Is it possible to create "Host lost" discovery triggers without scanning an IP range?
Currently we have hundreds of machines that come up and go down throughout the day. When the machines go down, they sometimes interfere with aggregate checks on their group -- their stale data is still counted in the aggregate check on the group.
What I'd like to do is to tell Zabbix that if it has not received a heartbeat from this machine in N minutes, then to disable the host and remove it from Zabbix. Ideally I'd be able to specify the group.
The machines are all added using auto-registration.
Edit: Some additional information
I'm essentially trying to configure a "host lost" discovery status for machines without scanning an IP range. We're unable to do ip scans because the machines are hosted on AWS and use auto-registration exclusively. Is there another way to do this?
Currently we have hundreds of machines that come up and go down throughout the day. When the machines go down, they sometimes interfere with aggregate checks on their group -- their stale data is still counted in the aggregate check on the group.
What I'd like to do is to tell Zabbix that if it has not received a heartbeat from this machine in N minutes, then to disable the host and remove it from Zabbix. Ideally I'd be able to specify the group.
The machines are all added using auto-registration.
Edit: Some additional information
I'm essentially trying to configure a "host lost" discovery status for machines without scanning an IP range. We're unable to do ip scans because the machines are hosted on AWS and use auto-registration exclusively. Is there another way to do this?