Tier1 Operations Report 2016-03-16

From GridPP Wiki
Jump to: navigation, search

RAL Tier1 Operations Report for 16th March 2016

Review of Issues during the week 9th to 16th March 2016.
  • Nothing particular to report. Security updates being rolled out.
  • We have uncovered a problem where draining of Castor disk servers is now going very slowly. We need to drain a few old servers to provide spares - this being part of the deployment plan and server lifecycle. Although this does not yet impact services we are having to investigate this.
Resolved Disk Server Issues
  • GDSS677 (CMSTape - D0T1) crashed in the morning of Tuesday 8th March. It was returned to service on Friday (11th Mar). Two disk drives were replaced.
  • GDSS698 (LHCbDst - D1T0) crashed in the morning of Friday 11th March. It was returned to service that evening although no cause for the crash was identified.
  • GDSS646 (LHCbUser - D1T0) crashed on Saturday 12th March. It was put back in service the following day (Sun) in read-only mode. An update of the RAID card firmware revealed some disk problems - which were fixed before the server was put back in read/write mode yesterday (Tuesday 15th)
Current operational status and issues
  • There is a problem seen by LHCb of a low but persistent rate of failure when copying the results of batch jobs to Castor. There is also a further problem that sometimes occurs when these (failed) writes are attempted to storage at other sites. A recent modification has improved, but not completed fixed this.
  • The intermittent, low-level, load-related packet loss seen over external connections is still being tracked. Likewise we have been working to understand some remaining low level of packet loss seen within a part of our Tier1 network.
Ongoing Disk Server Issues
  • GDSS620 (GenTape - D0T1) Reported a read-only file system yesterday (Tuesday) morning and was taken out of production. Two T2K files that were not yet copied to tape were corrupt and have been reported as lost. Investigations ongoing.
Notable Changes made since the last meeting.
  • The planned network intervention on Monday (14th) for a small network reconfiguration and firmware update in a pair of network switches was successful.
  • Patching for security updates ongoing.
Declared in the GOC DB
Service Scheduled? Outage/At Risk Start End Duration Reason
All Castor (All SRM endpoints) SCHEDULED OUTAGE 17/03/2016 09:30 17/03/2016 15:00 5 hours and 30 minutes Downtime on RAL Storage for security updates
Advanced warning for other interventions
The following items are being discussed and are still to be formally scheduled and announced.
  • At Risk on the whole Tier1 next Monday, 14th March, 11:00 - 16:00 while a network reconfiguration is made followed by the firmware update in a pair of network switches. (Note: The date of this work was updated after the meeting.)
  • The Castor 2.1.15 update is pending. Testing has shown a database related performance issue which is being followed up. We await successful resolution of that problem and completion of testing before scheduling.
  • Decommissioning of "GEN Scratch" storage in Castor.

Listing by category:

  • Databases:
    • Switch LFC/3D to new Database Infrastructure.
  • Castor:
    • Update SRMs to new version (includes updating to SL6).
    • Update to Castor version 2.1.15.
    • Migration of data from T10KC to T10KD tapes (affects Atlas & LHCb data).
  • Networking:
    • Replace the UKLight Router. Then upgrade the 'bypass' link to the RAL border routers to 2*10Gbit.
  • Fabric
    • Firmware updates on remaining EMC disk arrays (Castor, LFC)
  • Grid Services
    • A Load Balancer (HAProxy) will be used in front of the FTS service.
Entries in GOC DB starting since the last report.
Service Scheduled? Outage/At Risk Start End Duration Reason
Whole site SCHEDULED WARNING 14/03/2016 11:00 14/03/2016 16:00 5 hours Warning on Site during a network test and firmware upgrade on a pair of network switches.
Open GGUS Tickets (Snapshot during morning of meeting)
GGUS ID Level Urgency State Creation Last Update VO Subject
119841 Amber Less Urgent In Progress 2016-03-01 2016-03-08 LHCb HTTP support for lcgcadm04.gridpp.rl.ac.uk
117683 Green Less Urgent In Progress 2015-11-18 2016-02-17 CASTOR at RAL not publishing GLUE 2
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
09/03/16 100 100 100 100 100 100 100
10/03/16 100 100 100 100 100 100 100
11/03/16 100 100 100 100 100 94 100
12/03/16 100 100 98 100 100 97 100 Single SRM test failure on PUT.
13/03/16 100 100 100 100 100 100 100
14/03/16 100 100 98 34 100 100 100 Atlas: Single SRM test failure; CMS: Problem affecting CE tests across sites. However, RAL didn't subsequently run test. Will request correction.
15/03/16 100 100 100 40 100 100 100 CMS: Continuation of above.