[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Condor-users] file descriptor safety level exceeded error in schedd log??




We are investigating whether this indicates some kind of file descriptor leak, or whether there is a problem with the heuristic being used here.

You can prevent condor from applying this "file descriptor safety" check by adding the following to your configuration:

NETWORK_MAX_PENDING_CONNECTS = -1

--Dan

John Wheez wrote:

Aborting registration of socket <Startd Contact Socket> to startd <10.0.2.89:1030>: file descriptor safety level exceeded: limit 1600, registered socket count 15, fd 2696

I get the above error and it seems my pool nodes begin to fail and retry the jobs. This is odd because the jobs they are executing do not open nearly that amount of files nor create other files.

These are Windows XP machines.

Is there a way to increase this via Condor or Windows config? Are there any programs out there to help me debug how many file descriptors are being used?

Thanks for any info.

--JW
_______________________________________________
Condor-users mailing list
To unsubscribe, send a message to condor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/condor-users

The archives can be found at either
https://lists.cs.wisc.edu/archive/condor-users/
http://www.opencondor.org/spaces/viewmailarchive.action?key=CONDOR