[mrtg] FOLLOWUP2: Server statistics under Digital UNIX 4.0d

[John Losey]

My last question fixed by setting the option to "Gauge"

Now for one other question.  Does anyone have any scripts that can track
swap usage?  I've got a simple script executing a "swapon -s" and exporting
it to a text file, but I was hoping for something a bit less cumbersome than
trying to parse this log file.

Once I clean up my scripts and completely work out the bugs, I'll post my
config file & scripts that I use to parse out the vmstat results (I used the
vmstat from 4.0f because it fixed a formatting problem in 4.0d's vmstat).

Thanks,

John

-----Original Message-----
From: John Losey [mailto:JOHLOS at HBSI.COM]
Sent: Thursday, November 18, 1999 9:31 AM
To: MRTG List (E-mail)
Subject: [mrtg] FOLLOWUP: Server statistics under Digital UNIX 4.0d


I thought that I had this set up properly to record my CPU usage, but I
think something is still wrong:

my mrtg.cfg:
Target[cpu2]: `/usr/opt/local/metrics/logwatch_vm.sh
/usr/opt/local/metrics/vm.csv`
MaxBytes[cpu2]: 64000
Title[cpu2]: CPU Usage
PageTop[cpu2]: <H1>CPU Usage for lil_blarob</h1>

The script /usr/opt/local/metrics/logwatch_vm.sh produces the output for the
cpu usage with:
!/usr/local/bin/perl
print "$cpu_val\n0\n\n"

When I look at cpu2.log:
lil_blarob:root# head /webdocs/mrtg/cpu2.log
942946212 3 0
942946212 0 0 0 0
942945913 0 0 0 0
942945900 0 0 0 0
942945600 0 0 0 0
942945300 0 0 0 0
942945000 0 0 0 0
942944700 0 0 0 0
942944400 0 0 0 0
942944100 0 0 0 0

The cpu usage value appears correctly in the top line, but these values
never get moved to other lines.  I've had mrtg being called every 5 minutes
by cron and while the top line always has the current cpu usage value, the
other lines remain "0 0 0 0".

What am I doing wrong?

Thanks,

John

-----Original Message-----
From: John Losey [mailto:JOHLOS at HBSI.COM]
Sent: Wednesday, November 17, 1999 11:58 AM
To: MRTG List (E-mail)
Subject: [mrtg] Server statistics under Digital UNIX 4.0d


I am curious as to how to configure mrtg to track various server statistics.
Currently, I am running a script that runs vmstat and exports that
information into a comma delimited file that I can import into Excel.  I'm
looking to track the following statistics:  waiting procs, memory usage, cpu
usage, I/O stats, swap usage, and filesystem sizes.

I've looked through some of the script in the contrib directory, but wasn't
able to find something that would work under DU.  If anyone has managed to
track these statistics, could you please help point me in the right
direction?

Thanks,

John Losey

--
* To unsubscribe from the mrtg mailing list, send a message with the
  subject: unsubscribe to mrtg-request at list.ee.ethz.ch
* The mailing list archive is at http://www.ee.ethz.ch/~slist/mrtg

--
* To unsubscribe from the mrtg mailing list, send a message with the
  subject: unsubscribe to mrtg-request at list.ee.ethz.ch
* The mailing list archive is at http://www.ee.ethz.ch/~slist/mrtg