Hi Stephen
The panda monitor is the most up to date "snapshot" of production in
the last 12 hours (it's usually no more than minutes out of date). The
dashboard will never be more than a few hours out of date from this -
it merges results from various databases, but it offers better search
functionality, error aggregation and extensive historical records.
I would say that the dashboard should be a site's first port of call,
but that if you are extra keen you can look at the panda monitor as
well.
finished is the same as success - there are historical reasons why
this is labeled differently in each dashboard. I guess we could change
it, but people are very used to it...
Cheers
Graeme
On Tue, Jan 27, 2009 at 9:59 AM, Stephen Childs <[log in to unmask]> wrote:
> If I look at the panda monitoring
> (http://panda.cern.ch:25880/server/pandamon/query?overview=wnlist&type=production&hours=12&site=csTCDie&reload=yes)
> it says there are 50 jobs "finished" at my site, but on the dashboard
> (http://dashb-atlas-prodsys.cern.ch/dashboard/request.py/overview?site=csTCDie&grouping=cluster&cloud=SARA&start-date=2009-01-26%2020:00:00&end-date=2009-01-27%2008:59:59&grouping=site)
> it only has 24 in the state "success".
>
> Firstly, it's a little confusing having different naming schemes for job
> states. (That is assuming "finished" and "success" are the same.) Secondly,
> why are the numbers different, and which should I trust?
>
> Stephen
> --
> Dr. Stephen Childs,
> Research Fellow, EGEE Project, phone: +353-1-8961797
> Computer Architecture Group, email: Stephen.Childs @ cs.tcd.ie
> Trinity College Dublin, Ireland web: http://www.cs.tcd.ie/Stephen.Childs
>
--
Dr Graeme Stewart http://www.physics.gla.ac.uk/~graeme/
Department of Physics and Astronomy, University of Glasgow, Scotland
DEATH TO MEETINGS!
|