Re: [Condor-users] Resending: Solaris 10 - All jobs idling for ever...

[Prashant Lal <lalp@xxxxxxxxxxx>]

Hi 


Step 1) do append this at the end of your global config.
Append the below given lines in your condor_config (global file) at the end of the file, do a reconfig, if possible condor_restart -all and then check.

START : Owner == "whateverusername" || Owner == "condor" || Owner == "whateverusername" || Owner == "whateverusername" || Owner == "whateverusername"
START : True
SUSPEND : False
CONTINUE : True
PREEMPT : False
KILL : False

Step2) What your LOCAL_DIR points to (it is for each hosts and it holds the Logs and Execute Dir) 

Does it have all the require directories.

Step3) Please let us know what are the other changes you have made in the global config file.


On Mon, 2005-09-19 at 15:34 -0700, Michael Yoder wrote:
> > 	What do the ShadowLog (on the local machine)
> > 
> > I'm getting this:
> > 
> > bgoncal@lab1a> condor_config_val SHADOW_LOG
> > /home/condor/hosts/lab1a/log/ShadowLog
> > bgoncal@lab1a> more /home/condor/hosts/lab1a/log/ShadowLog
> > /home/condor/hosts/lab1a/log/ShadowLog: No such file or directory
> > bgoncal@lab1a>
>
> Oh!  That changes things.  (More below)
>
> > 9/17 06:55:46 DaemonCore: received command 440 (MATCH_INFO), calling
> > handler (command_match_info)
> > 9/17 06:55:46 vm1: match_info called
> > 9/17 06:55:46 vm1: Received match
> <170.140.151.126:38590>#1126713130#988
> > 9/17 06:55:46 vm1: State change: match notification protocol
> successful
> > 9/17 06:55:46 vm1: Changing state: Unclaimed -> Matched
> > 9/17 06:55:47 DaemonCore: Command received via UDP from host
> > <170.140.151.110:35503>
> > 9/17 06:55:47 DaemonCore: received command 440 (MATCH_INFO), calling
> > handler (command_match_info)
> > 9/17 06:55:47 vm2: match_info called
> > 9/17 06:55:47 vm2: Received match
> <170.140.151.126:38590>#1126713130#989
> > 9/17 06:55:47 vm2: State change: match notification protocol
> successful
> > 9/17 06:55:47 vm2: Changing state: Unclaimed -> Matched
> > 9/17 06:55:52 DaemonCore: Command received via UDP from host
> > <170.140.151.110:35618>
> > 9/17 06:55:52 DaemonCore: received command 443 (RELEASE_CLAIM),
> calling
> > handler (command_release_claim)
> > 9/17 06:55:52 vm1: State change: received RELEASE_CLAIM command
> > 9/17 06:55:52 vm1: Changing state: Matched -> Owner
> > 9/17 06:55:52 vm1: State change: IS_OWNER is false
> > 9/17 06:55:52 vm1: Changing state: Owner -> Unclaimed
>
> This means that the schedd is being matched with the startd, but for
> some reason the startd is getting a command (RELEASE_CLAIM) and letting
> go of that match.  This is happening before the job even thinks of
> starting.
>
> Can you give us more from your SchedLog?  (condor_config_val SCHEDD_LOG)
> (On the submit machine.)
>
> Can you provide us with an IP -> machine mapping for the above?  What
> roles do 170.140.151.126 and 170.140.151.110 play?
>
> > All the daemons mentioned in ... are running:
> > 
> > I'm probably missing something obvious, but I have no idea what it
> might
> > be... :(
>
> It's not obvious yet. :-)
>
> What interesting values have you modified in your condor_config file(s),
> if any?
>
> Mike Yoder
> Principal Member of Technical Staff
> Ask Mike: http://docs.optena.com
> Direct  : +1.408.321.9000
> Fax     : +1.408.321.9030
> Mobile  : +1.408.497.7597
> yoderm@xxxxxxxxxx
>
> Optena Corporation
> 2860 Zanker Road, Suite 201
> San Jose, CA 95134
> http://www.optena.com
>
> _______________________________________________
> Condor-users mailing list
> Condor-users@xxxxxxxxxxx
> https://lists.cs.wisc.edu/mailman/listinfo/condor-users

--
Prashant Lal <lalp@xxxxxxxxxxx>
Cadence Design Systems