[Nagiosplug-help] Cannot map "tcp" to protocol number - nrpe

[Taylor, Richard, VF UK - Technology (TS)]

> Some help for a newbie would be appreciated.
> 
> I am running nagios 1.3 on a host called beamish ( SunOS beamish 5.8 Generic_108528-14 sun4u sparc SUNW,Ultra-Enterprise ), and trying to get check_nrpe to work to another machine ( 10.98.65.92 ).
> 
> $ bin/nagios etc/nagios.cfg
> 
> Nagios 1.3
> Copyright (c) 1999-2004 Ethan Galstad (nagios at nagios.org)
> Last Modified: 11-17-2005
> License: GPL
> 
> Nagios 1.3 starting... (PID=21051)
> 
> I have compiled nrpe-2.0, both with all defaults, and with --disable-ssl, but still get the error 'Cannot map "tcp" to protocol number'.
> 
> I have even started up the nrpe daemon on host beamish - again with all defaults, and get the entry below in /var/adm/messages :-
> 
> Dec  5 22:25:50 beamish nrpe[12756]: [ID 601491 daemon.notice] Starting up daemon
> 
> Which indicates that the daemon has started OK.
> 
> If I "telnet localhost 5666", then I get connected OK, and when I disconnect I get the entry below in /var/adm/messages :-
> 
> Dec  5 22:52:03 beamish nrpe[20274]: [ID 421412 daemon.error] Could not read request from client, bailing out...
> 
> I do not know whether this is expected or not.
> 
> If I run check_nrpe with an invalid hostname, then I get a meaningful message, but get the 'Cannot map "tcp" to protocol number' error when I put in a valid hostname - including localhost and 127.0.0.1
> 
> root at beamish# ./check_nrpe -H beamishh
> Invalid host name 'beamishh'
> root at beamish# ./check_nrpe -H beamish
> Cannot map "tcp" to protocol number
> root at beamish# netstat -a | grep 5666
>       *.5666               *.*                0      0 24576      0 LISTEN
> 
> I don't *think* this is a problem with the nagios config, but am struggling to work out what it it.
> 
> Many thanks,
> 
> Richard
>