This is normally a networking setup issue.

Log in to your vm and see if you can ping your server running heat-api-cfn. If you can't you need to solve that problem.

Are you using nova-networking or neutron?

I'll explain how this works for the record:

1. when you configure heat there is a config option, it should be something like: heat_waitcondition_server_url = http://<heat-api-cfn ip="">:8000/v1/waitcondition</heat-api-cfn>
2. you create a stack with a waitcondition in it.
3. some metadata is saved in heat for the waitcondition
4. an ec2signed url is passed to the vm in the userdata
5. when cfn-signal is called it does a PUT to the ec2signed url
6. heat-engine modifies the waitcondition metadata.

So the only things you can really check are:

• is heat_waitcondition_server_url set correctly
• has cfn-signal been called
• did the call make it to heat-api-cfn (check the logs)

As I said earlier, if you have this issue it is normally because cfn-signal was called but it did not make it to heat-api-cfn because of networking issues.

-Angus