Historical Job Data Files
vovserver creates a CSV (comma-separated variable) file containing data about each job that ran. This format may be directly imported into many spreadsheet and database programs. The jobs files are stored in the jobs subdirectory of the vovserver configuration directory.
For example, for the default setup of Accelerator, the job files
are stored in the directory as shown below:
$VOVDIR/../../vnc/vnc.swd/jobsThe jobs files are rotated automatically each day by the vovserver, and older ones are compressed. Each file is named according to the day it
applies to, in the form YYYY.MM.DD, where YYYY is
the year, MM is the 2-digit month, and DD is the
two-digit day of the month.
Each file contains a few header lines that identify the format version and the order
of the fields. In the following example, the backslashes indicate line breaks
inserted for readability. The comment denoting the fields is actually one long line.
The available fields may vary by jobs file version as shown below:
# V 1
# T 1173942608
# FIELDS: ID,JOBCLASS,PROJECT,GROUP,USER,OSGROUP,DP,DIR,ENV,TOOL,JOBNAME,
SPRIORITY,XPRIORITY,RESOURCES,GRABBED,LICENSES,SUBMITHOST,EXEHOST,DPHOSTS,
SUBMITTIME,STARTTIME,ENDTIME,STATUS,EXIT,MAXRAM,CPUTIME,QUEUEJobs File Fields Description
The following table shows the type and meaning of the fields in the jobs file.
| Field Name | Data Type | Description |
|---|---|---|
| ID | %d | Numeric ID of the job |
| JOBCLASS | %s | Name of jobclass in which job was submitted; may be empty |
| PROJECT | %s | Name of management project on behalf of which job was run |
| GROUP | %s | Name of FairShare group in which job was run |
| USER | %s | Name of user as which the job ran |
| OSGROUP | %s | List of user.group |
| DP | %d | Boolean; 1 if distributed parallel job, 0 if not |
| DIR | %s | Working (run) directory of the job |
| ENV | %s | Name of the environment in which the job was run |
| TOOL | %s | Name of the program (first word on command line) that the job ran |
| JOBNAME | %s | User-assigned name of the job; may be null |
| SPRIORITY | %d | User-assigned scheduling priority of the job |
| XPRIORITY | %d | User-assigned execution priority of the job |
| RESOURCES | %s | Resources requested by the job |
| GRABBED | %s | Resources actually assigned to the job, after all mapping |
| LICENSES | %s | License features used by the job |
| SUBMITHOST | %s | Name of host from which job was submitted |
| EXEHOST | %s | Name of host on which job was executed |
| DPHOSTS | %s | List of host names on which distributed parallel job was executed |
| SUBMITTIME | %d | Time job was submitted, seconds since the Epoch |
| STARTTIME | %d | Time job started running, seconds since the Epoch |
| ENDTIME | %d | Time job finished running, seconds since the Epoch |
| STATUS | %s | String describing status of the job, e.g. 'Done' |
| EXIT | %d | Exit status of the job, e.g. 0 |
| MAXRAM | %d | Maximum RAM used by the job, in MBytes |
| CPUTIME | %d | Cumulative CPU time used by the job, in seconds |
| QUEUE | %d | Time the job waited in queue, in seconds |