Tier1 Operations Report 2015-03-04

From GridPP Wiki
Jump to: navigation, search

RAL Tier1 Operations Report for 4th March 2015

Review of Issues during the week 25th February to 4th March 2015.
  • On Thursday (26th Feb) there was a problem with our Argus server that stopped batch job submission starting for an hour or so.
  • A problem was uncovered by LHCb when a particular file could not be read from Castor. A manual fixup enabled access to the file again. However, investigations showed a problem when the file was created a few weeks ago - which was the second occurrence of a rare bug in Castor.
  • Five files have been declared lost to CMS. These were all picked up by the checksum checker at various times in the last weeks.
Resolved Disk Server Issues
  • GDSS757 (CMSDisk - D1T0) failed again early on Friday morning (27th Feb). As this server has failed multiple times it has been put back readonly and is being drained.
Current operational status and issues
  • We are running with a single router connecting the Tier1 network to the site network, rather than a resilient pair.
Ongoing Disk Server Issues
  • None
Notable Changes made this last week.
  • Removed DNs from FTS3 monitoring messages
  • Cap on maximum number of ALICE batch jobs raised from 3500 to 6000.
  • A new Stratum-1 system for the cern.ch domain set-up.
Declared in the GOC DB


Service Scheduled? Outage/At Risk Start End Duration Reason
Whole site SCHEDULED WARNING 11/03/2015 10:30 11/03/2015 11:00 30 minutes Warning on site access during firmware updates on pair of network switches.
lcgfts3.gridpp.rl.ac.uk, SCHEDULED WARNING 05/03/2015 10:00 05/03/2015 12:00 2 hours Warning on FTS3 service during upgrade to version v3.2.32
Whole site UNSCHEDULED WARNING 05/03/2015 07:45 05/03/2015 08:30 45 minutes Warning following installation of a replacement network router.


Advanced warning for other interventions
The following items are being discussed and are still to be formally scheduled and announced.
  • Investigate problems on the primary Tier1 router. Discussions with the vendor are ongoing. (Swap with replacement unit planned for tomorrow morning, 5th March).
  • Update firmware on resilient pair of switches in the Tier1 network - planned for Wednesday 11th March.
  • FTS3 update planned for Thursday 5th March.

Listing by category:

  • Databases:
    • Switch LFC/3D to new Database Infrastructure.
    • Update to Oracle 11.2.0.4
  • Castor:
    • Update SRMs to new version (includes updating to SL6).
    • Fix discrepancies were found in some of the Castor database tables and columns. (The issue has no operational impact.)
    • Update Castor to 2.1-14-latest.
  • Networking:
    • Resolve problems with primary Tier1 Router
    • Make routing changes to allow the removal of the UKLight Router.
    • Enable the RIP protocol for updating routing tables on the Tier1 routers. (Install patch to Router software).
  • Fabric
    • Firmware updates on remaining EMC disk arrays (Castor, FTS/LFC)
Entries in GOC DB starting since the last report.
  • None
Open GGUS Tickets (Snapshot during morning of meeting)
GGUS ID Level Urgency State Creation Last Update VO Subject
111856 Green Less Urgent Waiting Reply 2015-02-19 2015-03-03 LHCb Stalled LHCb jobs over night
111699 Green Less Urgent In Progress 2015-02-10 2015-02-27 Atlas gLExec hammercloud jobs keep failing at RAL-LCG2 & RALPP
109694 Red Urgent In Progress 2014-11-03 2015-03-04 SNO+ gfal-copy failing for files at RAL
108944 Red Urgent Waiting Reply 2014-10-01 2015-03-03 CMS AAA access test failing at T1_UK_RAL
Availability Report

Key: Atlas HC = Atlas HammerCloud (Queue ANALY_RAL_SL6, Template 508); CMS HC = CMS HammerCloud

Day OPS Alice Atlas CMS LHCb Atlas HC CMS HC Comment
25/02/15 100 100 100 100 100 100 99
26/02/15 100 100 100 100 100 100 99
27/02/15 100 100 100 100 100 100 95
28/02/15 100 100 100 100 100 100 95
01/03/15 100 100 100 100 100 0 98
02/03/15 100 100 100 100 100 100 97
03/03/15 100 100 100 100 100 98 100