Dave, gstat looks at the information system - I think it actually shows
jobslots-freeslots so it sees all the jobs submitted by other means -
such as Rodney's ATLAS jobs, that rtm doesn't see. This can be a large
number.
John
> -----Original Message-----
> From: Testbed Support for GridPP member institutes
> [mailto:[log in to unmask]] On Behalf Of Dr D J Colling
> Sent: 13 March 2006 20:20
> To: [log in to unmask]
> Subject: Re: GSTAT vs real time monitor
>
> Hi,
>
> I don't think that this is the reason. I am not sure how the
> gstat gets its figures but we certain have all the major RB
> (with the exception of NIKHEF). I believe the problem to be
> related to us having a power cut across the whole campus on
> Saturday and things not coming back properly.
> I have mailed Gidon about this, he is on holiday but I am
> hoping that he comes into contact with a computer and fix it.
> Otherwise I will ask somebody else to take a look.
>
> All the best,
> david
>
> On Mon, 13 Mar 2006, Burke, S (Stephen) wrote:
>
> > Testbed Support for GridPP member institutes
> >> [mailto:[log in to unmask]] On Behalf Of Michael
> George said:
> >> Does anybody know where there would be a mis-match between the
> >> number of jobs running as reported by Gstat and the number
> reported
> >> by the real time monitor?
> >
> > I think they get the information from different places,
> gstat from the
> > information system and the rtm from (a subset of) the
> resource brokers.
> > On the whole gstat is probably more likely to be accurate, unless
> > there are problems with the info providers, since it should
> see every
> > job known to the batch system.
> >
> > Stephen
> >
> >
>
|