Nagios Support Forum

Hi,

On NagiosXI 2014R2.6 I have just enabled the guist snapshot check in the VMWare monitoring wizard.

I am generating an warning for a snapshot that is more than 1 day old and a critical for a snapshot older than 2 days.
This works ok, with one issue: for some reason the last alert aborted saying the service was "flapping between stats". So we did not receive an e-mail, but the error was shown on our screen. However: how can the snapshot be flapping? It cannot be 1 day old first, than go back to being younger... Is there a fix for this? (Besides turning off flapping detection.)

What is the wizard that you are using? This cannot be the "default" vmware wizard, that comes with XI.

In XI, find the Service and click on it
There are four icons at the top
The second icon is "View Service Notifications"
The third icon is "View Service History"

For both of these icons:
Click the icon
Change the Period to This Week
Click Update
Take a screenshot

Please show us both screenshots.

Hi,

I have added the attachments below:
I just noticed the service was still in a "flapping" state. Eventhough it has not changed in some time...

So the flapping that is occurring is a Nagios behaviour, nothing to do with the actual snapshots themselves.

In Nagios, flapping detects when a service transitions from an OK state to a HARD warning/critical/unknown and then back to an OK state. When this behaviour occurs more than 20% of the time the service has been OK, then it enters a flapping state.

Unigarant wrote:I have just enabled the guist snapshot check in the VMWare monitoring wizard.

So if you have just added a new service, then it's OK state isn't going to be that long.



From your screenshot you can see that it has gone from OK to warning/critical/unknown and then back to an OK state a couple of times. So in this instance flapping is occurring.

Flapping can be disabled on a service by service basis.
Find the service in CCM
Edit it
On the Check Settings tab you can define the flap detection settings.

Box293 wrote:So the flapping that is occurring is a Nagios behaviour, nothing to do with the actual snapshots themselves.

In Nagios, flapping detects when a service transitions from an OK state to a HARD warning/critical/unknown and then back to an OK state. When this behaviour occurs more than 20% of the time the service has been OK, then it enters a flapping state.
...

From your screenshot you can see that it has gone from OK to warning/critical/unknown and then back to an OK state a couple of times. So in this instance flapping is occurring.

Great, thank you for your reply. I had already disabled flapping detection for all guest snapshots. I forgot that it was actually a percentage...

So in this case there is no issue. I'll still leave flapping detection off, I don't think it really serves a purpose for our snapshot checks

I assume then that we are clear to close this thread?