If I stop Scheduler, errors will stop too.

If I look properties of Scheduler service it says "EPiServer CMS 5 Scheduler Service". Maybe this is because of old Schedule version is not updated after EPiServer migration or something.