Difference between revisions of "RAL Tier1 weekly operations castor 27/04/2015"
From GridPP Wiki
(Created page with "[https://www.gridpp.ac.uk/wiki/RAL_Tier1_weekly_operations_castor List of CASTOR meetings] == Operations News == * Draining - Atlas still draining * SL6 Disk and Tape server ...") |
|||
Line 21: | Line 21: | ||
* storageD retrieval from castor problems - investigation ongoing | * storageD retrieval from castor problems - investigation ongoing | ||
* CMS heavy load & high job failure – hot spotting 3 servers, files spread to many servers which then also became loaded. CMS moved away from RAL for these files. Need to discuss this issue in more detail. | * CMS heavy load & high job failure – hot spotting 3 servers, files spread to many servers which then also became loaded. CMS moved away from RAL for these files. Need to discuss this issue in more detail. | ||
− | * castor functional test on lcgccvm02 causing problems - Gareth | + | * castor functional test on lcgccvm02 causing problems - Gareth reviewingr. |
− | + | ||
− | + | ||
Latest revision as of 09:25, 1 May 2015
Contents
Operations News
- Draining - Atlas still draining
- SL6 Disk and Tape server config is production ready - change control in for tape servers
- patch preprod oracle DB 11.2.04 - change control in
- testing CASTOR rebalancer (new version in 2.1.14-15)
Operations Problems
- GDSS757 cmsDisk / 763 atlasStripInput - new motherboards, fabric acceptance test finishes 23/4/15
- Elastic tape data corruption - CASTOR team assisting with extraction of data for review
- Seeing a high volume of SRM DB file dup errors - discuss with Shaun this week
- A higher number of checksum errors (Alice) investigation ongoing
- CMS high load / possibly still suffering from hot files - these have been duplicated on other disk servers
- OLDER>>>>
- Deadlocks on atlas stager DB (post 2.1.14-15) - not serious but investigation ongoing
- Issue with missing files in LHCb – race condition reported to CERN
- Atlas are putting files (sonar.test files) into un-routable paths - this looks like an issue with the space token used. Brian / Rob actions below.
- There is a problem with latest gfal libraries - not a new problem but Atlas are starting to exercise functionality and identifying these issues
- storageD retrieval from castor problems - investigation ongoing
- CMS heavy load & high job failure – hot spotting 3 servers, files spread to many servers which then also became loaded. CMS moved away from RAL for these files. Need to discuss this issue in more detail.
- castor functional test on lcgccvm02 causing problems - Gareth reviewingr.
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
Advanced Planning
Tasks
- SL6 tape server
- Bruno working on SL6 disk servers
- DB team need to plan some work which will result in the DBs being under load for approx 1h
- Provide new VM? to provide castor client functionality to query the backup DBs
- Plan to ensure PreProd represents production in terms of hardware generation are underway
- Switch from admin machines: lcgccvm02 to lcgcadm05
- Correct partitioning alignment issue (3rd CASTOR partition) on new castor disk servers
Interventions
Actions
- Rob - revisit deployment plans (esp people in office towards the end of March)
- Brian - On Wed, tell Tim if he can start repacking mctape
- Rob - change checksum validator so it maintains case when displaying
- Rob - talk with Shaun to see if its possible to reject anything that does not have a spacetoken
- Brian - to discuss unroutable files / spacetokens at at DDM meeting on Tuesday
- Rob to pick up DB cleanup change control
- Bruno to document processes to control services previously controlled by puppet
- Gareth to arrange meeting castor/fab/production to discuss the decommissioning procedures
- Chris/Rob to arrange a meeting to discuss CMS performance/xroot issues (is performance appropriate, if not plan to resolve) - inc. Shaun, Rob, Brian, Gareth
- Gareth to investigate providing checks for /etc/noquatto on production nodes & checks for fetch-crl
Staffing
- Castor on Call person
- Matt
- Staff absence/out of the office:
- Brian – out Wed/Thurs for GridPP