Fotis Georgatos [mailto:[log in to unmask]] said:
> > https://savannah.cern.ch/bugs/?func=detailitem&item_id=7582
> >
> > Basically you can give a list of brokers (NS) and each NS
> can have a set
> > of LBs specified.
>
> In fact, that looks more like a "fail-over" solution,
> rather than a true "multiple brokers"=="load balancing" solution.
As I understand it the brokers are chosen randomly, which is more than
just failover although it doesn't really balance the load.
> Some of the dev's we now, because they show up in this list,
> but not all.
> It might be that their emails are part of the source/bin code
> or something,
> but I haven't really checked it that well... does anyone know
> an rb developer?
This is the web page for workload management development:
http://egee-jra1-wm.mi.infn.it/egee-jra1-wm/
In theory there is supposed to be (Real Soon Now) a Pre-Production
System deploying the glite release which you can test; the RB is pretty
similar to what we have in LCG, at least from a user point of view. Bug
reports and enhancement requests can be submitted to the JRA1 savannah
area:
https://savannah.cern.ch/bugs/?group=jra1mdw
There is also the EGEE PTF:
https://svn.lal.in2p3.fr/EGEE/PTF/web/index.html
which collects requirements from all areas, including SA1/LCG, and
tracks them in savannah:
https://savannah.cern.ch/support/?group=egeeptf
> Yeap, so the least acceptable common denominator should be
> correct job runs
> from within the dteam VO, as seen from eg. lxplus.cern.ch, so
> that any site
> which is not able even for dteam is automatically pushed to
> maintenance mode.
As I already said, sites can't be pushed to maintenance from the
outside, it has to be done by the local admin, and in many cases once
the admin knows that a problem exists they can fix it anyway. Also you
don't want to stop dteam jobs going to a broken site, otherwise you
couldn't tell when it was fixed.
> Quickly or not, the grid is no longer a party of people in a
> single room,
> so there should be some more formal way of telling the other
> poor admins
> that something is not optimal with that site's maintenance.
There is, the results of the functional tests are public.
Stephen
|