RAL Tier1 weekly operations castor 15/09/2014

From GridPP Wiki
Jump to: navigation, search

Operations News

  • Plan to ensure PreProd represents production in terms of hardware generation are underway
  • Disk server redeployments continue (i.e. D1T0 reused in D0T1 etc) ... 5 servers in LHCb left
  • SL6 Headnode work progressing - hoping for rollout in Nov
  • xrootd security advisory with FAX component within xrootd

Operations Problems

  • a disk server issues at the weekend (filesystem errors) GDSS651, now back in production
  • A few Atlas SUM test failures throughout the week, may possibly be related to the below
  • xrootd memory leak seen on all dlf headnodes, affected Atlas lcgcdlf01 most which was swapping - need to investigate if a fix is already available, if not discuss at castor face to face
  • A number of duplicate entries in SRM userfile table (not the duplicate user issue) - discussed with Atlas, root cause still unknown
  • Break in connectivity Monday 8th, it appears that this did not affect castor internally in any way however if transfers were in process they would have failed and retried
  • Brian has found atlas scratch files on tape caches - concern about d0t1 decommissioning procedure, Brian to propose/update
  • There are 330 files on genTape that have zero physical size and zero NS size. There are also 10 files with no NS size but a physical non zero file - these need to be discussed with the VOs
  • fdsdss36 suffered an odd network config change, being corrected


Blocking Issues

  • grid ftp bug in SL6 - stops any globus copy if a client is using a particular library. This is a show stopper for SL6 on disk server.


Planned, Scheduled and Cancelled Interventions

  • Juan to patch castor dbs beginning of Nov PSU patches – standard change
  • 2.1.14-14 testing in preprod when Brian releases it
  • A Tier 1 Database cleanup is planned so as to eliminate a number of excess tables and other entities left over from previous CASTOR versions. This will be change-controlled in the near future.


Advanced Planning

Tasks

  • Possible future upgrade to CASTOR 2.1.14-15.
  • Switch from admin machines: lcgccvm02 to lcgcadm05
  • New VM configured to run against the standby CASTOR database will be created as a front-end for dark data etc queries.
  • Replace DLF with Elastic Search
  • Correct partitioning alignment issue (3rd CASTOR partition) on new castor disk servers


Interventions

  • None

Staffing

  • Castor on Call person
    • Rob


  • Staff absence/out of the office:
    • Chris – Out Mon-Wed
    • Brian – Out Tues - Thurs

Early warning – CASTOR Face to Face at CERN 22/23 Sept

Retrieved from "https://www.gridpp.ac.uk/w/index.php?title=RAL_Tier1_weekly_operations_castor_15/09/2014&oldid=5821"