Here is the Atlas report for this week.
Site problems:
=========
RAL:
* Analysis jobs filled space too quickly in castor and caused other transfers to fail. Also free space was
not well balanced on the disk servers. 7 disk servers have been put in read-only mode to allow draining to
emptier servers. Jobs have been throttled back to decrease the writing rate.
* One of the castor servers wasn't responding correctly to castor calls and outgoing transfers were affected.
This has been fixed.
* LFC packet loss on route into RAL, correlates with LFC timeouts: Networking people seem to have mostly fixed this,
i.e. there are still errors but they dropped to a scattered 1%. It seems there is also an improvement on the atlas
side of things and Elena has seen a drop in LFC errors.
UKI-SCOTGRID-GLASGOW:
* Networking problem over the weekend. Site put offline in production. Now solved.
UKI-SOUTHGRID-BHAM-HEP:
* Missing release: problem not with WNs as reported last week. They have two SW servers and are affected by usual
installation panda/bdii syncronisation problem.
UKI-SOUTHGRID-RALPP:
* Site is not receiveing pilots. EK contacted them via email, the problem is CMS jobs have swamped the cluster and
atlas jobs are not running. Since RALPP is also an atlas and lhcb site this shouldn't happen. A ticket was opened.
UKI-LT2-Brunel
* Space token UKI-LT2-BRUNEL_PRODDISK under 20% (Free:0.352 Total:2.199) since May 3 09:30
UKI-SOUTHGRID-OX-HEP
* Space token UKI-SOUTHGRID-OX-HEP_DATADISK under 20% (Free:9.785 Total:58.0) since Apr 29 18:24
UKI-SOUTHGRID-RALPP
* Space token UKI-SOUTHGRID-RALPP_HOTDISK under 20% (Free:0.15 Total:0.999) since Apr 29 18:24
FT Transfers:
=========
This is a category apart as they are functional tests and don't affect production but in the
new data distribution model they need to be followed up.
* Nothing to report.
Problems caused by Atlas:
=========================
* Nothing to report.
Generic problems
================
* Nothing to report.
Open Tickets
============
UKI-SOUTHGRID-BHAM-HEP:
https://gus.fzk.de/ws/ticket_info.php?ticket=69977 missing release in progress
UKI-SCOTGRID-GLASGOWL:
https://gus.fzk.de/ws/ticket_info.php?ticket=70131 tranfers failure in progres
RAL-LCG2:
https://gus.fzk.de/ws/ticket_info.php?ticket=68850 LFC on hold
https://gus.fzk.de/ws/ticket_info.php?ticket=69721 SRM in progress
UKI-NORTHGRID-MAN-HEP:
https://gus.fzk.de/ws/ticket_info.php?ticket=69336 networking/transfers on hold
|