Provisioning: Ensure merge/logcollect jobs get priority on the T1
The pilots are coming in as user "cmsprod" and competing for resources with all other pilots.
We should increase the priority for pilots coming from the HEPCloud demo.
#3 Updated by Burt Holzman almost 3 years ago
- Status changed from Work in progress to Resolved
- % Done changed from 90 to 100
Yes. We changed the cronjob on cmssrv279 to do voms-proxy-init with a different role (/cms/Role=cmst1admin). Tony had Merina map that role to cmsprod, and he (via puppet) changed the CMS gatekeepers to assign a HEPCloud accounting group to the role. He then changed the condor priority for the HEPCloud accounting group to take priority over global CMS pool pilots.