2023-09-07 Meeting Notes

 Date

Sep 7, 2023

 Participants

  • @Thomas, Jyothish (STFC,RAL,SC)

  • @Alexander Rogovskiy

  • @Alastair Dewhurst

  • Lancaster:

  • Matthew Steven D.

  • Gerard Hand

  • Glasgow:

    • Samuel Skipsey

Apologies:

@James Walder

 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

 

XRootD Releases

 

No news on 6.2 yet

 

Prefetch studies

Alex

prefetch works but needs an increased timeout env variable

 

Deletion studies through RDR

Ian

 

 

CMSD rollout

 

https://stfc.atlassian.net/browse/XRD-41

Status:

webdav alias has been pruned

 

 

CMSD operations: observations

 

gateways under heavy system load:

alice gateways were suspected to be causing issues with the ceph cluster due to read amplification caused by the xrdceph buffers:


however removing the buffers directly correlated with a general increase in op times so the change was reverted:

Oxford VP moved to QMUL
Sheffield is also going to be moved



 

CMSD outstanding items

 

Icinga / nags callout tests changes. - live and available

Improved load balancing / server failover triggering -

better 'rolling server restart script'

Documentation; setup / configuration / operations / troubleshooting / testing

Review of Sandbox and deployment to prod.

 

Tokens testing

 

NTR

 

AAA Gateways

 

Sandbox ready for review:

http://aquilon.gridpp.rl.ac.uk/sandboxes/diff.php?sandbox=jw-xrootd-aaa-5.5.4-3

 

SKA Gateway box

 

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

now working using ska pool on ceph dev

 

extra gateways deployment

 

DNS round robin for internal (batch farm) use of the gateways on the new network pending ipv6 external access ~2 weeks

 

ALICE WN gateways

 

(Birmingham using eos, Oxford no storage)

 

on GGUS:

Site reports

Lancaster - no developments. Matt is pondering how to get xrootd to perform “central banning”.

Sam Skipsey - higher load on ceph (possibly from validation), swapiness disk set to 0 from 10

Does OS upgrade preserve the disk in Ceph? probably yes but not tested on sites. Gerard (Lancs) might do this

 Action items

  •  

 

 Decisions