Dear All
Since Graeme raises this issue for ATLAS I am reminded that the GridPP PMB would like to encourage all sites to give some priority to readying themselves for all the LHC experiments for CCRC'08. Resolving problems and having good participation at this stage will greatly improve each site's situation later in the year. The base installs required are listed here: https://twiki.cern.ch/twiki/bin/view/LCG/GSSDCCRCBaseVersions. Quite a lot of information on the SRM situation is being exchanged on the GridPP storage list.
For more information on CCRC take a look here: http://indico.cern.ch/conferenceDisplay.py?confId=24844. The first part of the challenge is due to start on Monday 4th February.
Regards,
Jeremy
-----Original Message-----
From: GRIDPP2: Deployment and support of SRM and local storage management [mailto:[log in to unmask]] On Behalf Of Graeme Stewart
Sent: 31 January 2008 16:20
To: [log in to unmask]
Subject: T2 SRM2 Readiness for ATLAS
Can I poll ATLAS T2s as to their readiness for SRM2.2 in CCRC?
This means updating your SE to the right patch level for CCRC
(https://twiki.cern.ch/twiki/bin/view/LCG/GSSDCCRCBaseVersions) and
defining at least 3TB in the ATLASDATADISK space token, writeable by
atlas.
Glasgow are ready (see http://scotgrid.blogspot.com/2008/01/srm22-
configuration-for-fdrccrc.html), but it would be very good if we got
our other bigger T2s ready too: Lancaster, Manchester, Liverpool,
Oxford, RHUL, RALPP - but of course open to all ATLAS sites...
If you can let me know what your status is, and when you expect to be
ready, I would be grateful - there is some ATLAS specific
configuration which needs to be made to use your new SRM2 configuration.
Thanks
Graeme
On 30 Jan 2008, at 15:38, Greig Alan Cowan wrote:
> Hi Brian,
>
> Thanks for this, it's good information. This matches up with the
> information I posted last week:
>
> - 2 TB for ATLASDATADISK
> - 1 TB for ATLASENDUSER (better with 2 TB)
> - 1 TB for ATLASMCDISK (better with 2 TB)
> - 1 TB for ATLASGRPXXX
>
> Where the XXX is taken from your list below. It looks like we
> should also move the ATLASDATADISK from 2 to 3TB (going by Graeme's
> earlier posting).
>
> Greig
>
> On 30/01/08 15:32, brian davies wrote:
>> As promised, this is a forward of the mail thread regarding Tier2
>> group space tokens for atlas. I would make the suggestion that for
>> FDRs and CCRC we try atr least 2? TOP at GLASGOW and BEAUTY at Lancs?
>> Who are the active Users involved in FDRs and CCRC and what data are
>> they looking at?
>> ( Lancaster is involved with beauty, hence that suggestion and i
>> think
>> graeme suggested TOP for GLASGOW.) may want to try and have a site
>> with multiple GROUPS assigned?
>> Brian
>> ---------- Forwarded message ----------
>> From: Solveig Albrand <[log in to unmask]>
>> Date: 29 Jan 2008 11:20
>> Subject: Re: List of Group names for Space tokens at Tier 2 sites
>> To: Gilbert Poulard <[log in to unmask]>
>> Cc: Dario Barberis <[log in to unmask]>, Brian Davies
>> <[log in to unmask]>, atlas-project-adc-operations
>> <[log in to unmask]>
>> Thanks - I think our list IS up to date.
>> Gilbert Poulard wrote:
>>> Hi,
>>>
>>> The list of existing groups is available from the VOMRS web page:
>>> https://lcg-voms.cern.ch:8443/vo/atlas/vomrs
>>>
>>> Cheers,
>>> Gilbert
>>>
>>> -----Original Message-----
>>> From: Dario Barberis
>>> Sent: Tue 1/29/2008 11:46 AM
>>> To: Brian Davies
>>> Cc: atlas-project-adc-operations
>>> Subject: Re: List of Group names for Space tokens at Tier 2 sites
>>>
>>> Hi,
>>>
>>> I don't know if this list has been explicitly distributed, but
>>> for
>>> sure it must be the same ast the list of VOMS groups the ATLAS
>>> defined. Alessandro De Salvo may help here.
>>>
>>> Initially the list may be shorter for tests, but surely it will
>>> have to be complete by the beginning of data-taking.
>>>
>>> The expectation is that every Tier-1 supports all activities (and
>>> all groups). If this is too much, We'll have to assign groups to
>>> tier-1s and break the symmetry.
>>>
>>> Each Tier-2 site will support only a (small?) subset. The
>>> expectation is that all Tier-2s in the world will collectively
>>> support all groups.
>>>
>>> Cheers,
>>> Dario
>>>
>>>
>>> On 29 Jan 2008, at 11:36, [log in to unmask] wrote:
>>>
>>>
>>>> Does a list for the various space tokens for groups at Tier 2 sites
>>>> exist?
>>>> My assumption to start with we would have:
>>>> TOP
>>>> BPHYS
>>>> SUSY
>>>> HIGGS
>>>> SM
>>>> EXO
>>>> PBPB
>>>> MC
>>>> Is this correct naming convention?
>>>> Is the expectation that every Tier2 in a cloud will support all
>>>> groups?
>>>> That all Tier2s in a cloud will collectively support all groups?
>>>> Thanks
>>>> Brian
>>>>
>>>
>>>
>>>
>>> ______________________________________________________________
>>>
>>> Dr Dario Barberis
>>> CERN-PH Department Tel.: +41.22.767.1302
>>> MailBox E25510 Fax.: +41.22.767.8350
>>> CH-1211 Genève 23 (Switzerland)
>>> ______________________________________________________________
>>>
>>>
>>>
>>>
>> --
>> *************************
>> Solveig Albrand-Paget
>> LPSC
>> 53, Avenue des Martyrs
>> 38026, GRENOBLE cedex,
>> FRANCE
>> *******************
>> [log in to unmask]
>> Tel: (+) (33) (0)4 76 28 41 25
>> ********************
|