Soft State - handled as Normal

[enterdavertex]

Hi there , We have a machine that is going down too many time (Internet problem for this host). I was looking to get an option to set this machine as normal so that he won't pop into Nagios as Critical. If this is possible i would really like to know this options.

Thank you very much in advance !

[abrist]

You could use the negate plugin to change critical to warning or ok. This will change the availability metrics/historical data for the host, because future checks will be seen as ok instead of critical.

[enterdavertex]

You could use the negate plugin to change critical to warning or ok. This will change the availability metrics/historical data for the host, because future checks will be seen as ok instead of critical.

Where can i get this plugin, i searched on Nagios Exchange & the library and it seems it cannot be found..

[tmcdonald]

It should be /usr/local/nagios/libexec/negate

Code: Select all

Usage:
negate [-t timeout] [-Towcu STATE] [-s] <definition of wrapped plugin>

negate -h for more help output.

[enterdavertex]

I've read about this way to proceed, is there a way to manage if the host is not recheable in like 40 check then after if change status. Otherwise the only way to do is that the soft state are considered as Ok even if the host is not present.

[sreinhardt]

negate does not offer the ability to change the state to its original state after x number of retries. It only works off the current check results and does not store data about previous ones. Yes and no, what you should really be doing is using flapping conditions for that host\service. Otherwise a soft state for your 40 some retries, so that you can extend the time between hard ok and hard warn\crit will have a similar effect of limiting notifications.