Thanks. Can you ask at least the Tier 1's for their views on this and
summarise?
Ian
> -----Original Message-----
> From: LHC Computer Grid - Rollout
> [mailto:[log in to unmask]] On Behalf Of Jeff Templon
> Sent: 07 September 2005 11:04
> To: [log in to unmask]
> Subject: Re: [LCG-ROLLOUT] LCG-2_7_0 sooner or later ....
>
> yo
>
> i will give one of the site talks.
>
> Ian Bird wrote:
> > I suggest we have a time in the operations workshop to go
> through the
> > rationale behind the VO boxes, and all the issues
> associated with them.
> > Could I propose that someone representing the sites views
> (Jeff, John??)
> > could prepare a summary of fabric/site concerns and I will
> present the
> > view from the basline services. It would be nice if an experiment
> > person could summarise their side too. Theses boxes should not be a
> > free-for-all - there should be some constraints, and we need to
> > understand what those should be.
> >
> > Ian
> >
> >
> >>-----Original Message-----
> >>From: LHC Computer Grid - Rollout
> >>[mailto:[log in to unmask]] On Behalf Of Jeff Templon
> >>Sent: 06 September 2005 21:52
> >>To: [log in to unmask]
> >>Subject: Re: [LCG-ROLLOUT] LCG-2_7_0 sooner or later ....
> >>
> >>My take:
> >>
> >>I was at the 'hotly debated' meeting and indeed there was lots of
> >>resistance; I would have liked to have said no but I saw no other
> >>alternative, both from the HEP side (where is the generic
> >>software that
> >>will do what their VO software is going to do??) and from
> the Tier-1
> >>side (will anyone send us jobs if we don't install a VO box?)
> >>
> >>On the other hand, there are clear no-nos and these have
> not changed.
> >>We are providing the boxes as a service to the VO in the hope that
> >>something better will appear sometime. After all it looks
> >>like the SGM
> >>installation area may finally go away, and Tank and Spark may
> >>be dead,
> >>so there is reason to have at least some hope of poor solutions not
> >>becoming permanent. Service here should be understood as
> hospitality
> >>... we want to help. We all want this to work. However if
> >>you spit on
> >>your host's carpet he may ask you to leave. The solution for
> >>us to to
> >>help the VOs now, also help push for generic solutions for the long
> >>term, and to make it very clear that if the VO box starts to
> >>interfere
> >>with or degrade the quality of our generic Tier-1 services,
> >>it will be
> >>turned off. I think that message has been receieved by the
> >>experiments,
> >>I don't see problems.
> >>
> >> J 'my god what an optimist' T
> >>
> >>Gordon, JC (John) wrote:
> >>
> >>>Markus, it would be good to hear from Jamie and Ian on the
> >>
> >>decision. The
> >>
> >>>agreement of Tier1s has been with distinct reservations as we still
> >>>don't know what the VOs will do with these boxes. Alice
> >>
> >>have given some
> >>
> >>>information - lists of software but not details of how they
> >>
> >>will be used
> >>
> >>>- and even if we are happy with what runs at first, there seems no
> >>>control over what happens subsequently. So not to be
> >>
> >>obstructive we have
> >>
> >>>agreed but are keeping a close watch and will take action if we see
> >>>problems.
> >>>
> >>>RAL have set up a VObox for Alice although we still have a firewall
> >>>problem. The other UK sites will wait and see how we get
> on. For the
> >>>moment we are happy to talk about HOW but I am sure we
> >>
> >>won't be the only
> >>
> >>>site who considers WHETHER if VOs are felt to be abusing
> >>
> >>the privilege.
> >>
> >>>I am not sure I agree with you on the VO box being the more
> >>
> >>secure way
> >>
> >>>of working. By having the VO box installed you have
> >>
> >>persuaded sites to
> >>
> >>>put aside their security concerns. If the VOs had to
> >>
> >>justify what they
> >>
> >>>were doing sites would feel happier that they weren't
> >>
> >>installing some
> >>
> >>>trojan that will cause problems with their site security people.
> >>>
> >>>I don't mean that VOs should negotiate with each site but I
> >>
> >>would like
> >>
> >>>to think that someone somewhere knew what was being
> >>
> >>installed and had at
> >>
> >>>least thought about possible problems and could reassure
> >>
> >>the rest of us.
> >>
> >>>I don't really mind who this is. It could be you or some
> >>
> >>other person
> >>
> >>>who we all respect and who is independent of the experiments.
> >>>
> >>>John
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>>-----Original Message-----
> >>>>From: LHC Computer Grid - Rollout
> >>>>[mailto:[log in to unmask]] On Behalf Of
> Markus Schulz
> >>>>Sent: 06 September 2005 12:33
> >>>>To: [log in to unmask]
> >>>>Subject: Re: [LCG-ROLLOUT] LCG-2_7_0 sooner or later ....
> >>>>
> >>>>Hi Alessandra,
> >>>>yes, there is additional documentation made available by Simone
> >>>>Campana in wiki. http://goc.grid.sinica.edu.tw/gocwiki/VOBOX_HowTo
> >>>>In addition the release note contained quite some material.
> >>>>
> >>>>We can discuss the box there, however for the Tier-1 centers
> >>>>the fact
> >>>>that there are VO-BOXES has been already decided and
> agreed by the
> >>>>site managers of the T1s. Jamie and Ian can give you more
> >>>>details on
> >>>>the state of this decision.
> >>>>
> >>>>If we discuss this at the workshop we must ensure that we
> >>
> >>don't try
> >>
> >>>>to revert this decision. The discussion has to be focused on the
> >>>>"how" and not on the "wether".
> >>>>
> >>>>
> >>>> markus
> >>
>
|