Swiss Grid Operations Meeting on 2021-01-21 at 14:00

Next meeting: 11 Feb 2020 at 15h00 (note the unusual time)


Minutes:

1- Waiting for final confirmation about running 2 ARCs to the same partition:

  • ATLAS OK (Gianfranco)
  • CMS: OK (Derek confirmed after the meeting)
  • LHCb: OK (Roland confirmed during the meeting)
  • In case of further details the discussion will be continued on SLACK
New configuration expected to be deployed on Feb 10.

2- Logs sharing: Everybody agrees to share the logs

3- Mauro has hard time reconciling the plots provided by ATLAS (Gianfranco) and CSCS (Nick charts).

Nick asked again for the definition of a time interval to define the denominator to be used to produce the charts pledged/delivered core hours. Mauro (and people present agreed) to use a month, such that we are in sync with the OPS meeting.

Comment by Gianfranco: yes, all ATLAS monthly reports have always been based on calendar month

Mauro: I understand it's not only a matter of integral delivered resources, but also how they are delivered ("flat efficicieny" instead of waves).

On the other hand all present agreed that if CSCS is unable to deliver resources (e.g. security issues) they will try to catch up by temporarily adding some nodes (as they have already done in the past), while if experiments are not sending jobs, those cannot be recovered ("use it or lose it" as if it was on a standard cluster).

Mauro: while the ATLAS message is clear, I don't understand the meaning of the horizontal line "CSCS Pledge 64K" in Gianfranco's slides. It sits at ~105k ?

Comment by Gianfranco: as discussed in the past, the red line at ~105k on the plot is the combined "federation pledge". As requested in the past, I manually add the text "CSCS Pledge 64K" to quote the each site level of pledge

4- CSCS Nick report

100% availability from WLGC monitoring

slide 3: around 10-14 December ATLAS had no pending jobs. Gianfranco can you check what happened ?

Comment by Gianfranco: No. At the time, no anomalies from ATLAS side. Could not cross-check since CHIPP dashboard was not available (see unresolved slack discussion from 9th to 14th Dec)

Activities in progress:

  • Access to dCache data via Macaroons -->(Edited by Derek) Elia adapted the configuration to allow certificate-less access to the DAV-door. He needed an answer from the CMS central Ops whether they were able to now do macaroon authenticated/authorized transfers. They could, so this is solved.
  • ATLAS DPM Migration: What is the plan and timeline to move from test phase to production? (Gianfranco, sorry couldn't find it in your last report) Comment by Gianfranco: https://wiki.chipp.ch/twiki/pub/LCGTier2/MeetingSwissGridOperations20201210/UnibeT2ReportNov2020.pdf

5- CMS:

nothing to report

6- LHCb:

nothing to report (problem transferring files / open ticket should be OK)



Followup from previous Action Items

Action items

To be xchecked: https://wlcg-docs.web.cern.ch/reporting/accounting/Tier-2/2020/WLCG-T2accounting_12-20.pdf

VO reports

ATLAS

  • ATLAST2ReportDec2020.pdf: ATLAS CH Tier2 report

  • Highlights:
      • 4th month in a raw of underdeliver: 41% of pledge in December
      • 2620 avg running slots (at least 6k excpected)

CMS

LHCb

T2 Sites reports

CSCS

UniBe

T3 Sites reports

PSI

UniGe

EGI / WLCG

  • A/R report for December for NGI_CH OK: https://argo.egi.eu/egi/report-ar/Critical/NGI?month=2020-12
  • ARC CE disclosure of job owner names [EGI-SVG-2020-17013] does not affect LHC experiments
    • All jobs submitted with robot certificates
    • But if the site accepts jobs outside from the experiment frameworks, then a patch is needed
  • Retirement of Site-BDII checks from the availability profile is being discussed (deprecated service)

Review of open tickets

9 of 9 Tickets
Ticket-ID Type VO Site Priority Resp. Unit Status Last Update Subject Scope
150267 lhcb CSCS-LCG2 urgent NGI_CH waiting for reply 2021-01-21 Enable xroot and https with third-party ... WLCG
150263 atlas UNIBE-LHEP less urgent NGI_CH in progress 2021-01-19 UNIBE-LHEP: "Transfer failed: ... WLCG
150221 atlas CSCS-LCG2 less urgent NGI_CH involved in progress 2021-01-19 CSCS-LCG2: transfer error with ... WLCG
150220 lhcb CSCS-LCG2 very urgent NGI_CH in progress 2021-01-20 FTS3 transfers Failed to CSCS-LCG2 WLCG
150086   cms CSCS-LCG2 urgent NGI_CH assigned 2021-01-05 Lost files at T2_CH_CSCS EGI
150061 lhcb CSCS-LCG2 very urgent NGI_CH assigned 2021-01-01 Jobs Failed at CSCS-LCG2 WLCG
149166   cms CSCS-LCG2 less urgent NGI_CH assigned assigned 2020-12-17 CVMFS squids at CSCS EGI
147830   cms CSCS-LCG2 less urgent NGI_CH involved waiting for reply 2021-01-21 Enabling TPC transfers over davs WLCG
144485   none CSCS-LCG2 less urgent NGI_CH assigned in progress 2020-12-18 Upgrade to recent dCache release EGI

a.o.b
  • Attendants
  • CSCS:
  • CMS:
  • ATLAS:
  • LHCb:
  • EGI:

Topic attachments
I Attachment History Action Size Date Who Comment
PDFpdf ATLAST2ReportDec2020.pdf r1 manage 397.1 K 2021-01-21 - 11:51 GianfrancoSciacca ATLAS CH Tier2 report
PDFpdf CHIPPreportJan2021.pdf r1 manage 1223.8 K 2021-01-21 - 12:24 NickCardo  
PDFpdf UnibeT2ReportDec2020.pdf r1 manage 181.3 K 2021-01-21 - 11:51 GianfrancoSciacca UniBE Tier2 report
PDFpdf wlcg-swissops-cms-20210121.pdf r2 r1 manage 1360.9 K 2021-01-21 - 12:57 DerekFeichtinger CMS Report for Jan 2021

This topic: LCGTier2 > WebHome > MeetingsBoard > MeetingSwissGridOperations20210121
Topic revision: r8 - 2021-06-18 - MauroDonega
 
This site is powered by the TWiki collaboration platform Powered by Perl This site is powered by the TWiki collaboration platformCopyright © 2008-2024 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback