[Note from the taker of the minutes: Even if you dial in using the Phone bridge, please use a PC and click on the link to the audioconference given in the agenda. Enter your name and affiliation and that of the others in the same room - that is what will determine the list of attendees (more efficient than a tedious roll-call)! Thank you]
The meeting started 5 minutes late, at 16:05.
Feedback on Last Week's Minutes
None was given.
EGEE Items
Grid Operator Hand Over on Duty
Primary Team
Secondary Team
From
ROC UK/I
ROC SEE
To
ROC CE
ROC NE
In UK/I, a Certificate Authority update led to a few SAM CA failures.
The site RO-02-NIPNE has been given one more week by Kostas to sort out their problems or be suspended.
PPS Reports
Glexec has been installed since 2 weeks in PPS, but no experimental activity. Antonio requested that experiments test the software.
There were a few updates at the beginning of last week for CreamCE (ALICE are happy; FZK set up 300 cores on their CreamCE). The pilot is being extended to the end of August.
See the link in the agenda for more detail.
gLite Release News
Certified version of CreamCE (PPS update 33) released last week. It's "a few steps behind" the one used in the Cream pilot.
Release 34 being prepared (latest DPM and LFC, plus dCache with new configuration module).
The link to the detailed release news is on the agenda page, but is repeated here for your convenience.
Clemens had a question about MPI support. Is there a new release, better documentation? Antonio asked Clemens to open a Savannah bug against the documentation.
Issues from ROC reports
Nothing significant reported last week
Jeremy mentioned that a change in the CA hierarchy in the UK had affected quite a few users. Poor communication was the main culprit.
Upcoming SRM v2 tests for SAM
Konstantin Skaburskas from the SAM team gave an overview of the new SRM v2 tests. They are briefly described here, and in more detail here. The gist of the presentation was that GOCDB is being asked to refine the definition of the services SE and SRM, and that SAM will have different tests for the three services (SE, SRMv1 and SRMv2). The new tests (which remove dependencies on LFC and BDII) are in Validation, but fixes to lcg_utils are necessary before releasing them to production.
Alessandro mentioned that ATLAS would have some cleaning up to do because they'd been relying on SE as providing the union of SE and SRM tests. Konstantin asked where the criticality of tests should be discussed, and Diana said that she would send a pointer to the process (and she did).
Experiences with WMS
Jeremy wanted to know what others' experience of WMS was. How do they fare in production?
SE (Kostas): 3.1 seems more stable than 3.0., but lots of memory required.
DECH (Clemens) gave some feedback but was inaudible due to static on the line.
CERN said they'd give written feedback via the service managers.
RBs are deprecated from a support point of view, but many are still in use - is it because if problems with WMS?
At this point, John suggested that everyone send one paragraph on their WMS experience to him for compilation. It was also suggested to contact the site representatives.
WLCG Items
WLCG issues coming from ROC reports
None
Nick asked that Tier1s supply the FTM end-points that they have.
Upcoming WLCG Service Interventions
PIC downtime on 5 Aug 08:00-20:00 UTC
Stop press: RAL taking Castor down tomorrow (29/7/08) all day.
ATLAS Service
Alessandro had nothing special to report, but thanked the GGUS developers for team tickets and alarm tickets!
ALICE Service
No one present.
CMS Service
No one present.
LHCb Service
No one present.
WLCG Service Coordination
Operational review (Harry): new Castor patch level just come out, dcache coming, check link to baseline services.
Nick mentioned some of the standing items in the agenda, namely the link to baseline version information. A link to storage points of interest needs to be provided; for the moment, they’re in the agenda.
No tickets for discussion found by Nick, and no OSG representatives in the meeting. Harry mentioned that an important Atlas site (AGL-T2) had disappeared from the BDII, and it was urgent to fix this [John contacted the OSG people after the meeting and they solved the problem].