--------------------------------
Present: A. Forti, P. Love, P.Hodgson, M. George, A. Beresford, R. Allan
Situation at sites
------------------
Lancaster:
New kit (conona family processors) has arrived it hasn't been
installed yet. Peter is working on
the installation framework (kickstart) and there are problems with the
electric power supply to the room where the kit is going.
Paul has offered to go and help with the YAIM scripts if needed.
There is some interest in dcache but they will wait for me to say what
to do and keep on with classic SE until then. Post readvertised and contacted a PhD for interview but no interview yet.
Liverpool: Michael installed one node with LCFG server still redhat 7.3. If the other nodes can be restored with the same method quickly they will otherwise we will pass to YAIM in the middle term and Alessandra will work on it. Post readvertised and working on it but no news yet.
Manchester:
Tender for the farm is out, given a talk to the vendors last friday to
explain what type of kit is wanted.
Posts: Mark Kelly will start in March and Colin Morey will start in june
Sheffield: Farm up and running, filled with atlas software jobs after 10 days ago installation. There were problems with a file (permissions). Post Andrew has started ~3 weeks ago.
Daresbury: Robert said he'd like to work more in contact with us he cannot offer equipment but certainly he can offer knowledge especially involving e-science experts working on common areas like networking monitoring.
GrdiPP testbed
------------------
Alessandra reported the decision of having a Gridpp testbed to train new staff and work on pre-production environment. Agreed that northgrid has an abundance of old machines that can serve the pourpose once the new kits are in place.
Service Challenges
------------------
Lancaster volunteered to participate in the summer service challenges. It is unclear if this will happen in the NorthGrid context or if it will happen under ATLAS experiment umbrella using nevertheless NorthGrid equipment. Ask Roger Jones what does he intend to do since Peter doesn't know anything about it.
For the end of the year service challenges everybody will be involved. This requires planning and testing.
Nobody knows anything about their networking or uk-Light (which is the alternative).
Nobody has a proper storage strategy. dcache is considered appealing but everybody is reluctant to give it a go until it is demonstrated that it works.
Security
--------
* Sites are required to upgrade to 2.3.1 or upgrade the security rpms.
* Sites are asked to enable a cron job to update automatically the CA
rpms as suggested on the lcg-rollout thread New CA RPMS
* Asked for a security workshop to learn more about
security. Recognised that main problem is not yet the grid but
hardining of the system using unix tools. Share
knowledge and scripts.
* Security pages written with that pourpose please look at them.
Actions
-------
Alessandra:
* keep on working on dcache
* find out more about service challenges
* help Liverpool if needed
Andrew:
* find out about dcache
* find out about network topology at different universities using
network contacts
* send a link for the ukerna network workshop that is going to be in
Manchester
Peter:
* setup Lancaster new kit
Paul:
* help Lancaster with YAIM scripts if needed
Michael:
* Bring Liverpool kit back (on best effort basis)
Rob:
* Put us in contact with Daresbury networking people to setup
networking monitoring boxes at all sites.
Next Meetings: 1 April 2005 on the phone.
Last modified Mon 7 March 2005 . View page history
Switch to HTTPS . Website Help . Print View . Built with GridSite 1.4.3