wiki:tier-one-meeting-20171024
Last modified 13 months ago Last modified on 10/24/17 15:14:56

dCache Tier I meeting MONTH DATE, 2013

[part of a series of meetings]

Present

dCache.org(Tigran, Paul), IN2P3(), Sara(), Triumf(), BNL(), NDGF(Ulf), PIC(Elena), KIT(Xavier), Fermi(), CERN()

Agenda

(see box on the other side)

Site reports

KIT

Everything is going rather fine.

Upgrade plans ==

Plan to update dCache for LHCb tomorrow to 2.16. This is the last production instance being upgraded.

Ticket 9214

Updated BerkeleyDB configuration to try to keep some 1700 metadata files open at any time: the pool.plugins.meta.db!je.log.fileCacheSize property.

The pool currently has 850 such files. Speed improves from 5 hours to 2 hours.

Restart was on 20th October.

Memory consumption didn't change much (about the same as before).

Ticket 9282

The 'lost+found' story.

Are there any files in lost+found with a non-zero size and without any billing records?

Xavier to check.

ZK reboot

Vanishing cells within ZK reboots.

Which ZK version run at KIT? 349

xrootd GSI authn problems

RT ticket 9129

Xavier to try to capture debug output for failed transfers.

Close RT tickets

Privs to close tickets?

Currently it is all-or-nothing -- apparently no way to say this user is allowed to close their own tickets.

Tigran will investigate further.

PIC

Everything running smoothly.

NDGF

disable ==

Normal drain and reboot of head-nodes -- updating from 3.0.28 --> latest.

A GridFTP door that was removed from load balancer but still seeing transfers (small reads).

lb disable.

linkgropus

A 2.16 dCache forgot about linkgroups.

Updated other dCaches

Support tickets for discussion

[Items are added here automagically]

DTNM

Same time, in two weeks.