RAL Tier1 Experiments Liaison Meeting

Wednesday 6 Nov 2024, 13:30 → 15:50 Europe/London

Access Grid (RAL R89)

13:30 → 13:31 Experiment Operational Issues

13:35 → 13:45 VO-Liaison ATLAS

Speakers: Brij Kishor Jashal (TIFR, RAL, IFIC), Jyoti Prakash Biswal (Rutherford Appleton Laboratory)

13:45 → 13:55 VO Liaison CMS

Speaker: Katy Ellis (Science and Technology Facilities Council STFC (GB))

Katy missed the last 3 weeks - O&C week, CHEP and off sick.

SAM tests look good, other than the previously noted occasional irregular failure on the CE tests. AAA saw some failures in the middle of last week probably related to reboots made a few days before - typical behaviour. Service restarts in various order appeared to fix it.

Job performance: A week ago there were low performance jobs running at many sites - these were spotted and supposedly fixed. In the last 24 hours I see again low CPU efficiency at several T1s. Seems lots of T1s are running large amounts of Analysis jobs which are pulling the CPU eff down.

Mini-UK data challenge part 1 planned for week of 9th Dec (at least ATLAS and CMS). Intention is to mostly test Tier2s. Tier 1 will likely be used as a source, but the idea is not to push Echo significantly at this point. 

13:55 → 14:05 VO Liaison LHCb

Speaker: Alexander Rogovskiy (Rutherford Appleton Laboratory)

14:10 → 14:20 VO Liaison ALICE

Speaker: Alexander Rogovskiy (Rutherford Appleton Laboratory)

14:20 → 14:30 VO Liaison LSST

Speaker: Timothy John Noble (Science and Technology Facilities Council STFC (GB))

LSST limit currently set at 10. Checked with Wen Guan, and they are testing the functionality of not submitting pilots if there is nothing in the queue on the Dev PanDA now. When deployed on plod can change the limit and track behaviour.

 

Finalising of the messaging stack to enable RAL to be integrated into the site tests.

 

Ticket in with Grid Services to add Peter Love and Jennifer Adelman-McCarthy (CMS / LSST) to the UIs to allow them to access the LSST VO box, still waiting on the ticket to be addressed, as bounced between Jose, Tom, and James.

 

 Talking to DB team on 15th Nov about the scale and needs for the LSST Data Butler (10-100TB)

14:30 → 14:40 VO Liaison APEL

Speaker: Thomas Dack

14:45 → 14:55 WP-D - GPU, Data Management, Other

Speakers: Brian Davies (Lancaster University (GB)), Darren Moore

15:00 → 15:01 Major Incidents Changes

15:05 → 15:15 Summary of Operational Status and Issues

Speakers: Brian Davies (Lancaster University (GB)), Darren Moore

Callouts over weekend:

No call-outs over the weekend

Callouts:

There was one Echo call-out on the 28/10/24v: echo-manager01.gridpp.rl.ac.uk/ha-check_ceph_xrootd_webdav_service_callout. This was as the result of a load spike of FTS requests from Atlas.

Antares

Nothing of note to report.

Echo

The Echo team are continuing with their Nautilus → Pacific upgrade. To point, there have been no issues reported. Details of this intervention can be found at https://stfc.atlassian.net/wiki/pages/resumedraft.action?draftId=605061121

The tracking ticket for this intervention is here:

Batch Farm

Nothing of note to report

General 

The UK ARGUS service supported at the Tier1 at RAL will be decommissioned on the 2nd December 2024. This is after CERN’s notification of deprecating their instance back in July (https://twiki.cern.ch/twiki/bin/view/LCG/WLCGOpsMeetingWeek240729). 

After this date, the ARGUS service hosted at argusngi.gridpp.rl.ac.uk will be unreachable.

If sites are utilising their ARGUS instances for user mappings, we recommend migrating to native solutions found in Nordugrid ARC-CE, HTCondor-CE or similar.
Any questions or concerns, please inform the RAL Tier1 Grid Services team.

GGUS 

6 of 6 Tickets
Ticket-ID	Type	VO	Site	Priority	Resp. Unit	Status	Last Update	Subject	Scope
168905		lhcb	RAL-LCG2	less urgent	NGI_UK	in progress	2024-11-06	Access to the vobox	WLCG
168888		cms	RAL-LCG2	urgent	NGI_UK involved	on hold	2024-11-05	Request for Dual Stack Support on ...	WLCG
167369		cms	RAL-LCG2	urgent	NGI_UK involved	in progress	2024-10-14	CVMFS stratum-1 server at RAL	WLCG
165969		cms	RAL-LCG2	urgent	NGI_UK	on hold	2024-10-14	enabling IAM token access to ARC-CEs at ...	WLCG
165818		none	RAL-LCG2	urgent	NGI_UK	in progress	2024-09-02	ATTENTION - new TOKEN configuration for ...	EGI
165323		none	RAL-LCG2	less urgent	NGI_UK	on hold	2024-11-04	RAL-LCG2: Enabling tokens on ...	EGI

 

 

 

15:20 → 15:21 AOB

15:22 → 15:32 Any other Business

Speakers: Brian Davies (Lancaster University (GB)), Darren Moore