On 21/12/2015 21:55, Paul E Tader wrote:
Using your input and co-worker Steve T., we pieced together a configuration that we believe will work. We’re waiting for the single reconfigured node to drain first before knowing for sure.
Here's the config we have which works: (1) On the execution nodes in condor_config.local sm_pool = "dar" STARTD_ATTRS = SM_POOL, $(STARTD_EXPRS) (2) In the job submit file Requirements = ( sm_pool == "dar" ) That's it. The job only runs on those hosts. Regards, Brian.