Attending: Brian, Duncan, Gareth, Jens, John B, John H, Sam, Tom, Steve, Wahid, Elena, Matt D, Gang, Chris, David, Raul, Robert, Alessandra, Raja, Ewan, Govind, Adam Apologies: Dave C 0. Operational issues Good results with CERN@School on DIRAC, now need to address metadata. There is a tutorial on the DIRAC web site (https://github.com/DIRACGrid/DIRAC/wiki/DIRAC-Tutorials) Using portal at IC, easier to use. What is the status of DPM 1.8.8? Below see. 1. DPM collaroration meeting/developers meeting Jeremy, Sam, and Wahid attended the DPM collaboration meeting which was followed by a developers' meeting. The UK agreed to endeavour to contribute to IPv6, S3 testing, toolkit obviously, WebDAV, running tests, and migrating documentation (which is to migrate to Drupal). EPEL testing will be done by all. 1.8.8 is now in EPEL, and contains an important security fix (privilege escalation). John H reported dependency issues but this was probably due to migrating to EPEL and the phase of the Moon. GridFTP will be moving to DMLite so you should be able to get rid of the pool accounts. No Argus integrated yet, as Argus is not in EPEL. Sam has prodded, and asks Ewan to prod, too. Both Wahid and Sam have 1.8.8 in test. Used puppet, not Yaim. Eventually puppet should be the way forward, so it may be best to get the puppet snippets troubleshot rather than do more testing with Yaim. Wahid will give it another go with puppet, this week. We should now have a volunteer site upgrade and report (to the list). 2. Data access pre-GDB next week: http://indico.cern.ch/event/272787/ Wahid is going, as he is chairing the event. No one else expected to go (from GridPP T2s, maybe someone from T1), but everyone is welcome to attend remotely, too. 3. "Big data" update RAL hosted a research councils and public bodies data event and in many ways they are ahead of HEP in making big data available (the other question is what the public would do with the HEP data, but that is the wrong question to ask.) Humanities for example are pretty advanced with respect to making data available and usable and preserved in comparison with HEP, perhaps due to having taken the problem seriously for a long time. In the interest of time Jens would turn it into a blog post or something - maybe one now with quick thoughts and another one later once the presentations appear and a proper writeup thing can be, er, written up. 4. AOB Brian thought StoRMs were behind but it turns out something had broken in Wahid's monitoring, so it was the monitoring page which was behind. Chris's StoRM is at 1.11.4. The LDAP query should be accurate, though... Chris asked about having these very notes in CERN's indico. Chatlog John Hill: (10:13 AM) 1.8.8 is in epel, but I get dependency issues Ewan Mac Mahon: (10:17 AM) I saw it with the EMI metapackage looking for 1.8.7. components, and EPEL having 1.8.8. bits in it. I care about it. :-( Matt Doidge: (10:21 AM) 1.8.8 is the version that doesn't need pool accounts right? Samuel Cadellin Skipsey: (10:21 AM) Yes, Matt - the only thing that needed them was gridftp which is moved to dmlite in 1.8.8 Matt Doidge: (10:23 AM) Sweet. Samuel Cadellin Skipsey: (10:26 AM) https://svnweb.cern.ch/trac/lcgdm/attachment/wiki/Dpm/DPM_releasenotes_Mar2014.txt (These are the release notes for DPM 1.8.8 components) In particular, note the Known Issues near the top. Ewan Mac Mahon: (10:29 AM) The Oxford storm is not 'in production', nor is it running. Nor is it likely to come back. Brian@RAL: (10:29 AM) https://twiki.cern.ch/twiki/bin/view/LCG/WLCGBaselineVersions wahid: (10:31 AM) ah I remember its been broken since my old desktop was taken away from me. sorry I totally forgot that - will reinstalll the cron Jens Jensen: (10:32 AM) http://indico.cern.ch/event/272787/ Christopher John Walker: (10:36 AM) StoRM 1.11.3 EMI-3 / UMD-3 / SL5 / SL6 Release notes - security and bugfix release. So sites should be installing it. Duncan Rand: (10:36 AM) At IC we have : storm-backend-server-1.11.3-1.el6.noarch Christopher John Walker: (10:36 AM) Once Wahid has fixed his page, you should nag sites that are not up to date. Duncan Rand: (10:37 AM) is that what you're referring to? Christopher John Walker: (10:37 AM) Yes. Duncan Rand: (10:37 AM) that rpm I mean? Ewan Mac Mahon: (10:37 AM) I'm slighty sceptical about that view from the OS. Duncan Rand: (10:38 AM) It's been installed for a while: Feb 03 03:40:08 Updated: storm-backend-server-1.11.3-1.el6.noarch Ewan Mac Mahon: (10:38 AM) Given that their failure to open their data led directly to people having to implement openstreetmap from scratch. Michael Adam James Huffman: (10:38 AM) I'm working on the open release of data for CMS. Tom Whyntie: (10:38 AM) All CERN@school data will be open to school students Ewan Mac Mahon: (10:39 AM) The problem wasn't that no-one knew what to do with the data, it's that OS were trying to charge people a boatload of money for using it. wahid: (10:39 AM) yeah that was mental yes please do post slides on big data conference Tom Whyntie: (10:40 AM) Also useful for meeting material, slides etc.