1CONDOR_USERLOG(1)               HTCondor Manual              CONDOR_USERLOG(1)
2
3
4

NAME

6       condor_userlog - HTCondor Manual
7
8       Display and summarize job statistics from job log files.
9
10

SYNOPSIS

12       condor_userlog  [-help ] [-total | -raw ] [-debug ] [-evict ] [-j clus‐
13       ter | cluster.proc] [-all ] [-hostname ] logfile ...
14

DESCRIPTION

16       condor_userlog parses the information in job  log  files  and  displays
17       summaries  for  each  workstation  allocation and for each job. See the
18       condor_submit manual page for instructions for specifying that HTCondor
19       write a log file for your jobs.
20
21       If  -total is not specified, condor_userlog will first display a record
22       for each workstation allocation, which includes the following  informa‐
23       tion:
24
25          Job    The cluster/process id of the HTCondor job.
26
27          Host   The host where the job ran. By default, the host's IP address
28                 is displayed. If -hostname is specified, the host  name  will
29                 be displayed instead.
30
31          Start Time
32                 The  time  (month/day hour:minute) when the job began running
33                 on the host.
34
35          Evict Time
36                 The time (month/day hour:minute) when  the  job  was  evicted
37                 from the host.
38
39          Wall Time
40                 The  time (days+hours:minutes) for which this workstation was
41                 allocated to the job.
42
43          Good Time
44                 The allocated time (days+hours:min) which contributed to  the
45                 completion  of this job. If the job exited during the alloca‐
46                 tion, then this value will equal "Wall Time." If the job per‐
47                 formed  a checkpoint, then the value equals the work saved in
48                 the checkpoint during this allocation. If  the  job  did  not
49                 exit  or  perform  a  checkpoint  during this allocation, the
50                 value will be 0+00:00. This value can be greater than  0  and
51                 less than "Wall Time" if the application completed a periodic
52                 checkpoint during the allocation  but  failed  to  checkpoint
53                 when evicted.
54
55          CPU Usage
56                 The  CPU  time (days+hours:min) which contributed to the com‐
57                 pletion of this job.
58
59       condor_userlog will then display summary statistics per host:
60
61          Host/Job
62                 The IP address or host name for the host.
63
64          Wall Time
65                 The workstation time (days+hours:minutes) allocated  by  this
66                 host to the jobs specified in the query. By default, all jobs
67                 in the log are included in the query.
68
69          Good Time
70                 The time (days+hours:minutes) allocated on  this  host  which
71                 contributed  to  the  completion of the jobs specified in the
72                 query.
73
74          CPU Usage
75                 The CPU time (days+hours:minutes)  obtained  from  this  host
76                 which  contributed to the completion of the jobs specified in
77                 the query.
78
79          Avg Alloc
80                 The  average  length  of   an   allocation   on   this   host
81                 (days+hours:minutes).
82
83          Avg Lost
84                 The  average  amount of work lost (days+hours:minutes) when a
85                 job was evicted from this host without successfully  perform‐
86                 ing a checkpoint.
87
88          Goodput
89                 This  percentage  is  computed  as  Good Time divided by Wall
90                 Time.
91
92          Util.  This percentage is computed as  CPU  Usage  divided  by  Good
93                 Time.
94
95       condor_userlog will then display summary statistics per job:
96
97          Host/Job
98                 The cluster/process id of the HTCondor job.
99
100          Wall Time
101                 The  total workstation time (days+hours:minutes) allocated to
102                 this job.
103
104          Good Time
105                 The total time (days+hours:minutes)  allocated  to  this  job
106                 which contributed to the job's completion.
107
108          CPU Usage
109                 The  total CPU time (days+hours:minutes) which contributed to
110                 this job's completion.
111
112          Avg Alloc
113                 The average length of a workstation  allocation  obtained  by
114                 this job in minutes (days+hours:minutes).
115
116          Avg Lost
117                 The  average  amount  of  work lost (days+hours:minutes) when
118                 this job was evicted from a host  without  successfully  per‐
119                 forming a checkpoint.
120
121          Goodput
122                 This  percentage  is  computed  as  Good Time divided by Wall
123                 Time.
124
125          Util.  This percentage is computed as  CPU  Usage  divided  by  Good
126                 Time.
127
128       Finally, condor_userlog will display a summary for all hosts and jobs.
129

OPTIONS

131          -help  Get a brief description of the supported options
132
133          -total Only display job totals
134
135          -raw   Display raw data only
136
137          -debug Debug mode
138
139          -j     Select a specific cluster or cluster.proc
140
141          -evict Select only allocations which ended due to eviction
142
143          -all   Select all clusters and all allocations
144
145          -hostname
146                 Display host name instead of IP address
147

GENERAL REMARKS

149       Since the HTCondor job log file format does not contain a year field in
150       the timestamp, all entries are assumed to occur in  the  current  year.
151       Allocations  which  begin  in  one  year  and  end  in the next will be
152       silently ignored.
153

EXIT STATUS

155       condor_userlog will exit with a status value of 0 (zero) upon  success,
156       and it will exit with the value 1 (one) upon failure.
157

AUTHOR

159       HTCondor Team
160
162       1990-2022,  Center for High Throughput Computing, Computer Sciences De‐
163       partment, University of Wisconsin-Madison, Madison,  WI,  US.  Licensed
164       under the Apache License, Version 2.0.
165
166
167
168
1698.8                              Jan 19, 2022                CONDOR_USERLOG(1)
Impressum