pbsacct 3.4.6

Python library for parsing PBS accounting logs

The pbsacct Python module is an inteface for processing accounting log data from the various PBS variants (TORQUE, PBS Pro, and OpenPBS), including the support sending that data to a database for persistance and analysis. DB persistance is currently supported for MySQL/MariaDB, with SQLite and PostgreSQL in various stages of development.

For example, here is a simple example of using the module to read a TORQUE accounting log file and print information about the jobs logged therein:

#!/use/bin/env python
import pbsacct
jobs = pbsacct.jobs_from_file("/var/spool/torque/server_priv/accounting/20180627")
for jobid in jobs.keys():
  job = jobs[jobid]
  print "%s: user=%s group=%s account=%s" % \
        (jobid,job.user(),job.group(),job.account())
# Here endeth the program

More complete examples (including DB connectivity) can be found at:

https://github.com/tabaer/pbstools/blob/master/sbin/job-db-update https://github.com/tabaer/pbstools/blob/master/sbin/jobscript-to-db https://github.com/tabaer/pbstools/blob/master/sbin/pbsacct-dump https://github.com/tabaer/pbstools/blob/master/sbin/sw_app-cache https://github.com/tabaer/pbstools/blob/master/sbin/sw_app-index

The pbsacct Python module is part of the pbstools collection of utilities:

https://github.com/tabaer/pbstools