Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

Apologies:

CC:

\uD83E\uDD45 Goals

...

Item

Presenter

Notes

Operational Issues
Gateways and WNs:
- Current status and upcoming changes

Thomas, Jyothish (STFC,RAL,SC)

‘Memory allocation errors on WN proxy containers’

svc18 causing issues for cms SAM tests (davs refuse to connect)

memory allocation caused by 2 issues: non-pgrw vector read sizes were too large for direct reads, and the gateway containers were running out of memory.

WN container restart order can be improved

Compilation issues with XrdCeph and rocky 8 with 5.6+

Problems compiling XrdCeph alongside core xrootd (warning message → termination of compilation)

With standalone XrdCeph crashes

James Walder and Thomas, Jyothish (STFC,RAL,SC) to review the code and debugging.

.

CMSD Load balancing

Thomas Byrne Thomas, Jyothish (STFC,RAL,SC)

PR:
https://github.com/stfc/xrootd/pull/14
Code reviewed; item to be marked as done

Gateway Auth failures

Thomas, Jyothish (STFC,RAL,SC)

Auto-restarting on this failure mode is enable. Still observing occasional cases.

XrootD Workshop plan

Alastair Dewhurst

Shoveler

Katy Ellis

Shoveler installation and monitoring

Why is this on a cloud VM?
Access required into Cern.
To confirm that test WN is able to send data to the Shoveler instanceCC. for WN to report to Shoveler, not yet implemented.

Future developments ideas planning work

Ian Johnson Thomas, Jyothish (STFC,RAL,SC)

https://stfc.atlassian.net/wiki/spaces/X/pages/459997229/Notes+from+planning+meeting+22-04-2024?atlOrigin=eyJpIjoiNDRmNDEwOWI3Y2NhNDg5MDg4ZmZiYTNhNTliOWUwNmUiLCJwIjoiYyJ9

Deletion studies through RDR

Ian Johnson

Deletions

Jira Legacy
serverSystem Jira
serverId929eceee-34b0-3928-beeb-a1a37de31a8b
keyXRD-83

Preliminary figures, deleting 3000 small files from 10 gateways simultaneously, using 10 workers and 100 workers:

image-20240801-122652.pngImage Added

Seems encouraging, but small files so far. Need to check results for 100 workers as some target files had already been deleted. Moving on to record timings for deleting larger files in the GB range.

Planning for ALICE CMSD redirection

Thomas, Jyothish (STFC,RAL,SC)

restarted activities on this; looking at how server ‘logs in to the CMSD manager’

Checksums fixes

Alexander Rogovskiy Thomas, Jyothish (STFC,RAL,SC)

Prefetch studies and WN changes

Alexander Rogovskiy

View file
namexrootd_restarts.pdf

Prefetch activities complete.

Tokens Status

Thomas, Jyothish (STFC,RAL,SC) Katy Ellis

SKA Gateway box

James Walder

/wiki/spaces/UK/pages/215941180

AQ configuration for Bonded VLANs seems OK. Netbox config for {02,03} seem ok.

Issues with PXE booting and OS installation due to routing between AQ and hosts. JC and JA to look into this

Xrootd testing framework

Mariam Demir

Testing Framework - YTBN

 

 

on GGUS:

Site reports

Lancaster: Since dropping a bunch of data everything is a lot happier, although “scrubbing weirdness” is still an issue. Gerard’s on holiday so that’s the most precise we’ll get this week.Glasgow: Working on/with Reef; (Thomas, Jyothish (STFC,RAL,SC) links to https://github.com/ceph/ceph-build/pull/2272 )

Glasgow: Man with Reef on el9

✅ Action items

⤴ Decisions

...