RAL Tier1 Experiments Liaison Meeting
→
Europe/London
Access Grid (RAL R89)
Access Grid
RAL R89
-
-
14:00
Major Incidents Changes
-
1
Summary of Operational Status and IssuesSpeakers: Brian Davies (Lancaster University (GB)), Darren Moore, Kieran Howlett (STFC RAL)
-
14:10
Experiment Operational Issues
-
2
VO Liaison CMSSpeaker: Katy Ellis (Science and Technology Facilities Council STFC (GB))
Gsiftp now completely removed from CMS activities including WN uploads.
Tape REST API is in production and we see some successful davs transfers already. Katy is monitoring.
Antares updated to CTA 5 - no issues observed.
Cap removed on 12k CMS cores running simultaneously - however I have not seen the number go above 12k in Vande.
Concerning the DUNE usage of Echo - they acknowledge they went over pledge and are investigating (behaviour in Rucio, dark data and so on).
-
3
VO-Liaison ATLASSpeakers: Dr Brij Kishor Jashal (RAL, TIFR and IFIC), Jyoti Prakash Biswal (Rutherford Appleton Laboratory)
-
4
VO Liaison Others
-
5
VO Liaison LHCbSpeaker: Alexander Rogovskiy (Rutherford Appleton Laboratory)
LHCb:
- First 2024 data arrived to RAL
- LHCb submitted huge WGProduction last week, it overloaded ECHO and resulted in many vector read failures
- This indicates that chanes are indeed necessary
- No prefetch was rolled-out to 2021 gen last Friday
- In the long term we should aim for proxy removal, in my opinion
- This indicates that chanes are indeed necessary
- Checksums stopped working on 3 WNs from the preprod farm
- Due to tests
- Xrootd bug follow-up
- Lists of affected files created
ALICE:- Free space values in SRR does not match CRIC, needs correction (ticket CS-147)
- Do we have working xrootd space reporting?
-
6
VO Liaison LSSTSpeaker: Timothy John Noble (Science and Technology Facilities Council STFC (GB))
- LSST jobs running on batchfarm well
- new job type added to utilise the Butler Metadata hosted at RAL
- Currently failing due to configuration error but cannot correct until X509 user cert renewed
- new job type added to utilise the Butler Metadata hosted at RAL
- LSST Rucio currently in scaled back operations due to DB issues with load
- psycopg-binary for the client currently thought to be the issue
- I have suggested it is more of a DB issue as I have seen the error on my instance when close to its maximum connections allowed
- LSST jobs running on batchfarm well
-
7
VO Liaison APELSpeaker: Thomas Dack
-
14:45
AOB
-
8
Any other BusinessSpeakers: Brian Davies (Lancaster University (GB)), Darren Moore
-
14:00