Last modified 3 years ago Last modified on 10/25/16 15:06:47

dCache Tier I meeting October 25, 2016

[part of a series of meetings]

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


(see box on the other side)

Site reports


Xavier reported that KIT is working fine.

Slow Deletions

Paul mentioned the back-port of concurrent notification. Future work planned.

HPSS parallel test

Problem with concurrency has been investigated and the likely culprit found.

GPFS mount works fine with 200 concurrent copies, but HPSS fuse mount is limited to 20 concurrent transfers.

After 20 copies complete -- it took a couple of seconds to "wake up". Further investigation will take place.


Currently running multiple cleaners to keep up the back-log of deletions.

NDGF is now running fully HA, apart from transfer manager.

Ulf reported that NDGF is planning to exercise HA by doing a full upgrade on Thursday/Friday?.


No issues this week.

Pool overloaded

One issue with a pool. Initially empty, CMS wrote lots of data that went to this pool. Now they are reading it -- causing high io-load The partition is WASS.

Marc does not consider this critical, and will rebalance the pools.


Marc asked about storage accounting and APEL. This was triggered by a Spanish all-hands OPS meeting last Monday.

Marc will investigate further and open a support ticket.

Support tickets for discussion

[Items are added here automagically]


Same time, next week.