Hi David

> How can I configure it such that an *event* only occurs after two consecutive failures of that plugin
It'll piece of cake for Zabbix, but not for Zenoss :-)

If you want to create *notification* after 2 two consecutive failures(events), then it'll be easier (notification delay). But you want to depend on failures. Events are "states" for Zenoss, so I don't see easy solution in Zenoss. Even you implement it in your check script.

IMHO: Severity event escalation/transformation based on event count should be best Zenoss solution for you. So first event will have only INFO severity, but next in row will have CRITICAL severity.
http://community.zenoss.org/thread/8081
https://gist.github.com/cluther/1486570
I don't like this type of transformation, because it increases DB overhead. But if you have just ~70 devices it's OK.

Devops Monitoring Expert advice: Dockerize/automate/monitor all the things.

DevOps stack: Docker / Kubernetes / Mesos / Zabbix / Zenoss / Grafana / Puppet / Ansible / Vagrant / Terraform / Elasticsearch