Flip Flop Threshold in 3.1

[rnovoa]

Like it

Up

0

Down

Drop it

::

Hi summa_sp,

When you set a flip flop threshold x for a module, its status will not change until the status change condition is met x+1 times. It has nothing to do with alerts.

¿Did you configure your alert to fire when the ICMP module returns 0, or when the module becomes critical?

[summa_sp]

Like it

Up

0

Down

Drop it

::

The alerts is to set fire on critical status.  Would it work correctly if I set it to fire on 0?  I did some packet capture, and if FF is set to 5, the module changes status when it fails 5 times in total, not when it fails 5 times in a row.  The help file for FF indicates that it needs to fail 5 times in a row.

The FF help file is shown below.  Am I not understanding it correctly?  If the module changes state after 5 missed ping over something like 5 or 10 hours, regardless of when they happen, all that does is delay the amount of time it will take for the module to change states.  I need it to change state only when the module misses 5 times in a row.

“Take a classic example: A ping to a host where there is packet loss. In an environment like this, might give results as:

1 
1 
0 
1 
1 
0 
1 
1 
1

However, the host is alive in all cases. What we really want is to tell Pandora that until the host does not say you are at least three times down, not marked as such, so that in the previous case and would never be dropped, and only in this case it would be:

1 
1 
0 
1 
0 
0 
0”

[rnovoa]

Like it

Up

0

Down

Drop it

::

Hi summa_sp,

You got it right, the alert should be configured to fire when the module becomes critical. But I can’t reproduce the behavior you describe. Also, if you set FF to 5, the module should become critical when it gets the 6th consecutive 0 (not the 5th).

Did you upgrade your Pandora FMS Server to 3.1? Can you post the module configuration from the Pandora FMS database (tables tagente_modulo and tagente_estado)?