Last modified 7 years ago Last modified on 08/21/14 16:59:38

dCache Tier I meeting MONTH DATE, 2013

[part of a series of meetings]

Present, Paul), IN2P3(), Sara(), Triumf(), BNL(), NDGF(), PIC(), KIT(), Fermi(Natalia), CERN()


(see box on the other side)

Site reports


Files with zero length: created but never written. If they are zero length then it is OK for dCache not to have a location for them.

Can have a look in billing for activity to see if the file has been used recently.

Bug in the pool where, at start time, the file could be deleted.

Unclear if "pnfs register" was run on the pools.


Hot replication

Scripts to manually place files on pools. Provided using dCache (e.g., admin commands) for replication then everything should be fine.

Plan is to stay with 2.2 until

A way to understand whether a patch affects 2.2

Chimera dump: 1 instance for ATLAS, CMS, LHCb with 60,000,000 entries

US-CMS 17,000,000 entries

20 mins to dump the site -- sent to the French person (on experiments site)

IN2P3 - Lyon

Script to do a 'du -ks' within Chimera database

Support tickets for discussion

[Items are added here automagically]


Proposed: same time, next week.