Operations Bulletin Latest

From GridPP Wiki
Jump to: navigation, search

Bulletin archive


Week commencing 14th December 2015
Task Areas
General updates

Tuesday 7th December

  • A reminder that those going to the WLCG workshop in February need to register.
  • Janet networking problems. The need for closed discussion.
  • The GridPP website has now been moved - thanks to Tom and Andrew for getting it all ready.
  • An ARGUS developer meeting was held last week.
  • Please note the updated GridPP travel policy.


Tuesday 1st December

  • Switch over of the GridPP website. Soft launch this week on Tuesday with some unavailability expected during the afternoon.
  • The next WLCG GDB is on Wednesday 9th December. [ https://indico.cern.ch/event/319754/ Agenda].
  • Change of HEPSYSMAN dates due to Manchester room availability. Now anticipate 14th/15th.
  • Installing a UI on SL.
  • Core ops members - please provide your task list for Q3 15!
  • EGI: How many apel-publishers on sl5 in your NGI? (Instances in GOC-DB registered as glite-APEL.)
  • EGI: What is your WMS usage?
  • The dilemma: foreman/puppet or ansible?
  • Release date for GOCDB v5.5 is Wed 2nd December.
WLCG Operations Coordination - AgendasWiki Page

Tuesday 7th December

  • There was a WLCG operations coordination meeting last Thursday. Minutes | Agenda.
    • Please register for the WLCG workshop.
    • News: RedHat now fixing the openldap crash issue affecting Top BDII and ARC-CE. Stay tuned.
    • T0: LSF 9 software deployed on all worker nodes
    • T1: CC-IN2P3 & PIC: Globus host certificate validation changes
    • T2: NTR
    • ALICE: heavy ion run has been smooth from the grid perspective
    • ATLAS: Tier-0 performance in terms of events/second reconstructed from the whole cluster are quite low (few tents of Hz), observed huge I/O wait in Wigner spinning disks nodes. Plan full reprocessing campaign start for the 14th of December.
    • CMS: CMS Tier-0 workflows is driving some CERN Openstack hardware to its limits: GGUS:118056.
    • LHCb: Significant MC generation in-coming. MC simulation workflows have been executed successfully on commercial clouds, on both DBCE (up to 600 simultaneous jobs running) and Azure.
    • glexec: NTR
    • M/J features: Ongoing discussions clarifying key/value pairs. Next steps to review experience with implementations and installations, and update in view of technical note discussions.
    • HTTP TF: NTR
    • Infro sys future: Future Use Cases Document is now ready in the WLCG Document Repository (PDF). Looking at information system owned by WLCG (an interesting idea). Starting to prepare a Roadmap to GLUE 2.0.
    • IPv6: VOMS still does not work with IPv6. ARGUS not really tested but no problem expected as Java has good IPv6 support.
    • MW readiness: Virtual meeting on 2nd December. Verification workflows in progress listed here.
    • MC: NTR
    • Network/transfer metrics: NTR
    • RFC proxies: CMS have switched test pilot factories to RFC proxies.
    • Squid monitoring/HTTP discovery: NTR


Monday 30th November


Tier-1 - Status Page

Tuesday 8th December A reminder that there is a weekly Tier-1 experiment liaison meeting. Notes from the last meeting here

  • We are investigating why LHCB batch jobs sometimes fail to write results back to Castor (and the sometimes fail to write remotely as well). A recent change has improved, but not fixed, this problem.
  • We are continuing with the detailed network changes needed to remove our old core switch from the network. This work should be completed this week. As part of this we have moved the link from our core network to the "UKLight" router (that provides our external data path) this morning. There is another change that affects Castor tomorrow morning and a 'warning' has been declared in the GOC DB.
  • As reported last week, the tenders for the next round of Disk and CPU purchases are now out.
  • We have seen some increased data transfer rates to/from the Tier1. We have a plan to increase the bandwidth on the bypass link (to Tier2s) from 10 to 20 Gbit.
  • Tape plans are being reviewed. In particular we are looking to move LHCb and then Atlas from the T10KC (5TB) to T10KD (8TB) generation tapes as this enables us to put more data on each tape. Details yet to be finalized.
Storage & Data Management - Agendas/Minutes

Wednesday 09 Dec

  • Filesystem dump at Cambridge successful, no problems, awaiting feedback from ATLAS
  • T2C testing at Oxford "nearly ready", also need ATLAS to pick up the gauntlet now (CMS postponed till 2016)

Wednesday 02 Dec, not brought to you from the Brazilian Jungle.

  • Implications of diskless T2s (aka T2C) - ongoing discussion - need results from tests soon...
  • Several sites with old/decyaing/decommissioned hardware
  • Namespace dump testing updates

Wednesday 18 Nov

  • New member from Edinburgh!
  • Updates on PRODDISK cleanup at T2s, catalogue synchronisation (aka syncat), and what does "world readable" mean?
  • Summary of EGI community forum - interoperation and standards for moving data

Wednesday 11 Nov

  • Puppet for configuration

Wednesday 04 Nov

  • Brian presented GridPP achievements at the E2E workshop
  • Storage accounting with EGI
Tier-2 Evolution - GridPP JIRA

Tuesday 8 Dec

  • Liverpool has set up Vac with 126 VM slots. LHCb production MC and GridPP SAM tests running.
  • GridPP DIRAC VMs now working with new dirac.gridpp.ac.uk service: needed an ad-hoc configuration value adding (/Resources/Computing/CEDefaults/VirtualOrganization=gridpp) for the pilots due to the multi-VO support (GRIDPP-9)
  • For rollout, would like to set up a JIRA component for each site (several exist already.) Will be mailing sites to get the site contacts involved before we add each site.

Tuesday 1 Dec

  • Vac (0.20pre) now provides contexualization, metadata (EC2/OpenStack), and Machine/Job Features via HTTP rather than ISO image and NFS. (GRIDPP-27). This should allow VMs expecting to be run by OpenStack to work on Vac without modification.
  • CernVM image signature checking, and APEL Sync record generation (GRIDPP-10) are now also in Vac 0.20pre.

Tuesday 24 Nov

  • Cloud Init demonstrated with modified version of (old) GridPP DIRAC VMs
  • Cloud Init support in Vac 0.20pre (GRIDPP-27)
  • Progress on VMs for new GridPP DIRAC service: multi-VO config currently preventing matching. (GRIDPP-9)

Tuesday 17 Nov

  • New depo.gridpp.ac.uk service for uploading files to via HTTPS
  • ATLAS VMs now upload log files to depo.gridpp.ac.uk for debugging (GRIDPP-24)

Monday 9 Nov

  • Vac 0.19.0 released last week: new VacQuery protocol
  • Prototyping Vcycle(/Vac) GLUE2 publishing: talk at WLCG InfoSys Evolution TF on Thursday
  • Please now use https://repo.gridpp.ac.uk/vacproject/ URLs for user_data and cernvm3.iso files (in preparation for move to new www.gridpp.ac.uk website.)
  • See Laurence's GDB talk "Helix Nebula Price Enquiry Results" for more use of Vcycle

Tuesday 3 Nov

  • LHCb prototype of GOCDB pointers to resource BDII done
  • T2C tests at Oxford ongoing


Accounting - UK Grid Metrics HEPSPEC06 Atlas Dashboard HS06

Tuesday 24th November

  • Slight delay for Sheffield.

Tuesday 3rd November

  • APEL delay (normal state) Lancaster and Sheffield.

Tuesday 20th October The WLCG MB decided to create a Benchmarking Task force led by Helge Meinhard see talk

Documentation - KeyDocs

Tuesday 1st December

  • Sixt and hone have been removed from the GridPP list.

Tuesday 24th November Steve J: problems with voms server at fnal, voms.fnal.gov, resolved. Approved VOs document updated with newest records for those VOs affected, CDF, DZERO, LSST. Also, note changes to CA_DN for PLANCK and CDF. https://www.gridpp.ac.uk/wiki/GridPP_approved_VOs

Friday 6th Nov, 2015

SteveJ: Advice to admins published about a common GSS error, globus_gsi_callback_module: Could not verify credential etc.

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

Tuesday 20th October, 2015

Approved VOs document updated with temporary section for LZ

Tuesday 29nd September Steve J: problems with voms server at fnal, voms.fnal.gov, have been detected; I will resolve them soon and may issue an update to Approved VOs, alerting sites with TB_SUPPORT should that occur. Approved VOs potentially affected are CDF, DZERO, LSST. Please do not act act yet.


General note

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

Interoperation - EGI ops agendas

Tuesday 15th December

There was a meeting yesterday, agenda is here: https://wiki.egi.eu/wiki/Agenda-14-12-2015

  • UMD Preview repository
  • UMD releases
  • Decommissioning of dCache 2.6
  • Decommissioning of SL5
    • SL5 services must be decommissioned by end of April 2016; broadcast at December, probes will be warning since February 2016 to start helping with decommissioning
  • APEL on SL5
  • WMS Usage
  • New CE/batch system accounting integration
  • Raised a question on clarifying the future of the VO Nagios in a centralised ARGO world


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 30th November

  • A quiet week, mostly just cloing alarms for not-in-production services. There are no open ROD tickets.

Monday 23rd November

  • Still seeing the alarms for systems at QMUL and RHUL that are not in production.
  • Closed some tickets for the rolling availabilities. Confused by the rolling availability plots. It seems the "rolling average" period was cut back to 20 days.

Tuesday 3rd November

  • The long-standing UCL availability alarm went green yesterday on 29th October. We are not sure why!
  • Quite a lot of activity on the dashboard this week, but only one or two new tickets.
  • Tickets: Five for availability / reliability: Sussex, Sheffield, Liverpool, Lancaster and UCL. Two for GLUE2 validation: Liverpool and QMUL. One for the CEs at QMU

Tuesday 20th October

Lots of bits here and there, but no big pattern. Tickets about CE and storage problems open at several sites. QMUL notable as going on for a while, probably with some kind of configuration problem they're not identifying.


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

Tuesday 15th December

  • No updates applicable for the UK
  • News from elsewhere: EGI CSIRT activity is now suspending multiple sites following CVE-2015-7181/2/3, largely due to simple failure to acknowledge the notification, which was all that was required.

Tuesday 8th December

  • EGI SVG Advisory 'Low' RISK - OpenSSL announcement on 3rd December.
  • Keep OSes up-to-date.

Tuesday 2nd December

Tuesday 24th November

  • Call on NGIs to participate in "Security Threat Risk Assessment - with Cloud Focus" work.
  • Check Pakiti for CVE-2015-7183 issues.

Tuesday 17th November

  • Advisory-SVG-2015-CVE-2015-7183 issued 06/11/2015: a few UK sites show as unpatched in EGI monitoring. WNs, as tested by the monitoring, may be less vulnerable than affected middleware services but they could be taken as an indication of general site readiness and sites are encouraged to check their status.

The EGI security dashboard.


Services - PerfSonar 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 8th December

  • Given the recent network issues and role of GridPP DIRAC, there are plans to have a slave DNS for gridpp.ac.uk at Imperial and hopefully the T1 too. Andrew will seek an update to the whois records and name servers once the other host sites are confirmed.
  • Looking at the network problems this week will be of interest. Duncan supplied this link and Ewan the one for the dual stack instance.

Tuesday 6th October

Tuesday 14th July

  • GridPP35 in September will have a part focus on networking and IPv6. This will include a review of where sites are with their deployment. Please try to firm up dates for your IPv6 availability between now and September. Please update the GridPP IPv6 status table.


Tickets

Monday 14th December 2015, 13.00 GMT
37 Open Tickets to end the year on.

Separating out the atlas storage consistency check tickets, I don't think there's going to me much more done in 2015 on them from either end. Sites with tickets still open for this are: Sussex, Oxford, Birmingham, Glasgow, Sheffield, RHUL, QM, Imperial, Brunel and the Tier 1 (who are almost done). If you're not almost done then you'll want to On Hold the ticket (and knock the urgency down to "less urgent" whilst you're at it).

Onto the other tickets, site by site:

SUSSEX
118337 (14/12)
Some accidental not-reading-downtime-noifications by Sno+, I suspect this ticket can be closed. In progress (14/12)

117739 (20/11)
PRODDISK deletion at Sussex - as they're in the middle of a SE move this will want to be on hold! Unless you're just not moving the proddisk directories? In progress (25/11)

118289 (10/12)
Pilots roles at Sussex- the ticket only just rolled into the site but if you can you might want to roll this into your downtime if it's going smoothly (and if you're having a smooth downtime, can you tell me what that's like?). Assigned (14/12)

RALPP
117711 (19/11)
Pheno Pilot Role ticket. At last word the new groups "just" needed to be configured on the middleware. Any joy? In progress (19/11)

117740 (20/11)
The deletion of proddisk at RALPP has snowballed into a good clearout of the atlas area. Chris asked this morning if atlasdatadisk/step09 can be deleted too, and if Brian's had a chance to look at the filelists he provided. Waiting for reply (14/12)

BRISTOL
118311 (12/12)
CMS SAM3 tests were failing at Bristol. A tweak of their site configs solved the problem and they're both green (although Lukasz was surprised that they were throwing SRM warnings at a site with no SRM). In progress (looks like it can be closed) (14/12)

118299 (10/12)
CMS transfer timeouts on job stageout - has this one snuck past the Bristol ticket-sentries? Assigned (11/12)

BIRMINGHAM
118155 (4/12)
Biomed have used up too much space on Birmingham's SE - but need help finding the dark data and asked for a dump. We could co-opt the atlas tools to make a dump for them? No news for a while. In progress (4/12)

GLASGOW
117706 (19/11)
Pheno pilots at Glasgow -Gareth has updated that they're having a overhaul of identity management at the site and will roll supporting the "Dirac VOs" into that work. On Hold (2/12)

118052 (30/11)
An atlas ticket about http support at Glasgow's endpoints. Sam noticed that this is likely a problem with some older dmlite versions on some pools, and is looking into a non-nuclear solution. In progress (1/12)

ECDF
95303 (1/7/13)
Tarball glexec ticket-as usual see the Lancaster update as this is my bad. On hold (18/5)

SHEFFIELD
114460 (18/6)
Pilots roles at Sheffield - Elena is trying to roll this out this week. Any Joy? In progress (1/12)

LANCASTER
118305 (11/12)
Low availability ticket for Lancaster after the events of last week. We're just waiting for the numbers to pick up now. On hold (11/12)

95299 (1/7/13)
Tarball glexec. We have a new, updated build process but have had no time to test the end product - we have to build lcas/lcmaps alongside glexec to cope with different paths. On Hold (2/11)

RHUL
118236 (9/12)
Atlas "no route to host" errors due to a downed pool node - I think Govind has happy news about that disk pool so hopefully this ticket can be closed. In progress (11/12) Update -solved, looks like a transient problem.

118295 (11/12)
SE not working for Pheno when they use lcg-tools, although they can get stuff with uberftp. Odd. Assigned (11/12)

QMUL
117723 (19/11)
Pilots roles for pheno - these were rolled out but weren't behaving (normal pheno jobs are fine). Any joy? In progress (23/11)

IMPERIAL
118162 (4/12)
LFC at Imperial not working right for users from NA62, which coincided with the attack on JANET. DAMMIT. The LFC restarted and Janusz ran a test which worked fine the day after. No word back from the user. In progress (9/12)

BRUNEL
117710 (19/11)
Pheno pilots at Brunel - Raul was hoping to get to this. Any luck? On hold (22/11)

100IT
116358 (22/11)
Included for completeness - a "missing image" ticket for 100IT. There was a recent update so there's some progress here. In progress (7/12)

THE TIER 1
116866 (12/10)
Snoplus (and other) pilot roles at the Tier 1. Daniela is throwing dirac test jobs at the Tier 1 to try to debug the issues. A lot of effort going into the issue. In progress (9/12)

118044 (30/11)
Atlas glexec hammerclouds failing at the Tier 1. It was noted that the failures occurred about the time of a config change, but the glexec-guru was unavailable. In progress (7/12)

117683 (18/11)
Castor not publishing glue2. Stephen Burke has offered a hand with the task. In progress (19/11)

118338 (14/12)
Janusz, with his mice (mouse?) hat on, cannot copy data from the Imperial SE to Castor via the FTS, but can do uploads/downloads using the regular tools. Early days of the issue. In progress (14/12)

116864 (12/10)
CMS AAA xroot test failures. The ticket submitter replied to some queries with quite a lot of information that hopefully helps get to the bottom of this. In progress (20/11)

118130 (3/12)
Technically a ticket to the whole atlas UK cloud I think - atlas transfers throughout the UK has a large number of errors. Brian has been investigating and asked the FTS experts to help. Waiting for Reply (11/12)

Update: 118345 (14/12)
Jeremy has asked the Tier 1 about how well utilised their WMS currently is - Catalin has been put forward to comment.

Tools - MyEGI Nagios

Monday 30th November

  • The SAM/ARGO team has created a document describing Availability reliability calculation in ARGO tool.


Tuesday 6 Oct 2015

Moved Gridppnagios instance back to Oxford from Lancaster. It was kind of double whammy as both sites went down together. Fortunately Oxford site was partially working so we managed to start SAM Nagios at Oxford. Sam tests were unavailable for few hours but no affect on egi availibilty/reliability. Sites can have a look at https://mon.egi.eu/myegi/ss/ for a/r status.

Tuesday 29 Sep 2015

Following an air-conditioning problem at machine room in Oxford Tier-2 site on 26 September, gridppnagios(OX) was shut down and gridppnagios(Lancs) became active instance. Oxford site is in downtime until 1st Oct and it may be extended depending on the situation. VO-Nagios was also unavailable for two days but we have started it yesterday as it is running on a VM. VO-nagios is using oxford SE for replication test so it is failing those tests. I am looking to change to some other SE.

VOs - GridPP VOMS VO IDs Approved VO table

Tuesday 19th May

  • There is a current priority for enabling/supporting our joining communities.

Tuesday 5th May

  • We have a number of VOs to be removed. Dedicated follow-up meeting proposed.

Tuesday 28th April

  • For SNOPLUS.SNOLAB.CA, the port numbers for voms02.gridpp.ac.uk and voms03.gridpp.ac.uk have both been updated from 15003 to 15503.

Tuesday 31st March

  • LIGO are in need of additional support for debugging some tests.
  • LSST now enabled on 3 sites. No 'own' CVMFS yet.
Site Updates

Tuesday 24th February

  • Next review of status today.

Tuesday 27th January

  • Squids not in GOCDB for: UCL; ECDF; Birmingham; Durham; RHUL; IC; Sussex; Lancaster
  • Squids in GOCDB for: EFDA-JET; Manchester; Liverpool; Cambridge; Sheffield; Bristol; Brunel; QMUL; T1; Oxford; Glasgow; RALPPD.

Tuesday 2nd December

  • Multicore status. Queues available (63%)
    • YES: RAL T1; Brunel; Imperial; QMUL; Lancaster; Liverpool; Manchester; Glasgow; Cambridge; Oxford; RALPP; Sussex (12)
    • NO: RHUL (testing); UCL; Sheffield (testing); Durham; ECDF (testing); Birmingham; Bristol (7)
  • According to our table for cloud/VMs (26%)
    • YES: RAL T1; Brunel; Imperial; Manchester; Oxford (5)
    • NO: QMUL; RHUL; UCL; Lancaster; Liverpool; Sheffield; Durham; ECDF; Glasgow; Birmingham; Bristol; Cambridge; RALPP; Sussex (14)
  • GridPP DIRAC jobs successful (58%)
    • YES: Bristol; Glasgow; Lancaster; Liverpool; Manchester; Oxford; Sheffield; Brunel; IC; QMUL; RHUL (11)
    • NO: Cambridge; Durham; RALPP; RAL T1 (4) + ECDF; Sussex; UCL; Birmingham (4)
  • IPv6 status
    • Allocation - 42%
    • YES: RAL T1; Brunel; IC; QMUL; Manchester; Sheffield; Cambridge; Oxford (8)
    • NO: RHUL; UCL; Lancaster; Liverpool; Durham; ECDF; Glasgow; Birmingham; Bristol; RALPP; Sussex
  • Dual stack nodes - 21%
    • YES: Brunel; IC; QMUL; Oxford (4)
    • NO: RHUL; UCL; Lancaster; Glasgow; Liverpool; Manchester; Sheffield; Durham; ECDF; Birmingham; Bristol; Cambridge; RALPP; Sussex, RAL T1 (15)


Tuesday 21st October

  • High loads seen in xroot by several sites: Liverpool and RALT1... and also Bristol (see Luke's TB-S email on 16/10 for questions about changes to help).

Tuesday 9th September

  • Intel announced the new generation of Xeon based on Haswell.



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