<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
Vanhee Frederik wrote:
<blockquote cite="mid:46F767EB.4080303@perso.be" type="cite">
<pre wrap="">Rodrigo Pacheco wrote:
</pre>
<blockquote type="cite">
<pre wrap="">Colleagues good day!
I have a Nagios perfectly , would like a tool to generate graph
integrated with the Nagios. I walked searching nagiostat and perfparse
<a class="moz-txt-link-freetext" href="http://perfparse.sourceforge.net">http://perfparse.sourceforge.net</a> and nagiosgrapher. he would like to
know which of them more is detailed. that I can extract former the
weekly and dialy reports monthly: curve of load average, cpu usage,
memory, processes, downtime.
I am thankful since already attention
Rodrigo Pacheco
</pre>
</blockquote>
<pre wrap=""><!---->Hello,
I'm using NagiosGrapher and I'm very pleased with it, since it very
flexible. You can define graphs based on the performance output, but
also based on normal plugin output, using regexp.
This is very interesting when the plugins you use don't deliver the
performance output you want.
There is also a program you should look at, it is called PNP, but I
think it's only documented in German.
Frederik
</pre>
</blockquote>
PNP is nice and easy to use unless you are looking for exact data with
your graphing, since it uses rrdtool and averages all of the numbers at
specified intervals. Perfparse will keep exact numbers until you tell
it to remove them or you remove them from the database manually. The
problem with that, of course, is that it will use a great deal of disk
space (database storage), especially if you decide to keep your graphs
indefinitely. We use Perfparse currently, because we like to have exact
data, but we only keep data indefinitely for production hosts and
devices. I have not used NagiosGrapher, so I cannot comment on that.<br>
<br>
Jeff<br>
</body>
</html>