Operations Bulletin Latest

From GridPP Wiki
Jump to: navigation, search

Bulletin archive


Week commencing Monday 13th February 2017
Task Areas
General updates

Monday 6th February

  • There is a Benchmarking Working Group F2F in the pre-GDB slot this Tuesday. It covers experiment views (including benchmarking with production jobs), cloud benchmarking, issues with Haswell processors (magic boost investigation).
  • The GDB for February is this Wednesday. Topics include AFS phase-out, benchmarking, feedback on the EOS workshop and a baseline for WLCG T1 operations.
  • January R/A figures for the Tier-2 sites have been circulated:
    • ALICE. All OK
    • ATLAS
      • Sheffield 80%:82%
      • ECDF 87%:98%
      • RALPP 80%:80%
    • CMS. All OK
    • LHCb
      • Sheffield 71%:73%
      • ECDF 88%:100%
      • RALPP 83%:84%
    • Site responses so far:
      • RALPP reports that the load on their dCache is causing intermittent SAM failures.
      • Sheffield had an issue with ARGUS.
  • There was a short downtime to undertake updates of the GridPP website and the Vac websites on Friday 3rd February. No issues reported.
  • Cambridge/John: avoid frontier-squid-3.5.23-4.1
  • Winnie: Help to debug ARC-CE "This metric ... cannot be executed independently"
  • Minutes from the WLCG weekly ops meeting are available.


Monday 30th January


WLCG Operations Coordination - AgendasWiki Page

Monday 6th February

  • Minutes from the WLCG Data Steering Group meeting of 2nd February are available on the meeting agenda page.

Tuesday 31st January

  • There was a WLCG ops coordination meeting last Thursday 26th. Agenda | Minutes.
  • The next related meeting will be on 2nd February and this is the WLCG Data Steering Group. (Closed meeting).


Monday 16th January

  • The next WLCG ops coordination meeting is on 26th.
    • Downtimes proposal followup --> please send feedback by Jan 20;
    • Tape usage performance analysis.
Tier-1 - Status Page

Tuesday 14th February A reminder that there is a weekly Tier-1 experiment liaison meeting. Notes from the last meeting here

  • As reported last week the Castor 2.1.15 upgrade has been completed. However, there have been problems with CMS xroot redirection tests following this. The next change for Castor is to to update the SRMs.
  • The migration of LHCb data to the T10KD tape media has been completed. All Tier1 data is now on this media.
Storage & Data Management - Agendas/Minutes

Wednesday 18 Jan

  • Continued ZFS discussion
  • Updates from the non-LHC VOs, data-wise. Mostly no news, except for DiRAC, Cambridge HPC and Edingburgh EPCC are debugging and testing; LSST expecting to do MC workflows end of Jan.

Wednesday 11 Jan

  • ZFS presentation/discussion

Wednesday 04 Jan 2017

  • Happy! Generally everything ticked along nicely over the break
  • Interesting topics to pick up? Interest in ZFS again (see also Marcus's blog posts on the topic. GLUE2 for accounting?

Wednesday 21 Dec 2016

  • Report from Computing Insights UK (nee MEW)
  • Highlights and lowlights of the year (storage and data management only - which seems to have fared better than the world in general)

Wednesday 07 Dec

  • Good GridPP presence at the CLOUD-SIG workshop at Crick last week. Real Work™ included GridFTP-endpoint-in-the-cloud and Jupyter notebooks. How do you get data into a Jupyter notebook (from the grid)?
  • Opportunity to do more with data transfer between infrastructures.


Tier-2 Evolution - GridPP JIRA

Friday 3 February 2017

  • ATLAS MCORE VMs now successfully running PanDA production jobs.

Tuesday 31 January 2017

  • Vac 02.00 deployed at Liverpool and Manchester, and being used for installation underway at Birmingham.
  • ATLAS MCORE VMs in testing. Correctly fetch HTCondor jobs associated with MCOREVAC PanDA queues.
  • LHCb interruptible MC jobs in testing in VMs (and on LHCb HLT.) This will provide multiprocessor VMs that are flexible in duration and number of processors to fit available space.
  • Vcycle with OpenStack ID v3 support being used to manage LHCb tenancy at Yandex.ru, and in production at CERN, IN2P3-CC and Datacentred Ltd.

Thursday 5 January 2017

  • Vac 02.00 released, with SuperSlots mechanism for handling a mixture of different VM sizes on the same hypervisors.


Accounting - UK Grid Metrics HEPSPEC06 Atlas Dashboard HS06

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

Monday 30th January

The fusion VO has been removed from GridPP Approved VOs.

Monday 9th January

  • Tom put these together:
    • A snapshot of the GridPP UserGuide is now available as an offline document in the Zenodo repository:
    • ...and the v1.0 release of the LaTeX code used to compile it is here.
    • There are some minor differences (mainly hyperlink-related) to the online version [1], which has also had a v1.0 release here.


Tue 22nd Nov

  • More on Accounting audit tools

The process for ARC, Toque and VAC is now documented here.

https://github.com/gridpp/audit/wiki


Tue 15 Nov

  • Accounting audit tools

GridPP project in github used to host documents and scripts to audit work done accounts independently of APEL. So far, methods exist for ARC CE, and Torque batch system. Method for VAC still rough and being worked out by (e.g.) next week. To get the material:

git clone https://sjones-hep-ph-liv-ac-uk@github.com/gridpp/audit.git

Substituting you own github username. Or go to:

https://github.com/gridpp/audit

For write access, ask Andrew McNab.

  • Changes to CMS resource requirements in Approved VOs

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

Scroll to the Resources table at the bottom and check the new CMS requirements.

Tue 1 Nov

Publishing tutorial updated to use new wording for various measurements.

https://www.gridpp.ac.uk/wiki/Publishing_tutorial#Accounting_transmissions



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


General note

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

Interoperation - EGI ops agendas

Monday 16th January

  • There was an ops meeting on 9th January. Agenda.


Monday 7th November

  • There was an EGI Ops meeting today: agenda
  • UMD 3.14.5 released today
    • VOMS 3.5.0, which makes RFC proxies the default for voms-proxy-init
  • UMD 4.3.0 'October' release, release candidate ready, to be released by end of this week, including:
    • ARC, GFAL2, XROOT, Davix, dCache, ARGUS, Gridsite, edg-mkgrid, umd-release for CentOS7
  • please start using UMD4/SL6 or UMD4/CentOS7 instead of UMD3/SL6 & please don't use anymore EMI3
    • (think there may be a campaign around this soon)
  • Downtimes due to the vulnerability CVE-2016-5195: request an A/R recomputation
    • All the resource centres that were affected by the vulnerability CVE-2016-5195 and that declared a downtime between 2016-10-20 16:00 UTC and 2016-10-31 18:00 UTC are invited to request a recomputation of A/R figures for the days in which the downtime was ongoing.
Monitoring - Links MyWLCG

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.


Tuesday 31st March

Monday 7th December

On-duty - Dashboard ROD rota

Monday 16th January

  • Rota to be updated this week. Gordon is on shift this week.

Monday 9th January

  • Daniela posted on 25th Dec to say: "there's quite a lot of tickets - I tried to extend them all into January.

Otherwise apparently the grid is still standing."


Monday 5th December

  • Another fairly quiet week. There were a few new tickets, which are mostly now resolved. There was an intermittent SRM alarm at RALPP which sporadically appeared then disappeared; it finally hung around long enough today to raise a ticket.


Rollout Status WLCG Baseline

Tuesday 7th December

  • Raul reports: validation of site BDII on Centos 7 done.

Tuesday 15th September

Tuesday 12th May

  • MW Readiness WG meeting Wed May 6th at 4pm. Attended by Raul, Matt, Sam and Jeremy.


References


Security - Incident Procedure Policies Rota

Monday 6th February

  • One security incident reported by EGI CSIRT labelled EGI-20170107-01. Not known to affect any UK site.
  • Closure report circulated for incident EGI-20161214-01. Not known to affect any UK site.
  • Advisory-SVG-CVE-2016-9587: Up to 'HIGH' risk input validation vulnerability concerning Ansible reported by EGI SVG [1](see also Bulletin report 17-Jan below)
  • Summary of last week's EUGridPMA meeting by David Groep here [2]

Tuesday 31st January

  • Updated IGTF distribution 1.80 available for download [3]CHANGES
  • EUGridPMA certificate authorities meeting this week [4]
  • Communications challenge sent to 20 UK sites. 2 "failures" - no response within 4 hours. Thanks to all who participated.

Tuesday 24th January

  • EGI SVG UPDATE advisory concerning availability of patches for 'HIGH' risk CVE-2016-7117 [5]
  • EUGridPMA certificate authorities meeting next week [6]

Tuesday 17th January

  • SL6 updates published for CVE-2016-7117 which was downgraded by EGI SVG to HIGH risk [7]
  • Vulnerability disclosed for Ansible CVE-2016-9587. Upgrades are available. [8]
  • 2016/01/16 - Incident report involving one non-UK site published by EGI CSIRT as EGI-20161229-01.
  • EGI CSIRT F2F meeting this week.

The EGI security dashboard.


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).

Tuesday 14th February

  • The development dashboard is now operational again. However the production dashboard is not currently displaying latency results (OSG ticket submitted).
  • Birmingham pS host had lost its entire network connection and both Lancaster's hosts their IPv6 connection. Durham results are incomplete, GGUS ticket submitted (#126587).

Tuesday 25th October

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


Monday 19th September

  • UK eScience CA - certificate issuance problems. Jens reported that on 15th a partial but significant database corruption occurred on the signing system for the CA. Data was restored from (offline) backups but the rebuild was not correctly configured.
  • A large number of site admins and other GridPP supporters appeared to be suspended from the dteam VO last week. “During a planned upgrade operation of VOMS service, a system malfunction occurred. As a result, some users received false notification about membership expiration. We are in contact with the software development team in order to identify the cause.”


Tickets

Monday 6th February 2017, 14.30 GMT
21 23 Open UK tickets this month

FRESH IN THIS MORNING - BRISTOL
https://ggus.eu/?mode=ticket_info&ticket_id=126454 (7/6) As seen on TB-SUPPORT, CMS are having test failures at Bristol and Winnie is left without a CMS site support at the moment. I see some replies already on the list, I'll leave this slot here for hopefully helpful discussion. On Hold (7/6)

SUSSEX
125503 (9/12/16)
Sno+ file download failure ticket, due to the wrong SE name in the LFC for the files. Jeremy M reports that he is looking into created a DNS alias and asking the CA sage (aka Jens) to shape the necessary certificate. In progress (30/1)

122772 (11/7/16)
Webdav/xroot deployment ticket from atlas. Jeremy M reports the appointment of their new admin, which is great stuff. This is one of the first things on his todo list. I'll repeat the usual "we're here to help" message. No point suffering in silence! On hold (26/1)

Fresh in last night - 126438 - atlas seeing srmPut failures, but the error is 'file already exists'. A problem with rucio?

RALPP
125743 (27/12/16)
An availability ticket. A few blips on the nagios page, but I don't think there's anything to see here really. On Hold (29/1)

125815 (5/1)
Atlas ticket regarding space not being released after deletion. Chris has beaten his dcache into shape, and asked for the deletions to be re-attempted. Waiting for reply (30/1)

OXFORD
126371 (4/2)
Atlas transfer failures. Kashif spotted that the dpm-gsiftp daemon and failed, and got it back up. I suspect this ticket it can be closed if the daemon is stable? In progress (4/2)

121924 (2/6/16)
Perfsonar rate ticket? Any news? If not, is there likely to be any? On Hold (5/12/16)

125822 (5/1)
The Oxford edition of the "Space not released after deletion" issue. Kashif too has been tinkering his SE, tweaking and (re-)starting httpd daemons and asks for a fresh list of files to check. Waiting for reply (27/1)

BIRMINGHAM
126131 (24/1)
Availability ticket. The numbers are on the mend so the ticket is On Hold (30/1)

GLASGOW
125867 (9/1)
LHCB seeing cvmfs-related job failures on WNs at Glasgow. Gareth has updated cvmfs across the Glasgow nodes and asks if the issue has calmed down. Waiting for reply (31/1)

124052 (25/9/16)
Another LHCB ticket, about the arc publishing incorrect job numbers. Gareth provided an update regarding the Glasgow plans, rolling fixing this into the Centos7 migration. Thanks Gareth! On Hold (31/1)

EDINBURGH
126349 (3/2)
Another availability ticket, although today's numbers look to be okay so hopefully the cause of the troubles has passed. Looks like this ticket hasn't been noticed yet though. Assigned (3/2) Andy noted that the argo numbers seem nonsensical with negative availability for a few days! But things are on the mend now. Looks like a simple case of On Holding the ticket for the next 26 days.

LIVERPOOL
124819 (3/11/16)
The last AFS ticket, John B reports that the university has stopped firewalling UDP port 7001 and asks if things are better now. Waiting for reply (3/2)

126167 (25/1)
Decommissioning ticket for the last CREAM CE at Liverpool (which will also see the end of torque at the site). Downtime for the service will be on the 14th (Happy Valentine's Day?) and the service will be switched off properly come the 28th. In progress (30/1)

QMUL
125627 (19/12/16)
Atlas transfers failing to the QM test SE. Dan increased the space to 10TB to sooth the last batch of failures, just waiting to here if that worked. Waiting for reply (26/1)

126261 (30/1)
Biomed nagios tests not working for ce4 at QM. The problem persists. In progress (2/2)

126312 (1/2)
Atlas spotted QM's squid had fallen over. Dan has noticed problems since upgrading to v3 of frontier-squid, although the issues could also be related to IPv6 on the hosts (of the two squids at QM the one that fell over was also the one that has an IPv6 address in DNS). Keeping the ticket open to see if things stay up. In progress (1/2)

TIER 1
126296 (1/2)
CMS SAM tests failing against srm-cms-disk.gridpp.rl.ac.uk. All transfers "by hand" pass without trouble, and Gareth points out that this service is not in production in the GOCDB, so tests shouldn't even be running against it! Waiting for reply (6/2) Update - CMS got back that this is the endpoint specified in PhEDeX so this is why it was tested. If this is wrong it will need to be changed.

126376 (5/2)
Another batch of CMS SAM test failures. This includes the srm-cms-disk issue again. John K restarted the CMS xroot directors to try to clear the CE test errors that were being seen - things were looking up. In progress (6/2)

126184 (26/1)
Request from atlas for input on the new site monitoring schemes, linked in the ticket. The appropriate people were being chased. In progress (26/1)

124876 (7/11/16)
echo instance at RAL failing nagios tests due to the tests not using the right path. The ticket addressing this (125026) has had no progress since just before Christmas and so could do with a shake up. On Hold (1/1)

117683 (18/11/15)
Glue 2 publishing for Castor ticket. Did Jens and Rob have any luck tackling this in the pre-Christmas get together? On Hold (7/12/16)

Tools - MyEGI Nagios

29th November 2016

LSST monitoring is available through vo-nagios

https://vo-nagios.physics.ox.ac.uk/nagios/cgi-bin/status.cgi?servicegroup=lsst&style=detail

13th September 2016


VOs - GridPP VOMS VO IDs Approved VO table
Site Updates

Tuesday 23rd February

  • For January:

ALICE: All okay.

RHUL 89%:89% Lancaster 0%:0%

RALPP: 80%::80%

RALPP: 77%:77%

  • Site responses:
    • RHUL: The largest problem was related to the SRM. The DPM version was upgraded and it took several weeks to get it working again (13 Jan onwards). Several short-lived occurrences of running out of space on the SRM for non-ATLAS VOs. For around 3 days (15-17 Jan) the site suffered from a DNS configuration error by their site network manager which removed their SRM from the DNS, causing external connections such as tests and transfers to fail. For one day (25 Jan) the site network was down for upgrade to the 10Gb link to JANET. Some unexpected problems occurred extending the interruption from an hour to a day. The link has been successfully commissioned.
    • Lancaster: The ASAP metric for Lancaster for January is 97.5 %. There is a particular problem with ATLAS SAM tests which doesn’t affect the site activity in production and analysis and this relates to the path name being too long. A re-calculation has been performed.
    • RALPP: Both CMS and LHCb low figures are due to specific CMS jobs overloading the site SRM head node. The jobs should have stopped now.



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