Abnormal alert notification behavior

[eric.lin826]

Hi Nagios,
For example, in the "E-Form-Pro-Event Log" worksheet in the attached excel file, at 7:52, the first Host Down occurred.
The service "http://e-form.sunnybank.com.tw/" is directly converted to Hard Status when the condition of "max_check_attempts=5" is not met, and an alarm notification is issued at 7:57.
After the service is restored, No reinstatement notice has been issued. Is this normal?

[dchurch]

Here's the rationale for the current behavior:

- If a host is in a DOWN/UNREACHABLE state (including SOFT DOWN), non-OK services will immediately go into a HARD state.
- Because the service immediately goes into a HARD state, Core will try to send notifications for those services.
- At the same time, we want to suppress service notifications in the event that the host state is the cause of the problem (this is written in the documentation as "If the host is in hard non-OK state, notifications for services on this host won't be sent out").
- So, we suppress service notifications for SOFT DOWN hosts as well.

More on state types and logic.

[eric.lin826]

@dchurch Thank you for helping me clarify the problem.

Here's the rationale for the current behavior:

- If a host is in a DOWN/UNREACHABLE state (including SOFT DOWN), non-OK services will immediately go into a HARD state.
- Because the service immediately goes into a HARD state, Core will try to send notifications for those services.
- At the same time, we want to suppress service notifications in the event that the host state is the cause of the problem (this is written in the documentation as "If the host is in hard non-OK state, notifications for services on this host won't be sent out").
- So, we suppress service notifications for SOFT DOWN hosts as well.

More on state types and logic.

[benjaminsmith]

Hi,

Thank you for helping me clarify the problem.

Glad that helped! Did you have any other questions or may we close this thread?

[eric.lin826]

Sure, You can close this thread, thanks.