 |
 |
Thanks Ewan > We have an EMI2 on SL5 test system behind the CE 't2ce02.physics.ox.ac.uk'. > It's an EMI 2 on SL5 Cream CE, with a pair of 8-core EMI 2 on SL5 worker nodes > (so a grand total of sixteen cores). It's otherwise identical to the rest of the > site, same VO software area, tags, cvmfs config, glExec support and whatnot. > > I'm rather assuming that there's no point testing EMI 2 on SL 6 on the worker > nodes because we know the VOs aren't anywhere near ready for it yet. In general I would say that EMI-2 on SL5 is the safest path to follow and that configuration still needs experiment validation. However, the VOs will also shortly need some SL6 test clusters so there is some point. Raul just informed me that he has EMI-1 Cream, EMI-1 WN, glexec, Argus and also one EMI-2 Cream (with glexec) running on SL6 with 16 cores. "CMS is using it to port software to SL6. After two months of debugging, first tests running since yesterday. First major block was to get the SL6 WN to access data on DPM". Perhaps we should aim for at least one other EMI-2 SL6 site to work closely with ATLAS and LHCb? We can discuss that on Tuesday. regards, Jeremy