We have an issue with condor 7.8.7 where I believe it is not a bug but rather the way condor is either setup or works by default.
The issue is that when I submit 20 jobs from machine say ucsc1, and each job has a requirement that is can run on either ucsc1,2,3,4,5, and when checking the output log of the script it tells me which machine it's running on.
So I know that scripts are running on machines ucsc1-5.
However when I do condor_queue -g, all the jobs that I started show up under ucsc1, the machine that I submitted it on.
condor_status shows me that other machines have claimed slots on ucsc1-5, and that ucsc1 either has none or a few claimed slots, not the 20 jobs represented in the queue.
I was wondering if it was possible to make condor_queue show me which jobs are running on which machine instead of showing me from which machine the jobs where submitted from, so that I know which job is where, instead of blinding having to open file output logs in random. This is very inconvenient.