On Thu, 6 Mar 2003, Lawrence Lowe wrote:
> > Lawrie, you have amber. This means globus jobs can be run. I am not sure
> > whether you can distinguish from your end how Andrew's job arrived - ie
> > direct, via IC, or via CERN. You need to go to the RBs to see what they
> > think.
>
> Yes, it's interesting that Andrew's jobs *all* have gatekeeper log entries
> with "Got connection 194.36.3.182" (which is manchester) and never
> from elsewhere. Whereas, say, my jobs from RAL gppui via the IC RB have:
> Notice: 6: Got connection 155.198.216.19 at Thu Mar 6 14:14:07 2003
> which is gm03.hep.ph.ic.ac.uk, as I guess you would expect.
So the 194.36.3.182 ones are direct Globus job submission (hence the Amber
dot) but it's never worked for me (including me running as the cron
script) via the IC RB for BHAM.
> Maybe Andrew will be able to shed some light on things when he's back.
> But it may be worth knowing that on about 10% of attempts with
> jobs which request a match to BHAM, I get a status from the IC RB
> of
> Status = Aborted
> Last Update Time (UTC) = Thu Mar 6 14:58:11 2003
> Status Reason = No matching resource found
> and if for some reason that happened every time to Andrew's IC jobs, then
> I would never see them. But that's all speculation - there may be
> a much simpler and more stupid reason for my problem!
I've tried doing manual job submissions via the UI at Manchester that the
job monitoring uses and via the gppui.gridpp.rl.ac.uk and I always get
that error. So I believe this must be some MDS/RB interaction problem and
really nothing to do with the map.
Cheers,
Andrew
------------------------------------------------------------------------
[log in to unmask] http://www.hep.man.ac.uk/~mcnab/ +44-161-275-4227
"/C=UK/O=eScience/OU=Manchester/L=HEP/CN=Andrew McNab"
Grid Research, High Energy Physics Group, University of Manchester, UK
------------------------------------------------------------------------
|