GriddPP PMB Meeting 822 (13 June 2022)
Present: David Britton (Chair), Davide Costanzo, Jonathan Hays, Roger Jones, Steve Lloyd, Andrew McNab, Sam Skipsey (Minutes), Jill Sambrook (Minutes)
Apologies: Tony Cass, Peter Clarke, David Colling, Alastair Dewhurst, Tony Doyle, Katy Ellis, Peter Gronbech David Kelsey, Andrew Sansum
1) GridPP Vision Statement [DB]
DB sent an updated copy of the vision statement to Jeremy Yates after the group discussed this at the last meeting. Jeremy sent a brief reply saying he was writing a similar thing for UK SRC at the moment. No further comment for now.
Dave also sent this onto a couple of colleagues in STFC, including Nicola Ashman, in order to keep things moving along. We will now wait for a few weeks and then plan for next steps.
Thanks to all for their input.
2) GridPP48 Ambleside [SS]
The event is now live, and Dave has circulated the details and registration link.
Thanks to Sam for organising the registration and Roger for confirming details with the venue.
Dave has also sent details to Mark Wilson and Jeremy Yates. JISC replacement, when announced, will also be invited. Dave asked if anybody else should be included. JH suggested Sarah Verth from STFC. DB highlighted the event structure can sometimes make it difficult for people to pop in and out of as different things are spread across each of the three half days and the venue is not the most accessible.
At the moment, Thursday is set aside for both the network discussion and conf dinner. This is perhaps the best day for Jeremy to attend. Sarah may wish to attend on this day or wait for a more accessible meeting.
RJ is seeking more clarity from the venue about cottages and twinning room’s costs. No update yet, but will provide this soon.
Already 12 or 13 registered attendees.
AD unable to attend today’s PMB meeting, so we will look at a sponsorship update at a later time. DB to remind Alistair about this.
SI-0 Tier-1 Manager’s weekly report & Technical Meetings [AD]
Alistair was unable to attend the meeting, but sent the following Tier 1 Managers report:
I was at CERN last week so there were a few things I didn’t make much progress on.
- CERN meeting
As well as the pre-GDB several members of RAL were out at CERN to meet with the CTA and FTS teams.
We had a useful conversation around operational procedures for data taking.
The EOS developers have given us instructions for providing a core dump when the MGM crashes, which will hopefully allow us to resolve this problem.
The FTS developers have been able to replicate the LHCb slow deletion problem, so they are providing us with a script to do that. If we still can’t replicate the problem at RAL, it would indicate a network problem.
The Webdav REST API for Tape is being back ported into the current version of EOS. The protocol is still evolving and it was thought that it would be good for RAL to lead the adoption of this as we are most in need of this (as the other Tier-1s are happy with dCache’s SRM + http and CERN doesn’t need all the features we need as their tape system is partially hidden from the Grid).
- Technical Meetings
No update. Still trying to organise meetings on Rucio / DIRAC, perfSonar and new OS.
- Tier-1 Liaison Recruitment
No update. The LHCb Liaison and the 50: Liaison : Swift-HEP roles were both accepted by our first placed candidate. They are now going through the visa process, which realistically will take 2 – 3 months. No problems have been reported so far.
- Tier-1 Operations
Things seem to be mostly stable. The load spikes that have been seen on the Webdav servers did not seem to happen last week.
- LHCb Analysis Jobs
I didn’t manage to get round to this last week but I shall start this week. We aim to contact LHCb this week (and I hope to meet Chris Haen at CERN), we will ask for analysis jobs to be directed to a specific CE and ensure that jobs sent to this are schedule to SSD WN with all the fixes in place. We will then encourage LHCb to slowly increase the jobs sent here.
We had a meeting with Dell on Friday 10th to go through the CPU quotes. Our prices are being held until the end of July, (when we expect them to rise). Lead times are variable, but if we placed orders now we would expect delivery in November.
CPU nodes are 14% more expensive, I believe the storage nodes will be closer to a 5% increase.
SI-1 ATLAS Weekly Review and Plans [DCos]
Down time scheduled for this weekend going over to the Monday. Planned so that jobs should start to shut down 48 prior.
VAC migration plan for Birmingham is going ahead and we can now close this action.
CDF down time also planned.
SI-2 CMS Weekly Review and Plans [DC/KE]
SI-3 LHCb Weekly Review and Plans [AM]
The ongoing open ticket re ECHO access is still causing issues for the LHCb team. There has been no substantive reply since around the 16th of May. Researchers have been told to try a series of different protocols; however, each seems to have its own issues.
There is an LHCb NCB meeting later today as part of the 104th LHCb Week 13 – 17 June 2022. Andrew needs to provide an update and it is difficult to answer this question without a resolution to this ongoing issue.
AM to send DB the ticket number and a 1 paragraph summary. DB to raise at Liaison meeting on Wednesday (will send an update prior) and speak with AD to ensure there is some action on this.
Update to follow.
SI-4 Operations Meeting Report [SS,PG,PC] –
SI-5 LCG Management Board Report of Issues [DB] – Environmental Impact of WLCG – Update on
NTR – No meeting last week
SI-6 External Contexts (eg NGI/EGI) [PC/JH]
The STFC e-Infrastructure Advisory Group (SEAG) meeting took place in London last week.
The group looked at previous “actions” and updating these. Significant time spent on re-evaluation of actions from previous strategy.
IRIS applying for £2m DRI money.
REVIEW OF ACTIONS
782.4 DCos – Investigate VAC migration plan for Birmingham. [Ongoing – raised at UK Cloud Support meeting. No formal request as yet.] – follow up
Progress has now been made and action can now be removed.
800.5 AD – Arrange in person DIRAC/Rucio meeting at IC (Jan22). [on-going] – AD plans to re-establish virtual meetings and then set up an in person meeting. AD and DC to try and arrange a date.
818.4 – JS and SS to produce and add the PMB minutes to the CERN website every Monday morning when the Agenda is circulated to the PMB
Sam going to send a link to the group and then we can remove this item from this list.
818.5 – DB/SS Group to formalise VOs to be added to the approved list.
822.1 – AD to provide update about sponsorship for GridPP48 Ambleside.
822.2 – AM to send DB the ticket number for ongoing issues for LHCb group with a 1 paragraph summary. DB to raise at Liaison meeting on Wednesday (will send an update prior) and speak with AD to ensure there is some action on this.
Update to follow.