ATLAS UK Cloud Support
Zoom
https://cern.zoom.us/j/98434450232
Password protected (same as (new) OPs Mtg)
-
-
10:00
→
10:20
Status 20m
-
Outstanding tickets 10m
-
Outstanding tickets
-
155008 TEAM atlas UKI-SCOTGRID-ECDF urgent NGI_UK in progress 2021-11-22 14:29:00 UKI-SCOTGRID-ECDF failing transfers due to expired certificate EGI
- Transfers looking better, but need a confirmation
-
154940 TEAM atlas UKI-NORTHGRID-LANCS-HEP top priority NGI_UK in progress 2021-11-24 10:10:00 UKI-NORTHGRID-LANCS-HEP_DATADISK: Failed to stage-in file / File transfer timed out during stage-in EGI
- Discussion on how to proceed with decom of DPM and Commissioning of Ceph;
- To create Jiri’s for each action and prepare test cluster
- Previous tunings attempts did not help particularly
- Discussion on how to proceed with decom of DPM and Commissioning of Ceph;
-
154883 TEAM atlas UKI-NORTHGRID-LANCS-HEP less urgent NGI_UK in progress 2021-11-24 10:30:00 UKI-NORTHGRID-LANCS-HEP fails in transfers as destination EGI
- As above
-
154806 TEAM atlas UKI-LT2-QMUL less urgent NGI_UK in progress 2021-11-22 11:58:00 UKI-LT2-QMUL SOURCE transfer failures: [13] Result (Neon): SSL handshake failed EGI
- Still in progress, noting failures correlate with large batched transfer requests
-
154543 TEAM atlas UKI-SCOTGRID-ECDF urgent NGI_UK in progress 2021-11-12 11:44:00 DPM storage ACL configuration EGI
- Needs an update
-
154436 TEAM atlas RAL-LCG2 very urgent NGI_UK on hold 2021-11-10 13:52:00 RAL Echo Davs developments EGI
- Buffer
-
154200 TEAM atlas RAL-LCG2 less urgent NGI_UK on hold 2021-11-25 09:27:00 RAL-LCG2 deletion issues with error “The requested service is not available at the moment” EGI
- Testing for deletions working well so far
-
153367 TEAM atlas RAL-LCG2 urgent NGI_UK on hold 2021-11-10 14:08:00 HTTPS on RAL CTA EGI
- Awaiting testing
-
-
-
CPU 5m
New link for the site-oriented dashboard
-
Other new issues / tasks 5m
Renabling GPU queue for QMUL
https://atlas-cric.cern.ch/atlas/pandaqueue/detail/ANALY_QMUL_GPU/Upgraded dCache; Investigating dCache bug with copying output with NFS
SRR was looking ok, but had problems with SRR serviceUKI-LT2-BRUNEL_DATADISK (offline, needing new backplane)
RAL-LCG2 capacity reduction: https://its.cern.ch/jira/browse/ATLDDMOPS-5585
Xrootd 5.3.X (X>=3) needed for VP sites
Liverpool: retiring disk servers; space reduction needed
Wide variance in transfer speeds from Cern to RAL (via gridFTP); similar effects can be seen at other T1s, but RAL per-file throughput is slower than the average
-
-
10:20
→
10:40
Ongoing Items 20m
-
CentOS7 - Sussex 5m
-
TPC with http 5m
-
Storageless Site test (Oxford) 5m
-
- 10:40 → 10:50
-
10:50
→
11:00
AOB 10m
-
10:00
→
10:20