Investigate/Resolve API Gateway Flakiness

[brianlee731]

API Gateway (e.g. healthcheck) stops working periodically, and needs to be "kicked". Why is this? What needs to be fixed? Are the CORS settings in the API Gateway proper? Should the shared services procedure for setting up the API gateways always be run via terraform?

[jl-0]

Perhaps some relevant background. Today we had some issues with the health endpoint in Ops, upon investigation we saw an error code of 500 returned. Reviewing the logs for the gateway the following message was seen

Execution failed due to configuration error: No match for output mapping and no default output mapping configured. Endpoint Response Status Code: 200

After re-deploying the API gateway things seemed to work again (with no other configuration changes).

Perhaps setting VPC proxy integration is something to think about, Is there a reason we'd need to map the responses or just pass it through?