Next: condor_fetchlog
Up: 10. Command Reference Manual
Previous: condor_dagman
Contents
Index
Subsections
condor_drain
Control draining of an execute machine
condor_drain
[-help]
condor_drain
[-debug]
[-pool pool-name]
[-graceful | -quick | -fast]
[-resume-on-completion]
[-check expr]
machine-name
condor_drain
[-debug]
[-pool pool-name]
-cancel
[-request-id id]
machine-name
condor_drain is an administrative command used to control the draining
of all slots on an execute machine.
When a machine is draining,
it will not accept any new jobs.
Which machine to drain is specified by the argument machine-name,
and will be the same as the machine ClassAd attribute Machine.
How currently running jobs are treated
depends on the draining schedule that is chosen with a command-line option:
- -graceful
- Initiate a graceful eviction of the job.
This means all promises that have been made to the job are honored,
including MaxJobRetirementTime.
The eviction of jobs is coordinated to reduce idle time.
This means that if one slot has a job with a long
retirement time and the other slots have jobs with shorter retirement times,
the effective retirement time for all of the jobs is the longer one.
If no draining schedule is specified,
-graceful is chosen by default.
- -quick
- MaxJobRetirementTime is not honored.
Eviction of jobs is immediately initiated.
Jobs are given time to shut down and produce checkpoints,
according to the usual policy, that is,
given by MachineMaxVacateTime.
- -fast
- Jobs are immediately hard-killed,
with no chance to gracefully shut down or produce a checkpoint.
Once draining is complete, the machine will enter the Drained/Idle
state. To resume normal operation (negotiation) at that time
or any previous time during draining,
the -cancel option may be used.
The -resume-on-completion option results in
automatic resumption of normal operation once draining has completed,
and may be used when initiating draining.
This is useful for forcing a machine with a partitionable
slots to join all of the resources back together into one machine,
facilitating de-fragmentation and whole machine negotiation.
- -help
- Display brief usage information and exit.
- -debug
- Causes debugging information to be sent to
stderr, based on the value of the configuration variable
TOOL_DEBUG
- -pool pool-name
- Specify an alternate Condor pool,
if the default one is not desired.
- -graceful
- (the default) Honor the maximum vacate and
retirement time policy.
- -quick
- Honor the maximum vacate time,
but not the retirement time policy.
- -fast
- Honor neither the maximum vacate time policy
nor the retirement time policy.
- -resume-on-completion
- When done draining,
resume normal operation, such that potentially the whole machine could
be claimed.
- -check expr
- Abort draining,
if expr is not true for all slots to be drained.
- -cancel
- Cancel a prior draining request,
to permit the condor_negotiator to use the machine again.
- -request-id id
- Specify a specific draining request
to cancel,
where id is given by the DrainingRequestId
machine ClassAd attribute.
condor_drain will exit with a non-zero status value if it fails and
zero status if it succeeds.
Condor Team, University of Wisconsin-Madison
Copyright © 1990-2012 Condor Team, Computer Sciences Department,
University of Wisconsin-Madison, Madison, WI. All Rights Reserved.
Licensed under the Apache License, Version 2.0.
See the Condor Version 7.7.6 Manual or
http://www.condorproject.org/license
for
additional notices.
Next: condor_fetchlog
Up: 10. Command Reference Manual
Previous: condor_dagman
Contents
Index
condor-admin@cs.wisc.edu