Zabbix traps and interface flaps
We've currently configured LLD on a interfaces for Cisco devices.
A trap item exists with a corresponding trigger. When an interface drops and the device in question sends a trap notification the trigger is set off (moves to PROBLEM state) and, in our particular setup, the corresponding action is for PagerDuty to call the on-call engineer.
This works well, however when we have an interface on the network that is flapping, the engineer in question gets inundated with calls every time the interface goes up and comes down.
Does Zabbix have the functionality to do something like this:
"If this trigger has been in a PROBLEM state anytime in the past x seconds, then suppress any further triggering until said timer expires" ?
This way if we set the timer to 20 minutes (for example), the engineer would at most get contacted every 20 minutes rather than every 1 minute (or however frequently the flap is taking place).
A trap item exists with a corresponding trigger. When an interface drops and the device in question sends a trap notification the trigger is set off (moves to PROBLEM state) and, in our particular setup, the corresponding action is for PagerDuty to call the on-call engineer.
This works well, however when we have an interface on the network that is flapping, the engineer in question gets inundated with calls every time the interface goes up and comes down.
Does Zabbix have the functionality to do something like this:
"If this trigger has been in a PROBLEM state anytime in the past x seconds, then suppress any further triggering until said timer expires" ?
This way if we set the timer to 20 minutes (for example), the engineer would at most get contacted every 20 minutes rather than every 1 minute (or however frequently the flap is taking place).
Comment