The alert system wasn't designed with selective metric checking.

You can only call to check an entire entity, not a single value. If you call check_entity with only the operstatus, it'll make a mess of the state for other metrics.

This may be addressed in future with different alerting pathways outside of the existing alert test framework.

Adam.

Sent from BlueMail

On Apr 11, 2016, at 15:52, Kevin Wormington <kworm@missouri-telecom.com> wrote:
Hi,

Is it possible to run the alert checks outside of the normal poll cycle?  I have a simple perl script that receives SNMP traps and calls the snmptrap.php observium script and the interface is correctly marked down/up but the alert does not happen until the next poll cycle.

Thanks,

Kevin

observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium