Nagios Support Forum

Hi,

Just wondering if anyone knows how to ignore "socket timeout" errors (in general)?

I have a host sitting on a dodgy network that is not in my power to fix. I have a number of checks running on the host and I only want to know about errors with the actual check (when it connects successfully) and want to ignore any "socket timeout" errors.

Is this possible?

Thanks.

Hello,

There isn't a direct way to filter out "socket timeouts" but you can adjust the configuration settings to minimize the number of errors. For example,

1. Increase the timeout out threshold for the check command. Most plugins will have a timeout flag, increase this to -t 60.

2. Increase the number of check attempts and the re-try interval.

3. If necessary, enable flap detection. This will keep the last 21 check for the host or service, and you can set high and low thresholds for alerting. This feature is useful for checks that change too frequently.

Detection and Handling of State Flapping