One host unable to report data to Nagios XI

[sathevaner]

We have one host that has refused to report itself as "Up" since October 2019. The error we have been unable to reproduce is, "CRITICAL - (IP address redacted): Time to live exceeded in transit @ (IP address redacted). rta nan, lost 100%"

How do I troubleshoot this?

[benjaminsmith]

Hi,

Let's start by opening up a terminal and pinging the host to see if you get a reply. For example,

Code: Select all

ping -c 5 <ip address>

If that works but the check is failing in Nagios XI, please post the full check command to the thread so we can review it. The following guide has instructions on how to find the check command and test it from the command line in XI.

Nagios XI - How To Test Check Commands From The Command-line

Also, not sure if this is the reason or not, but ICMP is disabled by default on some servers and a ping check will fail even though the host is up.
See:
https://serverfault.com/questions/51173 ... 2-instance

[sathevaner]

Benjamin,

This is what's happening on the Nagios XI server:

Code: Select all

[root@host libexec]# ./check_icmp -H (IP address redacted) -w 3000.0,80% -c 5000.00,100%,-p 8
CRITICAL - (IP address redacted) lost 0% >= 0% :: 8: rta nan, lost 100%|(IP address redacted)rta=11.014ms;3000.000;5000.000;0; (IP address redacted)pl=0%;0;0;0;100 (IP address redacted)rtmax=12.877ms;;;; 10.71.247.1rtmin=9.833ms;;;; 8rta=0.000ms;3000.000;5000.000;0; 8pl=100%;0;0;0;100 8rtmax=0.000ms;;;; 8rtmin=0.000ms;;;; 

and doing a ping to the same host is successful:

Code: Select all

[root@host libexec]# ping -c 5 (IP address redacted)
PING (IP address redacted) ((IP address redacted)) 56(84) bytes of data.
64 bytes from (IP address redacted): icmp_seq=1 ttl=252 time=10.4 ms
64 bytes from (IP address redacted): icmp_seq=2 ttl=252 time=9.40 ms
64 bytes from (IP address redacted): icmp_seq=3 ttl=252 time=9.94 ms
64 bytes from (IP address redacted): icmp_seq=4 ttl=252 time=10.4 ms
64 bytes from (IP address redacted): icmp_seq=5 ttl=252 time=10.3 ms

Please advise.

[benjaminsmith]

Hi,

Looking over the check command, there is a slight error in the syntax, it should be as follows:

Code: Select all

./check_icmp -H (IP address redacted) -w 3000.0,80% -c 5000.00,100% -p 8

That is, without the comma between 100% and -p. Test that out on the command line, then go to the CCM in Nagios XI and verify the syntax on the check command, and let me know if the issue is resolved.

Reference:
Using The Core Config Manager For Host Management

[sathevaner]

This thread may now be locked. Apparently all that was needed as "Force an immediate check" despite being up this whole time.

[benjaminsmith]

This thread may now be locked. Apparently all that was needed as "Force an immediate check" despite being up this whole tim

Thanks for following up. If the issue comes back, feel free to open another thread.

Benjamin