Hello all,
Minutes from today's meeting attached, indico still won't let me upload
them I'm afraid.
No actions per se, rather a reminder to sites that haven't responded to
Pete Clarke yet to please do so.
With londongrid being decommissioned we might want to review the other
regional VOs - perhaps a topic for next week's meeting?
Thanks all,
Matt
On 25/07/17 10:23, Peter Gronbech wrote:
> I am here and will chair.
> Pete
>
Ops meeting 25/7/17
Chair: Pete G
Minutes: Matt
Attending: Adam Boutcher, Andrew McNab, Andrew Washbrook, Brian,Elena, Gareth Roy, Govind, Ian Loader, Ian Neilson, John Bland, Leo, Oliver Smith, Peter Clarke, Robert Currie, Sam Skipsey, Steve Jones, Winnie.
Apologies: Jeremy C, Daniela
2 new tickets - 1 for QM (129684)- DDM failures.
Matt- their SE looks quite broken at the moment
Oxford - lost heartbeart errors for mcore jobs. Elena
Big panda wants to switch to https access, which doesn't work for people who don't have atlas membership. Alessandra has opened JIRA ticket (see email to TB-Support). Please try to access the site, and reply to the email (or ticket).
Please- check bigpanda to see if you have access or not.
LHCB
Nothing in the elog, RAL failures due to a downtime so an artificial problem, everything seems okay though.
-This was due to the network outage this morning.
CMS
No one present.
Other VOs
Only update - lsst from last Tuesday. Ongoing action to expand panda submission to other sites. In due course other sites will be contacted.
GridPP Dirac Sam Status
RHUL VAC site added last week.
ECDF arc not had an update for a few months (11th May).
Andy W - is this due to being banned due to SL7?
Andy M -it could be, curl is used in the jobs. Are you running any gridpp jobs?
Andy W - will check.
Bulliten Board
new VAPOR release (2.3) ready for testing. Report issues via GGUS 129391.
http://operations-portal.egi.eu/vapor_dev
https://ggus.eu/index.php?mode=ticket_info&ticket_id=129391
Site are encouraged to take a look
Tier 1
A reminder that there is a weekly Tier-1 experiment liaison meeting. Notes from the last meeting here
http://www.gridpp.ac.uk/wiki/RAL_Tier1_Experiments_Liaison_Meeting
https://www.gridpp.ac.uk/wiki/Tier1_Operations_Report_2017-07-19
The SOAP interface to the FTS3 service has been stopped on Monday 17th July. This will allow us to update FTS3 to the latest version.
Castor unavailable during morning of 25th July for OS patching.
EGI have announced withdrawal of support for the WMS at the end of 2017 (as announced elsewhere).
A test Centos7 queue has been enabled on one of our CEs.
A test gateway to Echo ceph-test-gw691.gridpp.rl.ac.uk) has been made dual stack and test transfers have been shown to work over IPv6 to/from CERN.
Data is flowing from the Dirac Leicester site - reaching a peak transfer rate of around 1Gbit/sec.
-
Storage and Data Management
Discussion of potential topics for data pre-GDB in September, without being able to commit yet
Switching to fortnighly calls over the rest of July + August
Tier 2 Evolution
RHUL Vac site in production for LHCb and GridPP with Vac-in-a-Box.
Some problems with Vac-in-a-Box installation being worked on (race condition in setting up the firewall)
GridPP DIRAC and ATLAS jobs running at Cambridge Vac site.
Problem with multiprocessor accounting and APEL identified but seems to only affect Glasgow significantly. Fixed in Vac 02.00.01 release.
Oxford should pick this up automatically?
Andrew M will follow up with APEL people, it will likely be more a headache for them then the sites.
Documentation
London grid (vo.londongrid.ac.uk) VO removed from Approved VOs.
https://www.gridpp.ac.uk/wiki/GridPP_approved_VOs
Londongrid decommissioned, are the other regional VOs getting much use?
On-duty
No news, Andy M was ill so someone should have covered him yesterday?
Security
-Nothing to report.
It's Ian's last meeting today. Thanks Ian! All the best.
Tickets
Any one from QMUL around? The SE seems very unhappy.
Other Topics
Pete C - Please can these sites send me their entries for the network forward look document (as per the email):
Imperial, QMUL, UCl, Manchester, Sheffield, Sussex
Aplogies if you've already sent me information.
Actions Review
O-170711-01 Robin may take over as GalDyn contact. Matt will check with him next week.
O-170711-02 Sussex look at why gridpp dirac jobs aren't being scheduled. Leo will look at it.
Other actions skipped due to people not in attendence.
No other business.
Chat Window.
Matt Doidge: (25/07/2017 11:03)
I can do it
The minutes that is. Hmm, don't think my mic is working though
Elena Korolkova: (11:09 AM)
bigpanda [1] will not be open anymore in the future and will require CERN SSO. There is a big banner on it saying so and giving some directions to fix possible problems. However there might be problems if you are not an ATLAS member. Glasgow already reported they have. I've opened a JIRA ticket [2] for this. I encourage people to test and see if they have any problem in accessing the monitoring and let me know or update the ticket directly.
thanks
cheers
alessandra
[1] https://bigpanda.cern.ch
[2] https://its.cern.ch/jira/browse/ATLASPANDA-388
Peter Gronbech: (11:15 AM)
https://www.gridpp.ac.uk/gridpp-dirac-sam
Gareth Douglas Roy: (11:18 AM)
Fire alarm here got to leave the building
Brian Davies @ RAL-LCG2: (11:19 AM)
john will be joinming
Steve Jones: (11:29 AM)
Good luck Ian.
Brian Davies @ RAL-LCG2: (11:38 AM)
off
|