Re: [Condor-users] Connection problem

["Jason Stowe" <jstowe@xxxxxxxxxxxxxxxxxx>]

Matthieu,
How are the machines configured? As I understand it you have:
Linux = Running the condor_schedd
VMWare image of Windows = Running the condor_startd for executing jobs.
Perhaps check that the central manager is configured/running for this
pool. Can you share your configuration files for the two machines? My
guess here is that the negotiator and collector aren't running or
available for some reason, which would explain why you're getting:
> 11/24 12:26:39 (pid:8775) DaemonCore: received command 421 (RESCHEDULE),
> calling handler (reschedule_negotiator)
> 11/24 12:26:39 (pid:8775) attempt to connect to <xx.xx.xx.xx:9618>
> failed: Invalid argument (connect errno = 22).  Will keep trying for 20
> total seconds (20 to go).
The reschedule command happens after you submit to a scheduler, and
the schedd attempts to connect to the negotiator to tell it that jobs
are available to be run. Getting the configuration files would help,
plus the results of running the following command from the linux box
running the scheduler:
condor_config_val -n <Hostname of central manager> DAEMON_LIST
I'm happy to help further if you need it.

Warm regards,
Jason

--

===================================
Jason A. Stowe

Phone: 607.227.9686
jstowe@xxxxxxxxxxxxxxxxxx

Cycle Computing, LLC
http://www.cyclecomputing.com


On 11/24/06, Cargnelli, Matthieu <Matthieu.Cargnelli@xxxxxxxx> wrote:
> Hi,
>
> I'm trying to get a Windows condor master/worker working with a linux
> box (just submitting jobs). The windows box is actually running inside
> VMWare.
>
> The linux box is running a schedd, and I'm able to run a condor_submit
> command which puts the job in the local queue. Only the job doesn't
> start. A condor_status fails to, indicating that the collector doesn't
> respond. Here is a par of the Schedd Log :
>
>
> 11/24 12:25:01 (pid:8775) attempt to connect to <xx.xx.xx.xx:9618>
> failed: Invalid argument (connect errno = 22).  Will keep trying for 20
> total seconds (20 to go).
>
> 11/24 12:25:21 (pid:8775) attempt to connect to <xx.xx.xx.xx:9618>
> failed: Invalid argument (connect errno = 22).
> 11/24 12:25:21 (pid:8775) ERROR: SECMAN:2003:TCP connection to
> <xx.xx.xx.xx:9618> failed
>
> 11/24 12:25:21 (pid:8775) Failed to start non-blocking update to
> <xx.xx.xx.xx:9618>.
> 11/24 12:26:00 (pid:8775) get_file: Zero-length file check failed!
> 11/24 12:26:00 (pid:8775) Failed to receive file from client in
> SendSpoolFile.
> 11/24 12:26:39 (pid:8775) DaemonCore: Command received via UDP from host
> <127.0.1.1:32788>
> 11/24 12:26:39 (pid:8775) DaemonCore: received command 421 (RESCHEDULE),
> calling handler (reschedule_negotiator)
> 11/24 12:26:39 (pid:8775) attempt to connect to <xx.xx.xx.xx:9618>
> failed: Invalid argument (connect errno = 22).  Will keep trying for 20
> total seconds (20 to go).
>
> The linux box can ping the windows box (with its full hostname) and the
> windows firewall is opened for port 9618, I can connect through this
> port via telnet.
> I don't know what is wrong. Does someone have a clue on this ?
>
> --
> Matthieu Cargnelli
> EADS CCR - Centre de Toulouse
> Centreda 1
> 4, Avenue Didier Daurat
> 31700 BLAGNAC
> Tel: (+33 5) 67.19.61.73
>
> _______________________________________________
> 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


--

===================================
Jason A. Stowe

Phone: 607.227.9686
jstowe@xxxxxxxxxxxxxxxxxx

Cycle Computing, LLC
http://www.cyclecomputing.com