Operations Bulletin Latest

From GridPP Wiki
Jump to: navigation, search

Bulletin archive


Week commencing Monday 5th March 2018
Task Areas
General updates

Monday 5th March

  • For the WLCG-HSF workshop: ideas, plans or proposals for technology watch activities requested. Please let Helge et. al. know by Monday 19 March. They would particularly appreciate proposals in the form of a single PowerPoint slide.
  • GridPP technical meeting this Friday will focus on CentOS7. (Here is a link to 2nd March agenda for Vidyo details).
  • WLCG baseline for Singularity - 2.4.2
  • Need for JISC list VO-ADMINS.
  • WLCG T2 availability & reliability - January 2018

Monday 26th February

  • WLCG-HSF Workshop in Naples
  • Minutes of the weekly WLCG weekly ops meeting are available here.
  • Alessandra: Mentioned here grid enabled GPU behind an ARC-CE implementation.
  • GalDyn and GridPP. Would like to submit work to DiRAC and make use of GridPP resources.
  • GridPP40 registration is open.
  • Rajan: SL7 UI Machine

Monday 12th February

WLCG Operations Coordination - AgendasWiki Page

Monday 5th March

  • The agenda and minutes from the WLCG ops coordination meeting last week are available.
  • Special topics were: CNAF recovery update; SAM recalculation policy draft discussion and Progress on WLCG Storage Space Accounting Prototype including tape metrics.

Tuesday 27th February

Monday 29th January

  • The majority of T1 sites and VOs confirmed that they do use SAM tests from the critical profiles and regularly check A/R reports.

Suggestions for improvements:

    • Propose policy for accepting A/R recalculation requests. The draft should be reviewed at the next meeting.
    • VO should have flexibility in the definition of the critical profile. Though the impact of the changes in the critical profile should be carefully tested by the VO and should be announced in advance to the sites, no approval of the MB is required.
    • Make tests as close as possible to the real production flow. ATLAS is planning some work in this direction.
    • The proposal to include real production flows in the critical profile was not supported.
    • Sites which do have local fabric monitoring like Nagios for example, are recommended to use an API to import test results into the local fabric monitoring. **This would help to avoid test failures staying unnoticed for months.
    • Transparent navigation from the SAM UI to the log files is required to facilitate test failure debugging. This feature has to be preserved in the new SAM UI being developed by the monitoring team.
Tier-1 - Status Page

Tuesday 1st March Report from the last Tier1 - Experiments Liaison Meeting at on 21st February is here.

  • Due to resource availability issues during the inclement weather last week the scheduled intervention/outage on Castor (originally planned for Thursday 1st March) was cancelled. This intervention has now been rescheduled for March 27th.
  • The investigations into the broken RAL FTS service and failing transfer to RAL-LCG2-ECHO are continuing. Update to applied to the relevant GGUS tickets as appropriate (ID#133752 and ID#133719 respectively).
Storage & Data Management - Agendas/Minutes

Wed 06 Dec

  • Another success story - CERN@School running machine learning at QMUL
  • Technical events coming up - hepsysman and cloud workshop at Crick. We ought to do something Useful or Interesting.

Wed 15 Nov

  • More xcache progress at some sites; less at others who may be blocking on dependencies or be busy with Other Things(tm)
  • Some interest in following the non-X.509 authentication/authorisation

Wed 08 Nov

  • Not (very) modest at all progress on xcache testing... from Chris at RALPP

Wed 01 Nov

  • (very) modest progress on xcache testing...
  • Except that RALPP has it fully working for dCache - hope to get full report next week


Tier-2 Evolution - GridPP JIRA

Tuesday 6 Mar

  • Preparing Vac 3.0 release with Docker (and Singularity) container support, and updated Vacuum Pipes which allow VOs to define multiple types of VM without having to notify sites each time.
  • Preparing Vcycle release with updated OpenStack API support, Vacuum Pipe support, and support for space-wide shutdown of VMs.
  • LHCb DIRAC VMs have been updated to shutdown gracefully when signalled by OpenStack; GridPP DIRAC VMs to follow. This is part of a wider mechanism to allow sharing of resources between OpenStack projects of different priority, with eviction of lower priority VMs on demand.


Accounting - UK Grid Metrics HEPSPEC06 Atlas Dashboard HS06

Tuesday 6th February


Tuesday 30th January

Tuesday 24th Oct


Monday 16th January

  • The discussion topic for next week will be accounting comparisons. Please note Alessandra's comments last week.

Monday 14th November

  • Alessandra has written an FAQ to extract numbers from ATLAS and APEL avoiding the SSB.

Monday 26th September

  • A problem with the APEL Pub and Sync tests developed last Tuesday and was resolved on Wednesday. This had a temporary impact on the accounting portal.
Documentation - KeyDocs

Tuesday 22nd Jan 2018

CHANGE TO APPROVED VOs

  • New Approved VO

A VO, solidexperiment.org, has been included in the GridPP Approved VOs document. Sites are requested to support this experiment, resources permitting.

https://www.gridpp.ac.uk/wiki/GridPP_approved_VOs#Approved_Local_VOs

  • In the Ops Portal, LHCb mentions a new cvmfs mount, /cvmfs/lhcb-condb.cern.ch
  • A voms server for magic, voms02.pic.es, has an updated DN.

Tuesday 9th Jan 2018 Changes to Approved VOs.

a) VOMS servers for planck, ipv6.hepix.org and enmr are now restored to as they were before the flooding at CNAF. b) The DN of a VOMS server for magic has changed.

https://www.gridpp.ac.uk/wiki/GridPP_approved_VOs

Tuesday 5th Sept 2017

New Approved VO. SKA European regional data centre, skatelescope.eu

https://www.gridpp.ac.uk/wiki/GridPP_approved_VOs

This VO is ramping up and has requested support. Andrew McNab is the contact person for GridPP.

General note

See the worst KeyDocs list for documents needing review now and the names of the responsible people.

Interoperation - EGI ops agendas Indico schedule

Tuesday 27th February

Tuesday 30th January

Tuesday 23rd January

  • There was an EGI Operations Meeting last Monday: Agenda
  • Middleware
    • Long list of updates in UMD 4.6
    • UMD3 deprecation - start using UMD4 as soon as possible
    • Products missing, let them know - this is a heads up
  • Operations
    • New weights in A/R
    • "Several sites are still missing the necessary information for computing the weights: check on VAPOR the values published by your sites in order to properly publishing in the GLUE2 schema the number of logical CPUs and the Hep-Spec06 benchmark."
    • http://operations-portal.egi.eu/vapor/resources/GL2ResSummary
    • IPv6
    • https://wiki.egi.eu/w/index.php?title=IPV6_Assessment
    • webdav probes in production
    • "During the January OMB we are going to discuss the inclusion of the probes in the operators and in the critical profile" [may be superceded]
    • Storage accounting deployment
    • "IMPORTANT: be sure to have installed the star-accounting.py script v1.0.4"
Monitoring - Links MyWLCG

Tuesday 4th July

  • There were a number of useful links provided in the monitoring talks at the WLCG workshop in Manchester - especially those in the Wednesday sessions.

Monday 13th February

  • This category is pretty much inactive. Are there any topics under "monitoring" that anyone wants reported at this ops meeting? If not we will remove this section from the regular updates area of the bulletin and just leave the main links.

Tuesday 1st December


Tuesday 16th June

  • F Melaccio & D Crooks decided to add a FAQs section devoted to common monitoring issues under the monitoring page.
  • Feedback welcome.


On-duty - Dashboard ROD rota

Monday 5th March

  • AM->GS. A quiet week last week.

Monday 29th January

  • There were a couple of alarms which showed up on the dashboard, but not on the corresponding nagios.
  • DB got fed up with Imperial failing the job-cancel test for CREAM -- the CE is just busy and returns the cancelled state about 30 s after the test 'fails' -- so she filed a ticket questioning the usefulness of the test: https://ggus.eu/?mode=ticket_info&ticket_id=133004.

Monday 20th November

  • Generally quiet. There are three outstanding tickets: low availability at Birmingham, one at Liverpool which might just have gone green, and out-of-date IGTF CAs at Sheffield.
Rollout Status WLCG Baseline

Monday 20th November



Historical References


Security - Incident Procedure Policies Rota

Monday 5th March

  • memcached exploits and ...


Tuesday 6th February on 2018

  • Update on Meltdown/Spectre
  • Security Team meeting this afternoon




Services - PerfSonar production dashboard |PerfSonar development dashboard | GridPP VOMS

- This includes notifying of (inter)national services that will have an outage in the coming weeks or will be impacted by work elsewhere. (Cross-check the Tier-1 update).

Monday 19th February

Please could sites upgrade their perfsonar hosts to CentOS7. Instructions are here. Current OS versions here.

Monday 5th February

  • 'Campus network engineering for data-intensive science workshop' in October 2016, had this agenda. Note particularly the Imperial and Cambridge overviews.
  • There is a JISC document which seeks to help set expectations for what the network can deliver.


Monday 22nd January

Monday 20th November

  • perfSONAR: No news.
  • GridPP DIRAC: DPM & xrootd version. Upgrade due 16th November.
  • RIPE ATLAS probes: Status & tests TBC
  • The next LHCOPN/LHCONE meeting will be at RAL 6-7th March 2018. The UK position on LHCONE is being reviewed. RAL T1 will (likely) connect via the LHCOPN in the coming months. T2s watch this space!


Monday 23rd October

  • The LHCOPN/ONE meeting took place last week at KEK, co-hosted with the HEPiX Fall meeting.
    • You can find a short report of the meeting here.
    • All the slides presented are available in the agenda.

Tuesday 4th July

  • Pete C is preparing another networking forward look document. Note that figures presented in the Manchester workshop were for a Tier-2 (not always a single site).

here

  • Duncan has recreated the UK perfSONAR mesh. Link here!


Tickets

Monday 5th March 2018, 14.30 GMT
4 Open Tickets this month.

IPv6 Deployment Tickets
Sussex: 131617
Possibly on hold until mid-2018.
RALPP: 131616
Chris had an encouraging update back in January, but hit some snags with a new Perfsonar install. Any joy?
OXFORD: 131615
No update since stating you had dual-stacked Perfsonar boxes back in November. Anything to add? Thanks for the update.
CAMBRIDGE: 131614
No progress expected until the Summer of this year. Is this still the case?
BRISTOL: 131613
Last update hoped progress could happen by February, any news? No recent news
BIRMINGHAM: 131612
Some progress on the v6 infrastructure news, hopefully the bugs Mark described a few weeks back can be ironed out.
GLASGOW: 131611
Gareth provided a recent, if not totally positive, update.
ECDF: 131610
There were some interesting times last week when taking the first steps in dual-stacking the ECDF DPM broke things. Keeping to dual-stacking their test DPM for now.
DURHAM: 131609
Last update at the end of January had no positive movement from central IT on v6 deployment.
SHEFFIELD: 131608
This ticket really could do with an update - even an unexciting one.
MANCHESTER: 131607
IIRC I think reverse lookup works only for the Perfsonar boxes - the ticket could do with an update about this.
LIVERPOOL: 131606
Another ticket that could do with an update, even if it's a boring one. John provided a brief update.
UCL: 131604
No news from central IT at last check back in January.
RHUL: 131603
Perfsonar dual-stacked, but DNS lookup not supported yet.

Common or Garden Tickets

SUSSEX
122772 (11/7/16)
Webdav/Xroot ticket. Some good looking progress on getting this to work, although at last check Leo hit some more problems. In progress (7/2)

133325 (6/2)
Availability ticket. Hopefully given another week of smooth running this can be closed. In progress (12/2)

RALPP
133819 (4/3)
LHCB asked RALPP to provide details of nodes without any SSE4.2 support. As Chris instructed the ticket was reopened by LHCB to request lhcb jobs do not land on these nodes. Reopened (4/3) Update - solved, the nodes are being decommissioned very soon.

OXFORD
133809 (3/3)
Availability ticket, caused by the AC troubles. On hold (5/3)

BRISTOL
133762 (1/3)
CMS Transfer problems, on hold until Friday. On Hold (5/3)

133806 (2/3)
CMS asked sites to deploy Singularity by March 2018, this ticket is the follow up. On hold (5/3)

BIRMINGHAM
129930 (4/8/17)
Atlas http SAM tests failing. Any luck with the puppet scripts Kashif shared with you? On hold (13/2)

GLASGOW
133667 (23/2)
LHCB data access problems at Glasgow. The ticket tailed off a bit, Andrew McNab has offered to help compare Glasgow and Manchester settings. In progress (5/3) Update - everything looks good now after Sam updated xroot across the Glasgow storage. Maarten noted in the xroot changelog the likely fix. I should imagine this ticket can be closed now.

DURHAM
133338 (7/2)
Atlas jobs failing at Durham, with the problems likely to be related to the Arc Control Tower handling of pilots. Adam rolled out some changes, have these fixed things? In progress (21/2)

SHEFFIELD
133019 (24/1)
Availability ticket. Ticking along. On hold (1/3)

133810 (3/3)
Sno+ jobs failing due to cvmfs errors on a node, which Elena has offline. I suspect that that's this ticket done with. In progress (4/3)

133770 (2/3)
LHCB jobs failing due to problems on some WNs, Elena has been fixing them, hopefully it's all sorted now. In progress (3/3)

MANCHESTER
133716 (27/2)
Atlas deletion errors - it looks like this ticket has been missed. Assigned (27/2)

QMUL
133402 (9/2)
A good portion of Sno+ jobs failing at QM, due to stage in/out errors. This is likely caused by the reduced network bandwidth being hogged by atlas. Hopefully this will be fixed soon (by restoring the 20GB/s site connection). In progress (22/2)

132713 (4/1)
hyperk.org support ticket. Any news? In progress (6/2)

132929 (18/1)
CMS having problems due APEL's problem parsing slurm logs (or something like that). APEL support have been called in, but no news yet. In progress (29/1)

IMPERIAL
133683 (24/2)
Atlas seeing a high job failure at Imperial, due to problems with their AGIS configs that they have no control over. Elena proposes closing the ticket and moving the conversation to JIRA. In progress (5/3) Update - atlas are waiting on seeing some running jobs before closing the ticket

133818 (4/3)
Another LHCB asking how many nodes do not have sse4.2 support. Simon reports there are no plans to decommission these nodes yet. Waiting for reply (5/3)

133723 (27/2)
This is a ticket for the Cloud site, Sno+ saw problems. Simon was investigating, and has offlined the cloud site in Dirac to prevent further failures. In progress (27/2) Update - Simon hasn't managed to reproduce any errors, and has suggested closing the ticket for now, reopening if needed.

132688 (3/1)
Another not really an Imperial ticket, I think this lost Pheno file ticket can be closed soon. In progress (29/1) Update - ticket closed

TIER 1
133719 (27/2)
Atlas spotted tranfers failing into Echo. It was being investigated, any news? In progress (27/2)

133752 (1/3)
Atlas noticed the FTS was broken. Was investigating Alastair noted that it appears to be an IPv6 issue. In progress (1/3)

133717 (27/2)
Likely related, a similar sounding CMS ticket. Any news? In progress (27/2)

133619 (21/2)
Missing unmerged CMS files at RAL. Chris has been helping a lot, but has asked CMS to double check his working. Waiting for reply (5/3)

133764 (1/3)
Sno+ ticket about the RAL BDII not having SFU information. It looks like the bdii information has recently changed (for the worse). Any news? In progress (2/3) Update - Karin has updated the ticket saying that things have got a lot worse for Sno+, upping the ticket's priority.

132589 (21/12/17)
LHCB killed pilots ticket. Some more investigations into this show that the problem is getting worse. Any luck with your investigation? In progress (23/2)

132708 (4/1)
WMS decommissioning ticket. Nothing to do here until next month I don't think. In progress (18/1)

127597 (7/4/17)
CMS network performance ticket. No news since Chris' comprehensive update in January. On hold (29/1)

124876 (7/11/16)
ECHO gridftp ROD tests not working, due to problems with the tests. No news on the counter ticket, still. On hold (13/11/17)

117683 (18/11/15)
GLUE2 publishing for Castor. A quick update in January reports a prototype version is being tested. On hold (3/1)


Tools - MyEGI Nagios

Monday 20th November

Tuesday 18th July

  • Following our ops discussion last week, Steve will focus his tests on supporting the GridPP DIRAC area and decommission the other tests.


VOs - GridPP VOMS VO IDs Approved VO table

Monday 20th November

  • Tom Whyntie has requested (and been granted) access to the GridPP VO to get some pipelines working for large-scale processing and analysis of MRI scans associated with the UK Biobank project.
  • All VOs in the incubation page being prompted for updates by the end of November (required input for OC documents).
  • QMUL (Steve L) is following up on the biomed MoU. GridPP want to be cited in research papers for the support our resources/sites provide.


Site Updates

Date



Meeting Summaries
Project Management Board - MembersMinutes Quarterly Reports

Empty

GridPP ops meeting - Agendas Actions Core Tasks

Empty


RAL Tier-1 Experiment Liaison Meeting (Wednesday 13:30) Agenda Meeting takes place on Vidyo.

Highlights from this meeting are now included in the Tier1 report farther up this page.

WLCG Grid Deployment Board - Agendas MB agendas

Empty



NGI UK - Homepage CA

Empty

Events
UK ATLAS - Shifter view News & Links

Atlas S&C week 2-6 Feb 2015

Production

• Prodsys-2 in production since Dec 1st

• Deployment has not been transparent , many issued has been solved, the grid is filled again

• MC15 is expected to start soon, waiting for physics validations, evgen testing is underway and close to finalised.. Simulation expected to be broadly similar to MC14, no blockers expected.

Rucio

• Rucio in production since Dec 1st and is ready for LHC RUN-2. Some fields need improvements, including transfer and deletion agents, documentation and monitoring.

Rucio dumps available.

Dark data cleaning

files declaration . Only Only DDM ops can issue lost files declaration for now, cloud support needs to fill a ticket.

• Webdav panda functional tests with Hammercloud are ongoing

Monitoring

Main page

DDM Accounting

space

Deletion

ASAP

• ASAP (ATLAS Site Availability Performance) in place. Every 3 months the T2s sites performing BELOW 80% are reported to the International Computing Board.


UK CMS

Empty

UK LHCb

Empty

UK OTHER
  • N/A
To note

  • N/A