Description of problem: I'm periodically seeing a non-wise selection of nodes. I have several LC with many hosts and when the utilization of one goes high, following happens: RS: node-1 - scheduled node-2 - scheduled node-3 - scheduled node-4 - queued This is not wise, as there are other LCs that would be suitable for the job and have free resources. So while I've plenty of resources available, the job is delayed until one particular LC is free. Version-Release number of selected component (if applicable): 0.11.3 How reproducible: always Steps to Reproduce: 1. schedule recipeset with more recipes than nodes available on one LC, while having another LC with enough resources to run it too 2. wait for it to happen :) 3. Actual results: job delayed, resources not used effectively Expected results: job scheduled on first free LC that could host it, if none is found, queue it Additional info:
*** This bug has been marked as a duplicate of bug 709012 ***