[mrtg] Fine tuning my email MRTG logger

[Scott Haneda]

Thanks to all the fine folks here, I have managed to get MRTG to report a
graph based on my incoming and outgoing email server stats.  I have some
small technical problems I would like to overcome, and just do not fully
grasp all the options MRTG has to help me.  This may get a little long, but
hopefully someone can take a stab at each issue:

The data comes from a remote machine, I supply it to MRTG via curl and send
back data such as this:
12
45
1134103425
eims

I don't entirely understand how to generate the cfg file, I did one by hand,
by copying bits of other cfg files.  There is some trailing html bits in the
file, and I am not sure I need that stuff.

Right now, I have eims.cfg, in it is this:
EnableIPv6: no
Options[_]: growright,perminute,pngdate,gauge
Htmldir: /Library/WebServer/Documents/mrtg/http/
ImageDir: /Library/WebServer/Documents/mrtg/http/images/
LogDir: /Library/WebServer/Documents/mrtg/logs/
IconDir: images_mrtg/


Target[eims]: `curl -s http://www.example.com/test.php`
MaxBytes[eims]: 12500000
Title[eims]: eims
PageTop[eims]: <h1>eims</h1>

I assume those options are global ones, and if I want them to just apply to
this report I need to do this:
Options[eims]: growright,perminute,pngdate,gauge

When I run indexmaker on the eims.cfg file, I end up sending it to >
eims.html and get the normal report, with the 4 graphs on it.  Good so far,
but I also have a 24 port switch, and would like to get the one 5 minute
average in that index.html page to link to the detail eims.html page.

The 24 port switch has a configmaker line I run, and it does everything
perfect, I will run it from time to time as I add more servers to the port.
So, I don't really want to always have to paste in the eims.cfg data to the
switch.cfg file and have it get overwritten.  So, how do I get a cfgmaker
line to run that will "include" my eims.cfg and not have it interfere with
other settings and options for the switch?

This is just a raw count of emails in and out, so it is not bytes over time
at all, it is merely emails per time. Right now, the 5 minute average says
Bytes per minute and the weekly month and yearly graphs also say similar.
Can that be changed to something else?

Within the left portion of the graph is the numbers that represent in this
case bytes, can I logically think of those as "emails"?

My yearly graph has the word 'eims' in the upper left of the x/y, none of
the others have this, how did this get here, and how do I eitehr get it on
all of them, or none of them?

I am getting:
The statistics were last updated Thursday, 8 December 2005 at 20:55,
at which time 'eims' had been up for 1134104026.
As you can see, uptime is totally off, I am sending in seconds since epoch
as line 3 to mrtg, how is this calculated?  I would probably like to just
remove this line, as it is not relative at all.

I think that covers most of the smaller issues, thanks again.
-- 
-------------------------------------------------------------
Scott Haneda                                Tel: 415.898.2602
<http://www.newgeo.com>                     Novato, CA U.S.A.


--
Unsubscribe mailto:mrtg-request at list.ee.ethz.ch?subject=unsubscribe
Archive     http://lists.ee.ethz.ch/mrtg
FAQ         http://faq.mrtg.org    Homepage     http://www.mrtg.org
WebAdmin    http://lists.ee.ethz.ch/lsg2.cgi