Example code just reconnects does not issue a usable error message

When running the example code, I’m getting errors but the client just reconnects and doesn’t issue any useful error messages.

For example:

node node_modules/aws-iot-device-sdk/examples/device-example.js -t 1  -f certs -d 4000 -D -T paul-pi3 -H a1vfc41l04drxa.iot.us-east-1.amazonaws.com

Just fails with

attempting new mqtt connection...
offline
connection lost - will attempt reconnection in 4 seconds...
close
reconnect
connection lost - will attempt reconnection in 8 seconds...
close

When I hook up CloudWatch, I get a semi-usable error message:

2017-12-09 06:32:12.657 TRACEID:636c2ee2-3f87-58c6-f18c-4248f3190037 PRINCIPALID:4d1552d14416ab4198353b308538604794bcc3a0aa03c79050dd61d3afc20ee9 [ERROR] EVENT:MQTT Client Connect MESSAGE:Connect Status: AUTHORIZATION_ERROR

The client should be fixed to issue usable error messages. It takes more undocumented steps to hook up cloud watch, making “Hello World” scenarios much more difficult to get going.

It would be helpful if the error messages also suggested remediation steps. I have keys issued according to the IOT wizard, so I don’t know what to do next…

[paul-cipherlex]

After adding this policy, of which I could not find as an example in the documentation, the above example works:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "iot:Publish",
        "iot:Connect",
        "iot:Receive",
        "iot:Subscribe",
        "iot:GetThingShadow",
        "iot:DeleteThingShadow",
        "iot:UpdateThingShadow"
      ],
      "Resource": [
        "*"
      ]
    }
  ]
}

[liuszeng]

Hi @paul-cipherlex ,

I have created an internal tracker for this documentation update request. It will be included in the future releases. Thank you again for your input.

Thanks, Liusu