Re: [HTCondor-users] Sock::bind failed ShadowLog

[Keith Brown <keith6014@xxxxxxxxx>]

I had this book marked several years ago,Âhttp://www.cs.wisc.edu/condor/condorg/linux_scalability.html

is there a new updated Linux scalability page?Â

On Fri, May 2, 2014 at 11:02 PM, Keith Brown <keith6014@xxxxxxxxx> wrote:

...bump...

On Fri, May 2, 2014 at 6:47 AM, Keith Brown <keith6014@xxxxxxxxx> wrote:

i don't believe this is a permission issue because the job actually executes fine and runs for hours and then suddenly I see "CEDAR:6001:Failed to connect to ..."Â

i am at a loss here.Â

On Wed, Apr 30, 2014 at 6:48 AM, Keith Brown <keith6014@xxxxxxxxx> wrote:

has anyone seen this?

On Tue, Apr 29, 2014 at 7:02 AM, Keith Brown <keith6014@xxxxxxxxx> wrote:

On a large pool I noticed several of my job are keep getting rescheduled.Â

By looking at the ShadowLog I noticed,Â

Sock::bind failed: errno = 98 Address already in use

RemoteResource::killStarter(): Could not send command to startd

Sock::bind failed: errno = 98 Address already in use

Can't connect to queue manager: CEDAR:6001:Failed to connect to <scheduler>

Failed to perform final update to job queue!

By checking the mailing list, I have these settings:

NO_DNS = false

Anything I should be looking at?Â