Hi Steve,
thank you for the information and it helped. I checked the file and that
was alright. There was however one essential difference. Your script
pointed to
/opt/lcg/libexec/lcg-info-dynamic-pbs (very much at the beginning of the
script)
and ours
to
/opt/lcg/libexec/lcg-info-dynamic-ce
I checked that file and it had been corrupted by me re-doing the
configuration and read like
#!/bin/sh
/opt/lcg/libexec/lcg-info-dynamic-pbs /opt/lcg/var/gip/lcg-info-generic.conf
#!/bin/sh
/opt/lcg/libexec/lcg-info-dynamic-pbs /opt/lcg/var/gip/lcg-info-generic.conf
#!/bin/sh
/opt/lcg/libexec/lcg-info-dynamic-pbs /opt/lcg/var/gip/lcg-info-generic.conf
That means that the re-configuration did not overwrite the file but rather
appended.
I removed the last for lines, leaving only
#!/bin/sh
/opt/lcg/libexec/lcg-info-dynamic-pbs /opt/lcg/var/gip/lcg-info-generic.conf
and we have ok on all fronts.
I would think that this must be a bug in the configuration procedure.
Lydia
On Sat, 16 Apr 2005, Steve Traylen wrote:
> On Sat, Apr 16, 2005 at 06:17:34PM +0100 or thereabouts, Lydia Heck wrote:
> > Hi Maarten,
> > Thank you for the explanation. But I must admit that I do not follow
> > it completely.
> >
> > If we can run the ldapsearch -x -H and we get full information internally
> > and externally why do we fail on the test?
>
> Hi Lydia,
>
> I guess you are talking about these tests
>
> http://goc.grid.sinica.edu.tw/gstat/Durham/
>
> from these your site bdii can be reached on 2170 so that
> is healthy as I see it.
>
> However there are bits of information missing that come from the
> local GRIS on CE:2135.
>
> In particular the GlueCEUniqueID instances.
>
> I've attached a working /opt/lcg/var/gip/lcg-info-generic.conf
> (path not confirmed)
>
> You will have a some missing sections, the question is why?
>
> Steve
>
> >
> > With respect to the ldapsearch -LLL : should we run that
> > locally only? without referring to our CE at all? I got the from
> > some document or other.
> >
> > Lydia
> >
> >
> > On Sat, 16 Apr 2005, Maarten Litmaath, CERN wrote:
> >
> > > On Sat, 16 Apr 2005, Lydia Heck wrote:
> > >
> > > > [...]
> > > >
> > > > In order to check we did run locally
> > > > ldapsearch -x -H ldap://helmsley.dur.scotgrid.ac.uk:2170 -b
> > > > Mds-vo-name=Durham,o=Grid
> > > >
> > > > which gives full information that on the face of it looks correct.
> > > > From our UI we have submitted jobs, and they work.
> > > > We logged into Mark Nelson's account in RAL and did the ldapsearch
> > > > from there to make sure that there were no firewall issues and that
> > > > gave the results previously seen.
> > > >
> > > > The output of
> > > > ldapsearch -LLL -x -H ldap://helmsley.dur.scotgrid.ac.uk:2135 -b
> > > > "mds-vo-name=Durham,o=grid"
> > > > No such object (32)
> > >
> > > On port 2135 runs the GRIS, which has "mds-vo-name=local,o=grid".
> > > On port 2170 runs the GIIS, which has "mds-vo-name=Durham,o=grid".
> > >
> > > The GRIS provides info about a single resource (CE|SE|RB|PX|...),
> > > whereas the GIIS provides information about all resources on a site.
> > >
> > > The GIIS must be externally visible, the GRIS need not (in LCG-2).
> > >
> >
> > --------------------------------------------------------
> > Dr E L Heck
> > Department of Physics | Tel.: + 44 191 - 334 3628
> > University of Durham | Fax.: + 44 191 - 334 3645
> > Durham, DH1 3LE |
> > U.K. | e-mail: [log in to unmask]
> > --------------------------------------------------------
>
> --
> Steve Traylen
> [log in to unmask]
> http://www.gridpp.ac.uk/
>
--------------------------------------------------------
Dr E L Heck
Department of Physics | Tel.: + 44 191 - 334 3628
University of Durham | Fax.: + 44 191 - 334 3645
Durham, DH1 3LE |
U.K. | e-mail: [log in to unmask]
--------------------------------------------------------
|