2024-06-20 Meeting Notes

 Date

Jun 20, 2024

 Participants

 

  • @Thomas, Jyothish (STFC,RAL,SC)

  •  

  • Lancs:

  • Glasgow:

Apologies:

  •  

  •  

CC:

 

 

 Goals

  • List of Epics

  • New tickets

  • Consider new functionality / items

  • Detailed discussion of important topics

  • Site report activity

 

 Discussion topics

Current status of Echo Gateways / WNs testing

Recent sandbox’s for review / deployments:

 

Item

Presenter

Notes

 

Item

Presenter

Notes

 

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

@Thomas, Jyothish (STFC,RAL,SC)

 

Rocky 8 and 9 migration planning

 

see above

 

CMSD Load balancing

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

PR:
revised load balancing algorithm - weighed random selection by Jo-stfc · Pull Request #8 · stfc/xrootd

also see above

 

 

Gateway Auth failures

@Thomas, Jyothish (STFC,RAL,SC)

Significantly increased rate that a gateway fails to authenticate (failure mode is permission denied).

 

CHEP Abstract ideas

@Thomas, Jyothish (STFC,RAL,SC)

Accepted abstracts announced
Load balancing talk approved

 

XrootD Workshop plan

@Alastair Dewhurst

The workshop agenda and registration page can be found at:

XRootD and FTS Workshop @ STFC UK

We would kindly ask you to consider contributing to the workshop in the following areas:

  •  Status, statistics and projected usage of FTS and XRootD for Run3 and beyond.

  • Storage access for analysis use, including performance and CPU efficiency with emphasis on realistic running conditions (e.g., local disk, EOS, Tier2, remote data access, with or without XCache).

  • Data distribution management, needs and use-cases within scientific communities: data management needs of their community, frameworks used and how they integrate FTS and/or XRootD within their workflow.

  • Structured input and future requirements from the experiments and community members on the FTS medium-to-long term evolution.

  • R&D projects in the realm of data management, data access and data distribution within the Grid world are also invited. This could touch on integrating Cloud Storage within scientific data management workflows or Network projects aimed at optimizing data movements.

  • Discussion on connected topics such as Authentication frameworks (IAM, CILogon, etc) and the role of FTS and XRootd for emerging communities (e.g.: SKAO) in the world of scientific data management .

We encourage you to submit the title of your talk, along with how much time you would like to have to present it, using the form at XRootD and FTS Workshop @ STFC UK
deadline 1 August

Should the title not be sufficiently explanatory, please include a sentence or two of context.


data pipeline for another project in SCD (FTS/xrootd/antares)


 

Shoveller

@Katy Ellis

chat with cloud got superseded by network issues during the all hands. Given the ongoing os migration for both cloud and echo, this is best picked up again next month

Shoveler installation and monitoring

 

Future developments ideas planning work

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

Notes from planning meeting 22-04-2024

 

Deletion studies through RDR

@Ian Johnson

 

 

 

Deletions

https://stfc.atlassian.net/browse/XRD-83

 

 

Planning for ALICE CMSD redirection

@Thomas, Jyothish (STFC,RAL,SC)

no updates

 

Checksums fixes

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

WNs are currently redirecting checksums to external gws

 

Prefetch studies and WN changes

@Alexander Rogovskiy

ipv6 DNS poisoning rolled out

nullfs testing for xcache hit rate - full scale test on WN TBD

 

Tokens Status

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

 

 

SKA Gateway box

@James Walder

https://stfc.atlassian.net/wiki/spaces/UK/pages/215941180

gws in a working state, aq config port pending

 

 

 

 

 

Xrootd testing framework

@Mariam Demir

 

 

 

on GGUS:

Site reports

Lancaster: Restarting OSDs eventually fixed the backfilling slowness, data moving around at GB/s rather then MB/s. None the wiser for the root cause, or why these OSDs had problems. Some suspicion at QoS, but need to dig.

 

Glasgow:

Migration started, OS first.

in place upgrade worked well

 Action items

 

  •  

  •  

 

 Decisions