<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#ffffff">
<br>
<blockquote
cite="mid:02B80625-B645-4C15-A3CC-9D2679F206C2@salon.com"
type="cite">
<div>Assaf - thanks for your time and patience outlining those
steps:</div>
<div><br>
</div>
</blockquote>
<br>
That is what the list is about .<br>
<br>
<blockquote
cite="mid:02B80625-B645-4C15-A3CC-9D2679F206C2@salon.com"
type="cite">
<div>I was able to at least get the connectivity between Nagios
server and the monitored host to show a valid connection:</div>
<div><br>
</div>
<div>
<div>/usr/local/nagios/libexec/check_nrpe -H 10.0.100.139 -c
check_users</div>
<div>USERS OK - 1 users currently logged in |users=1;5;10;0</div>
<div><br>
</div>
<div>there were a few things wrong on the monitoring server, or
at least what I changed:</div>
<div><br>
</div>
<div>* changed ownership of nrpe.cfg to nagios.nagios</div>
<div>* modified xinetd.d/nrpe to show Nagios server IP as
allowed connection</div>
<div><br>
</div>
<div>I struggled with a few other things, and connection
attempts, before I realized I hadn't restarted the nrpe
service after modifying the xinetd.d/nrpe file.</div>
<div><br>
</div>
<div>I did notice that the plugin location from another server
is different than the one that is missing:</div>
<div>
<div style="margin: 0px; font: 12px Helvetica;"><i><b>/usr/lib64/nagios/plugins</b></i>
vs <i>/<b>usr/local/nagios/libexec/</b></i> (guess this has
something to do with the way it was compiled)</div>
</div>
</div>
</blockquote>
<br>
NRPE by default is placing the path as the lib64 in the nrpe.cfg
sample file , while all other packages use the local path - this is
something that you get used to "ignore-modify " with time .<br>
<blockquote
cite="mid:02B80625-B645-4C15-A3CC-9D2679F206C2@salon.com"
type="cite">
<div>
<div>Guess I have to wait now a few mins before the host shows
in the panel? as its not showing now even after 10 mins.
Restarting xinetd shows this in /var/log/messages:</div>
<div><br>
</div>
<div><b>xinetd[22518]: bind failed (Address already in use
(errno = 98)). service = nrpe (</b>is this a bug or anything
in xinetd? I saw that referenced on a few other forums)</div>
<div><br>
</div>
<div>nrpe service is running and listening though on 5666</div>
</div>
</blockquote>
I have always run nrpe as a stand alone daemon , never with xinetd ,
to have better control on what is running and how it is run .<br>
<br>
From what i remember about xinetd , it has an internal "keep-alive"
mechanism that waits till all connections are terminated before it
releases a port (again , i may be wrong ) you need to stop it and
wait till the port is no longer shown as active , and then restart
it .<br>
<br>
Or move to a standalone daemon - that will allow you to control the
nrpe separately from any other service and better debug the issue.<br>
<br>
<br>
<br>
<pre class="moz-signature" cols="72">--
Never,Ever Cut A Deal With a Dragon
Next year I will be doing the London to Paris bike ride to
raise money for the DogTrust (<a class="moz-txt-link-abbreviated" href="http://www.dogstrust.co.uk">www.dogstrust.co.uk</a>) .
Please Sponsor me at <a class="moz-txt-link-freetext" href="http://www.justgiving.com/Assaf-Flatto">http://www.justgiving.com/Assaf-Flatto</a></pre>
</body>
</html>