<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><br><div><div>On Nov 10, 2010, at 12:13 PM, Jonathan Wiggins wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">I'm stuck troubleshooting an error message in the Service Status Details, where a monitored host is being reported as "No Route to Host".<div><br></div><div>I can do remote plugin tests from the Nagios server to the monitored node without any problems, but referencing this page: <a href="http://www.troubleshootingwiki.org/Troubleshooting_Nagios_3.0">http://www.troubleshootingwiki.org/Troubleshooting_Nagios_3.0</a> I see a paragraph about testing SSH key-exchange. </div><div><br></div><div>Nagios Server = 10.0.100.130</div><div>monitored node = 10.0.100.143</div><div><br></div><div>when I do an "ssh -vvvv 10.0.100.130" from the monitored node, it goes straight through, no login prompt. When the reverse is attempted: "ssh -vvvv 10.0.100.143" from Nagios box, I consistently get prompted for login. I was recommended to use DSA instead of RSA, but that made no difference. </div><div><br></div><div>I've appended the contents of both id_rsa.pub and id_dsa.pub to the "authorized_keys" (and "known_hosts") files the exact same way to and from each box respectively. Of course my tests on 2 test systems worked perfectly, but I'm missing something with this setup.</div><div><br></div><div>Additionally, if I do an NRPE check from Nagios > monitored node, it completes successfully.</div><div><br></div><div><div><i>/usr/local/nagios/libexec/check_tcp -H 10.0.100.143 -p 5666</i></div><div><i>TCP OK - 0.000 second response time on port 5666|time=0.000361s;0.000000;0.000000;0.000000;10.000000</i></div></div><div><br></div><div><b>checking logs on monitored node:</b></div><div><br></div><div><div>grep nrpe /var/log/messages</div><div><br></div><div><br></div><div>/var/log/messages:Nov 10 12:01:26 monitorednode xinetd[5672]: START: nrpe pid=8607 from=10.0.100.130</div><div>/var/log/messages:Nov 10 12:01:26 monitorednode nrpe[8607]: Error: Could not complete SSL handshake. 5 </div><div>/var/log/messages:Nov 10 12:01:26 monitorednode xinetd[5672]: EXIT: nrpe status=0 pid=8607 duration=0(sec)</div><div>/var/log/messages:Nov 10 12:03:29 monitorednode xinetd[5672]: START: nrpe pid=9220 from=10.0.100.130</div><div>/var/log/messages:Nov 10 12:03:29 monitorednode nrpe[9220]: <b>Error: Could not complete SSL handshake</b>. 5 </div><div>/var/log/messages:Nov 10 12:03:29 monitorednode xinetd[5672]: EXIT: nrpe status=0 pid=9220 duration=0(sec)</div></div><div><br></div><div>"Error could not complete SSL handshake" seems to confirm my suspicions about the SSH key-exchange</div><div><br></div><div><br></div></div></blockquote><br></div><div><br></div><div>Ok, i've solved the SSH key-exchange part of my problem (criss-crossed id_rsa.pub keys or something) - so I have that piece figured out. But the "No Route to Host" message is still visible in Service Host Details for my monitored node.</div><div><br></div><div>so.. am able to run the following against remote host:</div><div><br></div><div><div>/usr/local/nagios/libexec/check_tcp -H 10.0.100.143 -p 5666</div><div>TCP OK - 0.000 second response time on port 5666|time=0.000361s;0.000000;0.000000;0.000000;10.000000</div><div><br></div><div>and this:</div><div><div>ssh 10.0.100.143 /usr/local/nagios/libexec/check_procs </div><div>PROCS OK: 603 processes</div><div><br></div></div></div><div>I see this in the messages files:</div><div><br></div><div><div><b>Nov 9 00:00:00 nagiosbox nagios: CURRENT SERVICE STATE: monitorednode;Home Page;CRITICAL;HARD;1;No route to host </b></div><div><b>Nov 10 00:00:00 <span class="Apple-style-span" style="font-weight: normal; "><b>nagiosbox</b></span> nagios: CURRENT HOST STATE: monitorednode;UP;HARD;1;PING OK - Packet loss = 0%, RTA = 0.21 ms </b></div><div><b>Nov 10 00:00:00 <span class="Apple-style-span" style="font-weight: normal; "><b>nagiosbox</b></span> nagios: CURRENT SERVICE STATE: monitorednode;Home Page;CRITICAL;HARD;1;No route to host</b></div><div><b><br></b></div><div>which looks like there is no packet loss on the PING to the host, but then it shows No Route To Host</div><div><br></div><div><br></div><div>Thanks again for the assistance.</div></div><br></body></html>