1condor_vacate(1)            General Commands Manual           condor_vacate(1)
2
3
4

Name

6       condor_vacateVacate jobs that are running on the specified hosts
7

Synopsis

9       condor_vacate[-help | -version]
10
11       condor_vacate[-graceful  |  -fast]  [-debug] [-pool centralmanagerhost‐
12       name[:portnumber]] [ -name hostname| hostname| -addr  "<a.b.c.d:port>"|
13       "<a.b.c.d:port>"| -constraint expression| -all]
14

Description

16       condor_vacatecauses HTCondor to checkpoint any running jobs on a set of
17       machines and force the jobs to vacate the machine. The  job(s)  remains
18       in the submitting machine's job queue.
19
20       Given  the  (default) -gracefuloption, a job running under the standard
21       universe will first produce a checkpoint  and  then  the  job  will  be
22       killed.  HTCondor  will  then restart the job somewhere else, using the
23       checkpoint to continue from where it left off. A job running under  the
24       vanilla  universe  is  killed,  and  HTCondor restarts the job from the
25       beginning somewhere else. condor_vacatehas no effect on a machine  with
26       no HTCondor job currently running.
27
28       There  is generally no need for the user or administrator to explicitly
29       run condor_vacate. HTCondor takes care of jobs in  this  way  automati‐
30       cally following the policies given in configuration files.
31

Options

33       -help
34
35          Display usage information
36
37
38
39       -version
40
41          Display version information
42
43
44
45       -graceful
46
47          Inform the job to checkpoint, then soft-kill it.
48
49
50
51       -fast
52
53          Hard-kill jobs instead of checkpointing them
54
55
56
57       -debug
58
59          Causes  debugging  information  to  be  sent to stderr, based on the
60          value of the configuration variable TOOL_DEBUG.
61
62
63
64       -pool centralmanagerhostname[:portnumber]
65
66          Specify a pool by giving the central  manager's  host  name  and  an
67          optional port number
68
69
70
71       -name hostname
72
73          Send the command to a machine identified by hostname
74
75
76
77       hostname
78
79          Send the command to a machine identified by hostname
80
81
82
83       -addr <a.b.c.d:port>
84
85          Send the command to a machine's master located at "<a.b.c.d:port>"
86
87
88
89       <a.b.c.d:port>
90
91          Send the command to a machine located at "<a.b.c.d:port>"
92
93
94
95       -constraint expression
96
97          Apply  this  command  only  to  machines  matching the given ClassAd
98          expression
99
100
101
102       -all
103
104          Send the command to all machines in the pool
105
106
107

Exit Status

109       condor_vacatewill exit with a status value of 0  (zero)  upon  success,
110       and it will exit with the value 1 (one) upon failure.
111

Examples

113       To send a condor_vacatecommand to two named machines:
114
115       % condor_vacate  robin cardinal
116
117       To send the condor_vacatecommand to a machine within a pool of machines
118       other than the local pool, use the -pooloption.  The  argument  is  the
119       name  of  the  central  manager  for  the  pool.  Note that one or more
120       machines within the pool must be specified as the targets for the  com‐
121       mand.  This  command  sends  the  command to a the single machine named
122       cae17within the pool of machines  that  has  condor.cae.wisc.eduas  its
123       central manager:
124
125       % condor_vacate -pool condor.cae.wisc.edu -name cae17
126

Author

128       Center for High Throughput Computing, University of Wisconsin-Madison
129
131       Copyright  (C) 1990-2019 Center for High Throughput Computing, Computer
132       Sciences Department, University of Wisconsin-Madison, Madison, WI.  All
133       Rights Reserved. Licensed under the Apache License, Version 2.0.
134
135
136
137                                     date                     condor_vacate(1)
Impressum