Windows Servers Loose Ping Packets

[pteegarden]

I have pointed out to our Microsoft Windows Administrators that Nagios does not like the way that Windows machines respond to pings: most of the time response is fairly rapid and complete (no loss of packets). On regular intervals, however, any windows server will fail the ping checks and return the following message:

SERVICE ALERT: <hostname>;Ping;WARNING;SOFT;1;WARNING - xxx.xxx.xx.xx: rta 2348.401ms, lost 80%

True, this only lasts a short time, and is usually good on the next check.

But...

This also shows up "yellow" on both the Operations Center screen and the Service Detail screen.

Is there a way to adjust the ping service check for the windows machines to tollerate this abhorent behaviour?

[abrist]

Is there a way to adjust the ping service check for the windows machines to tollerate this abhorent behaviour?

(emphasis mine)

Yeah, you can edit the thresholds to something like:

Code: Select all

-w 3000.0,80% -c 5000.0,100%

[pteegarden]

And where/how do I do this?

If I look at the Service Status Detail"/"ping"/"configure"/"reconfigure this service"/"monitoring tab", the "Monitor the service with this command..." box is empty - no command - no parameters to modify.

If I look at the service using CCM Service Management "Common Settings" tab, the Check command box is empty - no command - no parameters to modify.

[abrist]

You are most likely using a template. Check the template, and change the values there if you wish. Do understand that changing the values on the template will cause the changes to propagate to *all* hosts running this check.

Alternatively, you could remove the template, create a new one, and apply that to the problematic hosts (best option), create a new check and apply them to the hosts in question (not a bad option either).

[pteegarden]

I think a template was supplied for me when I added the windows boxes using the configuration wizard.
Here is a representitive sample from the /usr/local/nagios/etc/services subdirectory:

Code: Select all

define service {
        host_name                       bay
        service_description             Ping
        use                             xiwizard_windowsserver_ping_service
        max_check_attempts              5
        check_interval                  5
        retry_interval                  1
        check_period                    xi_timeperiod_24x7
        notification_interval           60
        notification_period             xi_timeperiod_24x7
        notifications_enabled           1
        contacts                        tellpaul
        contact_groups                  Windows
        _xiwizard                       windowsserver
        register                        1
        }

You might understand why I don't feel comfortable messing/replacing with something labeled "xiwizard_windowsserver_ping_service"

[pteegarden]

Using the CCM, I have modified the xiwizard_windowsserver_ping_service template.
is the change picked up immeadiately or do I have to restart the nagios service? or some other service?

[sreinhardt]

You do have to restart the core engine after altering CCM values. However if you have applied configuration after making and saving the changes it should be restarted already.

[pteegarden]

Thanks.