Last modified 3 years ago Last modified on 11/08/16 17:05:03

dCache Tier I meeting November 8, 2016

[part of a series of meetings]

Present, Tigran), IN2P3(), Sara(), Triumf(), BNL(), NDGF(Ulf), PIC(Marc), KIT(), Fermi(), CERN()


(see box on the other side)

Site reports


Ulf reported that NDGF upgraded to a custom build from Gerd today. The changes were a couple of patches that fix problems observed during the last upgrade. NDGF is now running two SRM instances in production, so fully redundant. The deployment was this morning (~08:00 CEST), so has been in production for ~6 hours. No problems have been reported and the active testing hasn't shown any problems either -- previously the Nagios probe reported problems within an hour.

There has been one problem with SLAC; this is likely a issue with a firewall.

There are a lot of logged message with the info service "triggering too fast" message.


Marc reported that everything is OK at PIC, except for a problem that might not be dCache related.

They are experiencing problems with ATLAS name-to-name translation plugin and FAX.

PIC has started to fail tests, but this is only for Rucio transfers -- other transfers seem to be fine.

From a manual inspection, it seems that the transfers are incorrectly translated, but Marc is still investigating.

Paul pointed out there was a dCache (RT) support ticket, triggered by a GGUS ticket that was similar.

Marc mentioned that this was UAM, a Spanish Tier-2 centre that they (PIC) provide some assistance.

The GGUS ticket has been assigned to the dCache team.

Paul suggested he respond to the ticket, describing the issue and how the problem appears to be coming from the xrootd name-to-name plugin, which is a contributed component and not part of the dCache release. The contact person is Ilya -- He would also drop Ilya an email mentioning this issue.

Support tickets for discussion

[Items are added here automagically]


Same time, next week.