US ATLAS Computing Facility

Wednesday 26 Feb 2025, 13:00 → 15:25 US/Eastern

Facilities Team Google Drive Folder

Zoom information

Meeting ID:  993 2967 7148

Meeting password: 452400

Invite link:  https://umich.zoom.us/j/99329677148?pwd=c29ObEdCak9wbFBWY2F2Rlo4cFJ6UT09

 

 

13:00 → 13:05 WBS 2.3 Facility Management News

Speakers: Alexei Klimentov (Brookhaven National Laboratory (US)), Dr Shawn Mc Kee (University of Michigan (US))

A few top of the meeting items:

 

• We have a meeting to discuss possible Tier-2 shopping lists with the PI's and relevant managers tomorrow afternoon with a goal of agreeing on a plan we can send to Chris and John
• Friday morning those of us working on the Trusted CI engagement will meet to discuss our freeback from homework #1 and initial responses for homework #2
• Please remember to track milestone progress (WBS 2.3 working copy at https://docs.google.com/spreadsheets/d/1Y0-KdvsRVCXYGd2t-SqCEFlppZn_PjvUUVDGp2vJjc4/edit?usp=sharing )
• BNL has new (more strict) rules for international  travel
  • personal days, travel justification, number of partisipants per conference/WS

 

Upcoming meetings:  LHCONE/LHCOPN and HEPiX

13:05 → 13:10 OSG-LHC

Speakers: Brian Hua Lin (University of Wisconsin), Matyas Selmeci

Release (this week)

• XRootD 5.7.3
• cvmfs 2.12.6
• IGTF 1.133

13:10 → 13:30 WBS 2.3.1: Tier1 Center

Convener: Alexei Klimentov (Brookhaven National Laboratory (US))

13:10 Tier-1 Infrastructure

Speaker: Jason Smith

13:15 Compute Farm

Speaker: Thomas Smith

13:20 Storage

Speakers: Carlos Fernando Gamboa (Brookhaven National Laboratory (US)), Carlos Fernando Gamboa (Department of Physics-Brookhaven National Laboratory (BNL)-Unkno)

13:25 Tier1 Operations and Monitoring

Speaker: Ivan Glushkov (Brookhaven National Laboratory (US))

WBS 2.3.1.2 Tier-1 Infrastructure - Jason

• NTR

WBS 2.3.1.3 Tier-1 Compute - Tom

• gridgk06 upgraded to alma 9.5 condor 24
• gridgk07 closed to jobs, upgrade pending (this week)
• This will conclude upgrades to ATLAS T1 farm production CE infrastructure
• Added in production BNL_ARM resource (480 slots)

 

WBS 2.3.1.4 Tier-1 Storage - Carlos

• NTR

WBS 2.3.1.4 Tier-1 Operations & Monitoring - Ivan

• NTR

13:30 → 13:40 WBS 2.3.2 Tier2 Centers

Updates on US Tier-2 centers

Conveners: Fred Luehring (Indiana University (US)), Rafael Coelho Lopes De Sa (University of Massachusetts (US))

• Reasonable running in the past two weeks.
  • MWT2 (1.5 days) and OU (1 day) took downtime. 
  • CPB had a DNS incident over last weekend and was offline for about a day.
  • Otherwise good running...
• Two sites still finishing EL9 updates: MSU and UTA.
  • MSU is close to having their installation system working. 
  • UTA (SWT2_CPB) is done with all servers except the storage servers.
• We have decided to set a deadline of March 31 fo submission of this year's Procurement and Operations plans.
  • I will follow up on whether there are template milestones that we can adjust to the March 31 deadline.

13:40 → 13:50 WBS 2.3.3 Heterogenous Integration and Operations

HIOPS

Convener: Rui Wang (Argonne National Laboratory (US))

13:40 HPC Operations

Speaker: Rui Wang (Argonne National Laboratory (US))

TACC: running smoothly. Fully moved to token-based communication. No more issues seem on the shared file system

Perlmutter: 15% usage below the expectation. May need to improve the thoughtput

13:45 Integration of Complex Workflows on Heterogeneous Resources

Speakers: Doug Benjamin (Brookhaven National Laboratory (US)), Xin Zhao (Brookhaven National Laboratory (US))

still working on a solution to get nvidia libraries available to ATLAS jobs running on the NERSC GPU queue.  Testing using a container created from the merging of the nvidia cuda development RockyLinux 9 container and the Docker files from the Alma 9 adc grid containers developed and maintained by Alessandro DeSalvo.

 

Still need to pass into container needed environmental variables , create a work area and add mount points for /pscratch and /cvmfs.  then modify the pilot wrapper script.  etc

13:50 → 14:10 WBS 2.3.4 Analysis Facilities

Conveners: Ofer Rind (Brookhaven National Laboratory), Wei Yang (SLAC National Accelerator Laboratory (US))

13:50 Analysis Facilities - BNL

Speaker: Qiulan Huang (Brookhaven National Laboratory (US))

1. Working with openshift expert on mounting persistent storage like NFS within container. 

13:55 Analysis Facilities - SLAC

Speaker: Wei Yang (SLAC National Accelerator Laboratory (US))

14:00 Analysis Facilities - Chicago

Speaker: Fengping Hu (University of Chicago (US))

• Maintenence scheduled on Tuesday, March 4, 2025.
• Pod eviction issus on some of the nodes - Condor scratch is  on the root filesystem. We will put in some ways to mitigate the issue.

14:10 → 14:25 WBS 2.3.5 Continuous Operations

Convener: Ofer Rind (Brookhaven National Laboratory)

• WLCG DOMA BDT effort was restarted today - link to slides and minutes

14:10 ADC Operations, US Cloud Operations: Site Issues, Tickets & ADC Ops News

Speaker: Ivan Glushkov (Brookhaven National Laboratory (US))

• ADC Operations:
  • Data Carousel: Including analysis
    • A few test users already included
    • Still a few things to clarify (ATLASPANDA-1129)
  • MC Evgen
    • default maxFailure set to 3 (was 10)
  • ARC CE bug found and fixed.
    • XRD and EOS turls need davs replacing with https
    • fix should go into ARC7 and backportd to ARC6. Deployment timescale unknown.
    • This was the rewason for the SWT2 to SWT2 failing transfers
  • IAM to K8S switch scheduled for 3/10/25
    • Anything using voms-atlas-auth.app.cern.ch as token/proxy issuer will start failing
    • Still many tokens and proxies requested. Contacted all users (btw who is nathan.crawford@uci.edu?)
  • Started dedicated “Sites” section started in the new/developing “ADC Documentation”
    • First contribution: “How to add a remote_queue to your CE/gatekeeper”
    • Feel free to contribute anything that might be of help to other site admins
• US Cloud Operations:
  • Site Issues
    • NET2:
      • Now running all ATLAS workflows
    • OU_OSCER_ATLAS
      • Was still appearing in monitoring in downtime after end of downtime. Solved (ADCMONITOR-559)
    • Others
      • Due to Data Carousel configuration - a tape staging problem at TRIUMF was visible as high destination failure rate on all US sites. Solved (GGUS:2430)
  • Tickets
    • AGLT2:
      • GGUS:2431: Bad CVMFS mounts. Solved.
    • BNL:
      • GGUS:2428: High failure rate from gridgk04. Solved.
    • MWT2:
      • GGUS:2099: BGP tagging.
    • NET2:
      • GGUS:2404: Squid degraded due to power outage. Solved.
      • GGUS:2365: Failing transfers during Jumbo frames test. Solved.
      • GGUS:2097: BGP tagging.
      • ATLDDMOPS-5707: NET2 tape comissioning is advancing.
    • SWT2:
      • GGUS:2098: BGP tagging

14:15 Services DevOps

Speaker: Ilija Vukotic (University of Chicago (US))

• XCache
  • issues with gStream monitoring are being debugged
  • issues with 2 Oxford xcache nodes tests
• VP
  • working fine
• Varnishes
  • all working fine
  • writing documentation on how to deploy it
• ServiceY
  • writing documentation on how to deploy its Runner
  • stress testing of AF ads nodes 
  • stress testing of FAB
• AF
  • Assistant now can run bash commands and scripts.

14:20 Facility R&D

Speaker: Lincoln Bryant (University of Chicago (US))

• Testing flocking from UChicago AF to MWT2 with Docker-based containers + HTCondor overlay
  • Some early results during a MWT2 downtime, displacing OSG workloads with AF workloads. 
    • A bunch of parameters to tune, but for now we're submitting fairly non-aggressively, with each container set to 8 cores / 48 GB RAM (~VHIMEM equivalent)
    • Already identified a few things to fix - Singularity, for example, seems broken 
    • Users simply add "ALLOW_MWT2=True" in their job ad 
  • Should be generalizable to run elsewhere, but currently requires privilege. Might be possible without privilege for the containers, TBD.
• Starting a document describing WireGuard implementation requirements 

14:25 → 14:35 AOB