Hi,
I've configured a CloudWatch integration so that CloudWatch alarms are delivered to OpsGenie. I have a CloudWatch alarm sending alarm events to SNS topic and I've configured OpsGenie CloudWatch integration endpoint as an HTTPS subscriber.
Now I would like to configure a constant monitoring of this integration to ensure that it works especially at the moment where a severe incident happens and CloudWatch alarm needs to be delivered to OpsGenie so somebody is paged.
My idea was to configure a "cron" task to publish a "ping" message event periodically on the same SNS topic so it's delivered to OpsGenie as a heartbeat. I want OpsGenie to treat these events as heartbeats and page me when a heartbeat message is not received for some longer period of time (as it might indicate that alarm delivery from AWS to OpsGenie doesn't work properly). Is it possible?
I know there is a feature called HeartBeat Monitoring. I thought that maybe I could create an heartbeat endpoint and configure it as another HTTPS subscriber on the same SNS topic. However, heartbeat endpoint uses a different "protocol" - it requires HTTP authorisation header (whereas CloudWatch integration includes the API key in the URL).
Am I missing something? How can I make sure that the alarm delivery works and if it stops working I get paged by OpsGenie?
Best regards,
Piotr