1CONDOR_NOW(1) HTCondor Manual CONDOR_NOW(1)
2
3
4
6 condor_now - HTCondor Manual
7
8 Start a job now.
9
10
12 condor_now -help
13
14 condor_now [-name] [-debug] now-job vacate-job
15
17 condor_now tries to run the now-job now. The vacate-job is immediately
18 vacated; after it terminates, if the schedd still has the claim to the
19 vacated job's slot - and it usually will - the schedd will immediately
20 start the now-job on that slot.
21
22 You must specify each job using both the cluster and proc IDs.
23
25 -help Print a usage reminder.
26
27 -debug Print debugging output. Control the verbosity with the envi‐
28 ronment variables _CONDOR_TOOL_DEBUG, as usual.
29
30 -name **
31 Specify the scheduler('s name) and (optionally) the pool to
32 find it in.
33
35 The now-job and the vacated-job must have the same owner; if you are
36 not the queue super-user, you must own both jobs. The jobs must be on
37 the same schedd, and both jobs must be in the vanilla universe. The
38 now-job must be idle and the vacated-job must be running.
39
41 To begin running job 17.3 as soon as possible using job 4.2's slot:
42
43 condor_now 17.3 4.2
44
45 To try to figure out why that doesn't work for the 'magic' scheduler in
46 the 'gandalf' pool, set the environment variable _CONDOR_TOOL_DEBUG to
47 'D_FULLDEBUG' and then:
48
49 condor_now -debug -schedd magic -pool gandalf 17.3 4.2
50
52 condor_now will exit with a status value of 0 (zero) if the schedd
53 accepts its request to vacate the vacate-job and start the now-job in
54 its place. It does not wait for the now-job to have started running.
55
57 HTCondor Team
58
60 1990-2020, Center for High Throughput Computing, Computer Sciences
61 Department, University of Wisconsin-Madison, Madison, WI, US. Licensed
62 under the Apache License, Version 2.0.
63
64
65
66
678.8 Aug 06, 2020 CONDOR_NOW(1)