I had an interesting situation occur this past weekend. We have a host with a maintenance period for a service that has a scheduled restart. During this maintenance period the service failed to restart correctly causing an alert to be triggered once the maintenance was over. Shortly afterwards the service started and the host recovered and was removed from the dashboard, however the alerts were still escalated. These escalated alerts continued even after the serviced had recovered itself. The alerts only stopped once the event was acknowledged. Has anyone seen this before or know of a work around for this issue? I'm using Zabbix version 1.8.15.