- Compact style
- Indico style
- Indico style - inline minutes
- Indico style - numbered
- Indico style - numbered + minutes
- Indico Weeks View
EARLY DRAFT NOTES (Jeremy)
Alessandra; Chris B; Dan T; Daniela B; David C; Elena K; Ewan MM; Federico M; Gang Q; Gareth R; Gareth S; Ian L; Ian N; Jeremy C; John B; John H; Marcus E; Matt D; Kashif M; Oliver S; Winnie L; Pete G; Robert F; Robert F; Steve J; Sam S; Terry F.
018: H/w for main compute platform. Other in LMC.
CMS (Daniela): Nothing to report. Glasgow on blacklist as T3. For user job submission. Nobody knows what happened. Issues for T2s so may be CMS staff change issue.
ATLAS (Elena): Going through the tickets – discussed at ATLAS meeting last week. Quite a number asking for DPM dumps.
118732 – Glasgow. Transfers fail with no such file dir error. Sam confirmed the files do not exist. Now waiting for reply.
118052 – http deployment. From December 2015. Sam thinks the problem is solved. Some disk servers running old DPM dlite. Looking to fix. Link shows 60% tests are successful.
117889 – ATLAS request storage consistency check. Sam was asking which endpoints to have checked. Next on the list for Sam but have not been a high priority until now.
118695 – Lancaster. Storage consistency. Matt thinks he has solved the problem.
117894 – Sussex. Storage consistency. Not in progress yet.
118740 – Brunel. Multicore jobs failing. Raul thinks they are failing with errors of lost heartbeat and difficult to trace. May be related to power cuts.
117878 – Brunel. Consistency check. Raul says he is not sure what he should do. ATLAS response.
118756 – Manchester. Open last night.
118679 – Manchester. Http support.
117885 – Manchester storage. Two servers being deployed.
118728 – QMUL. T0 expert file cannot be copied to site. Dan investigating.
4 tickets on ATLAS request storage consistency: Oxford, Birmingham, Sheffield and RHUL. Each requires attention although Govind noted he will deal with it shortly.
ATLAS jamboree at the end of January. https://indico.cern.ch/event/440821/
LHCb (Raja): Not much running on grid. Little MC. New request to come in.
T2A: Cloud resources open nebula at T1.
LIGO (Sam) some discussion about a problem (re Condor). Paul has just started looking at it again.
LSST: VOMS servers showing up properly in VO ID card.
Marcus may update on test transfers from NERSC as now has an account. He can try to do some test transfers. Need to know from Joe what data to transfer – the location.
LZ: (Elena). Ran production jobs for technical design report to simulate background events. Ran at IC. Daniela was asking sites to support LZ. Checking in BDII only see RALPP and Sheffield.
Oxford – turn on support on the CPU. TBD.
ECDF – Still plan to support it. But currently switching compute cluster. Beginning Feb.
Liverpool – Accounts enabled.
Daniela: If you enable LZ you need to have pilot accounts enabled. RALPPD works.
Sheffield and Sussex do not have pilot roles for GridPP VO. Problem at weekend with VOMS server down at weekend. Admin interface at Manchester. Bug in Dirac triggered. Set up extra DNS servers but the way setup not correct.
RF: Issue – container that ran … log file 4000/lines second. Route cause not identified. DNS issues Andrew was dealing with this and there needs to be some updates higher up the chain. JANET/IT services. Q: Has Andrew opened the ticket?
Dirac dependency on VOMS? Was a bug in Dirac setup.
RAL – being followed up.
Brunel. No update
UCL sorted? VAC standard images have it.
Sheffield – enabled pilot role for several VOs. Still does not work. Maybe an ARGUS config issue.
Durham. Couple to do. IN =-hand.
ECDF. Nothing yet. Switched.
Glasgow. Was on-hold while looking at account management. Looking to centralize.
Birmingham?
RALPPD – done.
GDB – David’s talk may be in March.
Request to sites. Use of puppet repo.
Please register for HEPSYSMAN
Ask Mark to forward the requirements for the GANGA meeting