Re: [condor-users] Condor Problem 6.6.1

[Alain Roy <roy@xxxxxxxxxxx>]

> Should i be worrying about these crashes. Of late they have become more
> frequent

Gaurang--

I thought you were just using the scheduler and globus universe without 
matchmaking. Am I correct? If so, you don't need the startd to run, and you 
can disable it. Though the log below does look like you are running jobs in 
another universe.

If I'm incorrect, then the real question is: is this getting in the way of 
your work? Are jobs not completing? If so, then you should worry about it. 
Otherwise, you shouldn't.

We Condor developers should certainly worry about it though!

I would like to see a bit more information about this problem, if possible. 
It's probably not worth talking about it on this mailing list (to save 
people the back and forth of debugging information).

> *** Last 20 line(s) of file StartLog:
> 4/2 16:16:09 ERROR:
> SECMAN:2003:TCP connection to <128.9.72.82:42576> failed
> 4/2 16:16:09 Send_Signal: ERROR Connect to <128.9.72.82:42576> failed.4/2 
> 16:16:09 Error sending signal to starter, errno = 25 (Inappropriate ioctl 
> for device)
> 4/2 16:16:09 State change: Error sending signals to starter

What does the StarterLog say at the same time this happened? Does it have 
any error messages?

-alain


Condor Support Information:
http://www.cs.wisc.edu/condor/condor-support/
To Unsubscribe, send mail to majordomo@xxxxxxxxxxx with
unsubscribe condor-users <your_email_address>