Iâm running a small pool of machines with a central manager running Mac OS 10.13.6 and several Windows execute machines (running Windows 7, 10, Server 2012). Weâve noticed a significant slowdown in response from any of the condor_ commands on the central manager, e.g., condor_q, condor_submit, and condor_status (it takes about a minute). Even though weâve been running HTCondor for some time (over 5 years), I think Iâm pretty much of a noobie when it comes to diagnosing problems, since most of the time Iâve had issues, I was able to resolve them quickly. But this one has me stumped and I donât know where to look to figure out what is going wrong and how to correct it. I think it might have started when I upgraded to 8.6.1, but that was a while back and the problem only seems to be recent (within the last couple months). Iâm pretty sure all the execute nodes are also running 8.6.1.
I did do some searching and cleared up one problem, I think, and that was setting ENABLE_IPV6 to False, since I was getting an init_local_hostname. Also, since our pool is behind a firewall, I decided to set USE_SHARED_PORT to False. Neither has resolved the issue.
Any help, guidance, places to look, would be greatly appreciated!
Stephen C. Upton
Faculty Associate - Research
SEED (Simulation Experiments & Efficient Designs) Center for Data Farming
Operations Research Department
Naval Postgraduate School
SEED Center website: https://harvest.nps.edu