Hi
on lcb problem :
Apparently it was discussed on WLCG meeting
on Friday
LHCb reports -
• General: some sites are reporting wrong information on their queues in the BDII (queue length with 999999 or SI00 reported as 0). LHCb will submit GGUS tickets to all of them asking for a fix of the information. This concerns mostly Tier2s. List of LHCb queues characteristics.
on Monday:
LHCb reports -
• General: some sites have set a wall clock time limit == CPU time limit. This makes any matching using CPU work requirement useless as the wall clock limit will always fire first (unless the job has an efficiency > 1). Sites don't seem to understand the issue... Not a major problem in most cases as our job efficiency is good, but it could be degraded by events outside our control (machine too heavily loaded or too high overcommitment of the machine (slots/cores).
I disagree with the statement that we can't have wall clock time limit == CPU time limit. However, I set manually the numbers they asked for as I do not have time ATM to spend on the discussions.
Elena
On 26 Nov 2012, at 15:41, Matt Doidge wrote:
> Heya all,
> Here's this week's ticket bit.
> Cheers,
> Matt
>
> 35 Open UK tickets today.
>
> I had to set a few tickets "In Progress" this week. Remember if a VO reopens a ticket and you get back onto refix the issue the ticket should be re-in progressed.
>
> A few trends this week. t2k are on a software update spree, I know at Lancaster there were problems due to us not having installed the packages listed on the CiC portal. Biomed continue to give the silent treatment on their tickets, and there seem to be a few lhcb tickets about the place.
>
> Unsupported Glite Software:
> BRISTOL: https://ggus.eu/ws/ticket_info.php?ticket=87472 (17/10) In Progress (23/11)
> Cream CE & Workernodes are EMI2. This only leaves the the APEL box and I believe (the hard one) the Bristol Storm box. Good stuff though.
> EDINBURGH: https://ggus.eu/ws/ticket_info.php?ticket=87171 (10/10) In Progress (21/11)
> "pretty much done with our EMI deployment". Great news, once you're done you can close this ticket, it's not being done for us anymore.
> MANCHESTER: https://ggus.eu/ws/ticket_info.php?ticket=87467 (17/10) On hold (5/11)
> Manchester have the final push of upgrades planned for this week. Good luck!
> UCL: https://ggus.eu/ws/ticket_info.php?ticket=87468 (17/10) On hold (1/11)
> Things are quiet on the UCL front.
>
> QMUL
> https://ggus.eu/ws/ticket_info.php?ticket=88822 (23/11)
> This lhcb ticket concerning 99999 values for Max CPU time information publishing might have slipped under the QM radar. Assigned (23/11)
>
> Similar problem at Lancaster: https://ggus.eu/ws/ticket_info.php?ticket=88772
> Sheffield solved their ticket: https://ggus.eu/ws/ticket_info.php?ticket=88781
>
> https://ggus.eu/ws/ticket_info.php?ticket=86306 (22/9)
> Hard to Kill lhcb pilots on the QMUL cream. LHCB sent out another "to purge" list on the 21st. The corresponding ticket from Chris to the CREAM developers (https://ggus.eu/tech/ticket_show.php?ticket=87891) has been cheekily closed because Chris couldn't provide log snippets right away (due to Cream's overactive log rotation). In progress (21/11)
>
> BTW Sheffield has seen this problem too in the last week: https://ggus.eu/ws/ticket_info.php?ticket=88719
>
> LIVERPOOL
> https://ggus.eu/ws/ticket_info.php?ticket=88761 (22/11)
> Technically this is a "from the UK ticket". LHCB jobs have been swamping the Liverpool WAN. Cutting back the number of running lhcb jobs has alleviated the problem somewhat. Is the cause of the sudden lhcb bandwidth gobbling still thought to be simply a surge of lhcb work whilst atlas were quiet? In progress (23/11)
>
> LANCASTER
> https://ggus.eu/ws/ticket_info.php?ticket=88628 (20/11)
> t2k were having problems installing their software on one of our clusters, largely due to not taking t2k into account in our latest node image. However the latest problem is a crash when building ROOT which has us stumped. Apparently Oxford see this too, so expect a poke from me soon in an attempt to scavenge a solution. In progress (26/11)
>
> RHUL
> https://ggus.eu/ws/ticket_info.php?ticket=88417 (11/11)
> Atlas squid problems - any luck getting the firewall opened for your new squid config? In Progress (20/11)
__________________________________________________
Dr Elena Korolkova
Email: [log in to unmask]
Tel.: +44 (0)114 2223553
Fax: +44 (0)114 2223555
Department of Physics and Astronomy
University of Sheffield
Sheffield, S3 7RH, United Kingdom
|