Monday 5th November 2018, 14.00 GMT
41 Open UK Tickets this month
SUSSEX
https://ggus.eu/?mode=ticket_info&ticket_id=138071 (2/11)
A fresh ticket from atlas about SRM problems. The lack of links in the ticket made it hard for Leo to debug, and he has asked for clarification. Waiting for reply (2/11)
BIRMINGHAM
https://ggus.eu/?mode=ticket_info&ticket_id=138026 (31/10)
A ticket concerning the alice VOBOX at Birmingham. It looks like the problem went away on its on and this ticket can be closed, but there appear that there will be other conversations to have about alice needs at Birmingham at a later date. In Progress (can be closed) (3/11)
https://ggus.eu/?mode=ticket_info&ticket_id=137801 (17/10)
A Birmingham related ticket rather then a ticket for the site, the tracking of the decommissioning of their DPM SE. I can't quite remember how things are properly done, but shouldn't this be put On Hold until the 28th November? In progress (22/10)
BRISTOL
https://ggus.eu/?mode=ticket_info&ticket_id=138041 (1/11)
A CMS ticket concerning failing transfers. Lukasz has traced the problem due to the files being on disk but not in the namespace, and emailed the dpm support list for help fixing this (if a fix is possible). In progress (5/11)
OXFORD
https://ggus.eu/?mode=ticket_info&ticket_id=137941 (25/10)
Sno+ had problems accessing data on the Oxford SE due to BDII issues. Kashif fixed the IPv6 routing problems that were the cause of these, and things are working once again. Another ticket that can be closed. In progress (30/10)
GLASGOW
https://ggus.eu/?mode=ticket_info&ticket_id=134689 (23/4)
Request to upgrade Perfsonar boxes to CentOS7. Gareth gave his plan and (good) reasons why they won't be able to do this just yet at Glasgow - getting v6 working comes first. On hold (30/10)
ECDF
https://ggus.eu/?mode=ticket_info&ticket_id=137985 (29/10)
Atlas deletion errors at Edinburgh. Andy is reckoning this is a consistency problem as the system tries to delete files that aren't there anymore, and has asked if it's lots of different file deletion attempts failing or the same few deletion attempts failing repeatedly. I used to have a dodgy bash script that could help with that (by working on the downloaded xml from the DDM pages), but I don't think it made it off of our old SE I'm afraid. Waiting for reply (1/11)
DURHAM
https://ggus.eu/?mode=ticket_info&ticket_id=134687 (23/4)
Request to upgrade Perfsonar to CentOS7. It was mentioned verbally that this has been postponed to be part of "CentOS 7 Big Push" early next year, could that be put into the ticket. Be aware that Perfsonar support on SL6 will end soon though (we're already in the "grace period"). In progress (26/9)
SHEFFIELD
https://ggus.eu/?mode=ticket_info&ticket_id=137732 (15/10)
One of the ROD availability tickets, waiting for the time to pass. There's been a good stretch of green 100%s in the argo monitoring, so things are looking good. On hold (15/10)
https://ggus.eu/?mode=ticket_info&ticket_id=138095 (5/11)
Another ROD ticket, this is for the "APEL-Pub" tests. Set In Progress (5/11)
MANCHESTER
https://ggus.eu/?mode=ticket_info&ticket_id=137112 (11/9)
An atlas ticket about Manchester's Space Token numbers being broken after trouble with a draining script moving data outside of the tokens. The process to move them back was expected to take weeks. How are things looking now? Tim provided some figures from rucio a few weeks back, but that picture might be out of date now. On Hold (16/10)
LANCASTER
https://ggus.eu/?mode=ticket_info&ticket_id=136635 (9/8)
A very long running low availability ticket caused by issues with Lancaster's SE. The recent problems were caused by the CertLifetime check not working for a while during the move to DOME was underway, returning an "Unknown" status. As every other aspect of our DPM worked in that time I've requested a recomputation, which may or may not be a bit cheeky. On Hold (5/11)
https://ggus.eu/?mode=ticket_info&ticket_id=137996 (30/10)
Another ROD ticket (sorry ROD shifters), this time failing a non-critical http test. The issue has been tracked to a problem in the DOME code and a fix will hopefully be out this month. Until then... On Hold (5/11)
QMUL
https://ggus.eu/?mode=ticket_info&ticket_id=134573 (17/4)
CMS request to install singularity, on hold until the QM move to CentOS7. CMS has re-poked the ticket, asking again for the site's plans. On Hold (31/10)
BRUNEL
https://ggus.eu/?mode=ticket_info&ticket_id=133956 (9/3)
CMS ticket regarding Brunel's xroot configs. Raul has done some work involving DOME in the background, perhaps some of that progress could be used to update the ticket? In progress (16/10)
TIER 1
https://ggus.eu/?mode=ticket_info&ticket_id=138033 (1/11)
Atlas singularity jobs failing at RAL, with some reference to similar issues for SKA. It's being looked at, and Tim has provided some extra observations. In progress (1/11)
https://ggus.eu/?mode=ticket_info&ticket_id=137650 (9/10)
CMS seeing low HC xroot success rates at RAL. Lots of back and forth on the ticket, I don't think a conclusion has been reached yet though. In progress (2/11)
https://ggus.eu/?mode=ticket_info&ticket_id=138077 (2/11)
CMS SAM tests failing at RAL. Things seem to have healed themselves, but John has asked some team members to check the logs. In progress (5/11)
https://ggus.eu/?mode=ticket_info&ticket_id=138103 (5/11)
CMS transfers failing - the cause looks to be a zero-sized "stub file" causing issues, and it's being investigated. In progress (5/11)
https://ggus.eu/?mode=ticket_info&ticket_id=138002 (29/10)
CMS problems with the FTS, with a lot of sites seeing "bad transfer quality". Investigations pointed to a IPv6 problem that has since been fixed. However Gareth couldn't see an endemic issue with the RAL FTS whilst looking through the plots, and has asked for clarification. Waiting for reply (5/11)
https://ggus.eu/?mode=ticket_info&ticket_id=137897 (23/10)
enmr.eu jobs have zero normalised CPU hours in the accounting portals. It seems to have been a problem with the data the site reported. Catalin has asked the VO if anything changed around the 16th of October, and to resubmit some more jobs so they can watch them. Waiting for reply (5/11)
https://ggus.eu/?mode=ticket_info&ticket_id=137752 (15/10)
A request to replicate the OSG cvmfs repositories on the EGI stratum 1s. These have been replicated to the RAL servers, so I don't know what the next steps are. In progress (2/11)
https://ggus.eu/?mode=ticket_info&ticket_id=136199 (18/7)
One of a few LHCB FTS tickets, it looks like the work here has progressed nicely so maybe this ticket can be closed? Or is it waiting on all LHCB FTS issues to be solved? In progress (1/11)
https://ggus.eu/?mode=ticket_info&ticket_id=137822 (18/10)
FTS servers seemingly in a bad state for LHCB. I think this is being worked on, but no news in this particular ticket. In progress (22/10)
https://ggus.eu/?mode=ticket_info&ticket_id=138028 (1/11)
LHCB noticing files cannot be staged from tape to disk. The issue is somewhat understood (there is a copy of the file on disk elsewhere), but it's unsure why the resulting disk-to-disk transfer fails so the ticket is being kept open. In progress (1/11)
https://ggus.eu/?mode=ticket_info&ticket_id=136701 (14/8)
LHCB ticket regarding the high background rate of failures putting job output into RAL. I don't think a conclusion has been reached, the last update has Chris from LHCB collecting some more stats and saying that LHCB hope to be using direct xroot connections in the not too distant future. In progress (17/10)
https://ggus.eu/?mode=ticket_info&ticket_id=137153 (12/9)
T2K having trouble with zero sized files in the LFC. The LFC devs have been contacted for help, but that was a few weeks back. Any news from them? In progress (10/10)
|