I have child nodes that have actions that employ escalations like this for example:

detect a service is down: restart the service, email operations
15 minutes later: page the on-call guy if it's still a problem
30 minutes later: page his boss
etc...

In the dashboard, I see the actions that happen immediately but the actions at 15 minutes and 30 minutes appear to be 'in progress' on the master. On the slave, the actions on the dashboard are accurate.