[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Condor-users] relation between schedd and dagman



On Mon, 9 Aug 2010, Mag Gam wrote:

Is there any relationship between dagman and schedd? I accidentally
stopped schedd on my central manager host then I quickly restarted it
after 45 seconds. All jobs are ok but it seems the DAGMan job started
from the beginning again. Just curious if there is a relation.

Yes, DAGMan is run as a scheduler universe job, so it is closely tied to the schedd. If you stop and restart the schedd, that will stop and restart DAGMan as well. However, when DAGMan restarts, it should go into recovery mode, not restart from the beginning. Can you send your dagman.out file? I'd like to check what actually happened.

Kent Wenger
Condor Team