Dear Steve,
> No worries - it's gone away! Must have been restarting it - flushed the cache.
This was documented in the release notes:
http://gridinfo.web.cern.ch/sys-admins/bdii-releases/bdii-5220-1
In a future version of the BDII, I´m deleting the cache files when the BDII is restarted.
By the way, the obsolete bug in GLUE 2 is something that needs to be fixed for the whole BDII hierarchy. I mean, if your resource BDIIs suffer from it, it propagates to the site BDII and then to the top BDII. So you may be running an up to date site or top BDII but if the resource BDIIs still have this bug and for some reason the service is reconfigured, old entries remain and are propagated everywhere else.
By the way, I´m removing the validity test from glue-validator until version bdii 5.2.20 is widely deployed because indeed, it gives many errors due to this bug and since we already know the cause of the problem we can temporarily remove the test.
Glue-validator is still under development trying to include more tests or remove tests due to bugs in the BDII or info providers. You are welcome to try it and give me feedback, but please remember that this is still a beta version. There will be a training about it in the EGI TF.
Cheers,
Maria
> On 07/29/2013 01:09 PM, Stephen Burke wrote:
> > LHC Computer Grid - Rollout [mailto:[log in to unmask]] On
> Behalf Of Daniela Bauer said:
> >> that's an ancient bug, you need to upgrade hepgrid10 to something more
> recent than 1.14.
> > This problem is usually an interaction between that bug and the one fixed by
> this:
> >
> > http://www.eu-emi.eu/releases/emi-3-monte-bianco/updates/-/asset_publi
> > sher/5Na8/content/update-5-03-06-2013-v-3-3-0-1#What_s_new
> >
> > Basically the site BDII was keeping old objects which had stopped being
> published by the underlying service - most likely Liverpool has upgraded the
> CREAM but the fix for that bug triggered the second one. Restarting the site BDII
> is usually enough to clear it, but upgrading is better to stop it happening again.
> >
> > Incidentally, the nagios tests for GLUE 2 are still in development so sites aren't
> yet expected to follow them up, although you can of course look if you're
> interested.
> >
> > Stephen
>
>
> --
> Steve Jones [log in to unmask]
> System Administrator office: 220
> High Energy Physics Division tel (int): 42334
> Oliver Lodge Laboratory tel (ext): +44 (0)151 794 2334
> University of Liverpool http://www.liv.ac.uk/physics/hep/
|