[Nagios-users] RE: HELP

[Carroll, J. P [Contractor] <jca...@sp...>]

I'm tempted to say that there's already something listening on that port.  I
would also say that while the nrpe daemon will bind to a port, I don't
believe you can get it to bind to a specific socket.  I'm not aware of any
way to get xinetd to force nrpe to bind to a specific socket, either.

Is it possible you have the daemon running standalone while you're also
trying to get xinetd to manage it?

Is it possible you have two files in your /etc/xinetd.d/ directory, one
named nrpe, one named something else?  I helped someone else a month or two
ago, and he was suffering much the same way, and it was stumping both of us.
In the end, he discovered a file in /etc/xinetd.d/ named something like "-i"
or perhaps ":".  He wasn't really looking for it, so it went unnoticed for
quite a while.  You might want to try changing to the /etc/xinetd.d/
directory, then do a "grep -l nrpe `ls -a`" and see what all pops up.

In case I haven't been clear, it really does seem like something else has
already grabbed that port, and that shouldn't be the case.  That would agree
with the strace output you sent me.

jc

> -----Original Message-----
> From: Kaplan, Andrew H. [mailto:AHKAPLAN@PARTNERS.ORG]
> Sent: Thursday, January 09, 2003 3:19 PM
> To: Carroll, Jim P [Contractor]
> Cc: Nag...@li...
> Subject: HELP
> 
> 
> The server has two network jacks attached to it. One jack is on a
> 192.168.1.0 network while the other is on
> a 132.183.12.0 net. I am still getting the Network server 
> bind failure (98:
> Address already in use) error
> message. Because I am trying to use xinetd.d, is there a line 
> that would
> tell nrpe to bind to the 132 network?
> I am grasping at straws here, but I can't think of anything 
> else at this
> time.
>