Brian Bockelman <bbockelm@...> writes:
> I donât know what might be going on, but it sounds like they are being
reset by the negotiator during the
> negotiation cycle. Two ideas:
> - Verify the negotiator is running 8.2.
> - Crank up the logs to D_FULLDEBUG on the negotiator and see if there are
obvious problems logged.
Thanks for your advice. I ended up stopping condor on all nodes, making
sure all processes were turned off, then restarting. That itself didn't do
the trick, though.
The problem seems to be solved after running "condor_prio -setfactor
blahblah" (for each user) on each node separately. So far, the modified
priority factors are sticking.
Bryan
|