check_ping works in ssh but not on web-gui?

[staun]

Is there a difference on the check_ping call when running the command from ssh vs. webgui?

It seems that mine fails on the WebGUI.

from SSH
[root@srv-mon-nsb1 libexec]# ./check_ping -H srv-files-nsb1 -w 10,100% -c 50,100%
PING OK - Packet loss = 0%, RTA = 0.41 ms|rta=0.415000ms;10.000000;50.000000;0.000000 pl=0%;100;100;0

from WebGUI
COMMAND: /usr/local/nagios/libexec/check_ping -H srv-web-nsb1 -w 10,100% -c 50,100% -p 5
OUTPUT: PING CRITICAL - Packet loss = 100%|rta=50.000000ms;10.000000;50.000000;0.000000 pl=100%;100;100;0

[sreinhardt]

Just to verify, are you getting this output from the test command button or by letting the command run via normal nagios scheduler? Otherwise to my knowledge there is not a difference, unless selinux or something does not give the executing user permission to exit the system with a ping.

[staun]

This is run from a SSH command line when it has success with the PING.

[abrist]

Is selinux enabled?

Code: Select all

getenforce

Can you run the check from the cli as user apache?

Code: Select all

su - apache -c "./check_ping -H srv-web-nsb1 -w 10,100% -c 50,100% -p 5" 

Have you tried configuring the check in XI and letting the nagios scheduler run the check normally (rather than using the test check command button).

[staun]

hmm cannot run this command with the user apache.

Code: Select all

[root@srv-mon-nsb1 libexec]# su - apache -c "./check_ping -H srv-web-nsb1 -w 10,100% -c 50,100% -p 5"
This account is currently not available.

[lmiltchev]

Run the following commands and show the output:

Code: Select all

ll /bin/ping
ll /usr/local/nagios/libexec/check_ping