Hi Rod, Fokke,
Fokke please confirm, but AFAIK the Atlas data is on a storage element
(teras.sara.nl) that is NOT affected by the overall downtime of the compute
site (matrix.sara.nl).
We must make sure though that the publishing of teras is continued
(like for the RLS).
Fokke, can you confirm?
DavidG.
Rod Walker wrote:
> Sara is among the Tier 1 sites storing data critical to the ongoing Atlas
> production. I`m sure it`s impossible to schedule downtime to suit all VO's
> but might it be possible to devise an upgrade strategy which keeps the
> services running. I think that this could work for sites with sufficient
> flexability in terms of resources, i.e. the critical tier 1's.
>
> Cheers,
> Rod.
>
>
> On Mon, 18 Apr 2005, Fokke Dijkstra wrote:
>
>
>>Hi,
>>
>>Next week we will be migrating our site (SARA-LCG2) to LCG-2.4.0. At the same time we will also switch the OS to Scientific Linux. Together this forms a rather drastic change.
>>
>>We are also running an RLS server on our site, which is used by a number of VOs. Since the RLS is not supported within LCG-2.4.0 we will leave it as it is. And upgrade to an LFC in the near future. This has, of course, to be discussed with the VOs first.
>>
>>In the meantime our site will be down for about a week, which is unacceptable for the RLS. I have therefore installed a BDII on the RLS as well that contains its information.
>>
>>My suggestion would be to have the RLS as a separate entry in the all-sites list, removing the dependency of the existence of the RLS endpoint information on the availability of the Computing Element.
>>Is this a good idea?
>>
>>If yes, can someone add the following entry to the all-sites list:
>>SARA-RLS ldap://rls.matrix.sara.nl:2170/mds-vo-name=sararls,o=grid
>>
>>Please notify me, because I will then have to remove the RLS entries from the BDII on our CE.
>>
>>Kind regards,
>>
>>Fokke Dijkstra
>>
>>
>>--------
>>Fokke Dijkstra
>>High Performance Computing
>>SARA - Reken- en Netwerkdiensten http://www.sara.nl
>>Tel. +31 20 592 8004 Fax. +31 20 668 3167
>>
>>
>
>
> --
> Rod Walker +1 6042913051
--
David Groep
** National Institute for Nuclear and High Energy Physics, PDP/Grid group **
** Room: H1.56 Phone: +31 20 5922179, PObox 41882, NL-1009DB Amsterdam NL **
|