Nagios Host Down but Services Up

binkocd · 02-03-2012, 03:08 PM

Information Before hand:
Ubuntu Server 10.4 x64
Nagios 3.2.0

When I got to Host Detail it shows that the host is down (red) and when I click on the host all the checks show green and the host is show as down with "Warning limit is greater than Critical limit" as the status information. I have cleared the log_file, object_cache_file, log_archive_path, and state_retention_file. Only other thing I was thinking was maybe there was something in the mysql db, but wasn't sure where to look. Any help would be greatly appreciated. Thank you.

arashi256 · 02-04-2012, 07:33 PM

Isn't whether the host is up or down dictated by ping requests via check_ping? Then I would guess that either your host is rejecting ping packets buy accepting check_nrpe requests on port 5666. Either that or the Host Detail page is cached in your browser, but more likely the first option.

chrism01 · 02-05-2012, 07:01 PM

Quote:

"Warning limit is greater than Critical limit"

This means what it says; somehow you/someone has edited the check_ping and reversed those nums. Warn limit must be < Critical limit.

Here is an example from my (hostname excluded) setting in /usr/local/nagios/etc/objects/localhost.cfg

Code:

define service{
        use                             local-service         ; Name of service template to use
        host_name                       hostname_here
        service_description             PING
        check_command                   check_ping!100.0,20%!500.0,60%
        contact_groups                  linux-admins
        }

binkocd · 02-06-2012, 07:59 AM

Here's what I have for ping:

Code:

check_command   check_ping!50,20%!80,90%

When I was building the configs previously I did have the check_snmp_disk values in reverse order and it gave me that error for the service, but I have since cleared that up and it reports correctly. Could there be any way that that error has been cached somewhere on the server and just hasn't cleared out?

nasiruddin · 02-06-2012, 09:41 AM

check whether in your command.cfg the check-host-alive is defined properly or not.It should be something like the following

# 'check-host-alive' command definition
define command{
command_name check-host-alive
command_line $USER1$/check_ping -H $HOSTADDRESS$ -w 3000.0,80% -c 5000.0,100% -p 5
}