 |
 |
Hi All, I've just deployed the first few SL4 Worker nodes into my batch pool and only waiting on DNS/Firewall changes before deploying the second CE that will route jobs to them but before I set that up I've got a couple of questions. 1) Do we still need per VO queues? - I'd quite like to get my queue list down to express, short, medium and long again and I really don't want to have to have the 50-60 queues I'd need to have SL3 and SL4 versions of each of the VO queues before I even think about different memory limits. I know the VOViews stuff in the info system was supposed to make this possible again but does it work? 2) Is there any reason not to leave the second CE in place long term - it would be useful for load balacing and upgrades to have two Ces feeding jobs to the same queues but would it lead to double counting of resources? The accounting should handle it correctly so maybe it won't matter (or maybe I'll steal everyones jobs). 3) I still haven't come up with a good scheme in maui to fairshare first between Grid/Non-Grid then between groups/VOs and finally between users in the groups. Is anyone doing this? Can you send me your maui.cfg file? And an explaination. Please. The SL4 worker node install seems to have been fairly straight forward so far (though I have had jobs testing the nodes yet). I installed the glite-WN and glite-TORQUE-client metarpms (version 3.1.0-1) from the CERN yum repositorys and used yaim with my old site-info.def file with only a couple of modifications: TORQUE_SERVER -> BATCH_SERVER GLOBUS_TCP_PORT_RANGE in now comma not space separated JAVA_LOCATION has changed with the new version of java Yours, Chris.