Sam chairing and reporting:

No Brian, and Elena had no knowledge of the ATLAS deletion rate
improvement gubbins at RAL, so we had to skip that item.

On DPM:

Raja felt that providing srmless interfaces was quite nice for LHCb
(but didn't feel that there was much of high priority that needed done
for DPM, as they just use xrootd interfaces).

Ewan was of the opinion that there's some conflict between the aims of
"providing SRMless operation" for DPM, and "making DPM relevant" - the
core objection was that by removing the need for the SRM, we also make
it much easier to support sites which are just commodity parallel
distributed storage volumes sitting behind xrootd or gridftp
interfaces (with no need for the extra DPM gubbins).

There was a general agreement that the Puppet configuration stuff
could be improved, at least documentation and stability wise. (This
had already been forwarded to the DPM core team at the dev meeting
last week.)

We will have a presentation in a few weeks from Lukcas on the storage
at Bristol, which may become the first production instance of the DPM
on HDFS interface.

Chat log:
John Bland: (01/04/2015 10:09)
space tokens/quotas without srm would be nice
puppet working 100% is the only thing really holding me back from 1.8.9
Paige Winslowe Lacesso: (10:29 AM)
uess mike not work
Yes, Lukas is planning to give a talk
once got some config/data
No production yet
Alas - I'm out of loop did 0%
so far LK did it all
I WISH I was more involved :(
Ewan Mac Mahon: (10:31 AM)
:-)
Now the modules are supposedly installable I'll be able to resume
trying to do a puppet DPM, so I should be able to run through the
whole process, so if there are other problems, I should run straight
into them.