holland-discuss team mailing list archive

["David Maynard" <dmaynard@xxxxxxxxxxx>]

What is the best way to monitor whether Holland is running successfully?

 

I am running Holland 1.0.6 from the EPEL RPMs on CentOS 5.x using the
mysqldump provider.  It is running on a Rackspace next-gen cloud server.

 

When I run it from the command line, it produces the expected output to
stderr according to the log level.  When I run it from cron, it doesn't
appear to produce any output to stdout or stderr.  The messages do get
logged to /var/log/holland/holland.log.  I assume it notices when it is
running without a tty and doesn't generate the console output.

 

I considered trying to write a logwatch parser for the holland.log file,
but it doesn't use the standard syslog time/line format.  I'm not a
logwatch expert, so creating a new plugin from scratch looks like more
work than I was hoping for.  Holland is "nice" enough to clearly flag
the different log levels, so picking out errors should be
straightforward once you  get past the basic date filtering.

 

Any other suggestions on the best approach to monitoring holland for
failures?  Something that could be turned into a Nagios plugin would be
great.  I'm sure I could create a ugly hack that grepped the
holland.log, but it seems like there should be a better way.

 

Thanks!

 

-dpm

--
  David P Maynard, OutServ.net, Inc.
  Email: dmaynard@xxxxxxxxxxx <blocked::mailto:dmaynard@xxxxxxxxxxx> ,
Tel: +1 (512) 977-8918 x0, Fax: +1 (512) 501-6366
--