This is an outdated version of the HTCondor Manual. You can find current documentation at http://htcondor.org/manual.
next up previous contents index
Next: 7.7 Troubleshooting Up: 7. Frequently Asked Questions Previous: 7.5 Grid Computing   Contents   Index

Subsections

7.6 Managing Large Workflows

How do I get meaningful output from condor_q with so many jobs in the queue?

There are several ways to constrain the output of condor_q when there are lots and lots of jobs in the queue. To see only the jobs that are currently running:

  condor_q -run
To see only the jobs that are currently on hold:
  condor_q -hold
To see other output, combine options. For example, to see only running jobs submitted by A. Einstein that belong to cluster 77:
  condor_q -run einstein 77
Another example uses the -constraint option to condor_q. To see only the jobs in the queue that started running, but were interrupted and then started again from the beginning, perhaps more than once:
  condor_q -constraint 'NumJobStarts > 1'
Complete details of condor_q are contained in the manual page at page [*].

What does HTCondor offer that can help with running a large number of jobs?

Many of the features of DAGMan are targeted at helping with the administration and running of large numbers of jobs. See section 2.10.13 at page [*].


next up previous contents index
Next: 7.7 Troubleshooting Up: 7. Frequently Asked Questions Previous: 7.5 Grid Computing   Contents   Index
htcondor-admin@cs.wisc.edu