Amplify-cli: Amplify push fails after adding amplify auth

[jtan25]

Describe the bug
Amplify push fails after adding amplify auth

To Reproduce
Steps to reproduce the behavior:

1. amplify add auth
2. I just used default properties
3. amplify push
4. Resource is not in the state stackUpdateComplete, CREATE_FAILED

Screenshots
CREATE_FAILED testauth AWS::CloudFormation::Stack Sat Sep 22 2018 19:04:37 GMT-0400 (Eastern Daylight Time) Parameters: [authRoleArn, autoVerifiedAttributes, allowUnauthenticatedIdentities, smsVerificationMessage, userpoolClientReadAttributes, mfaTypes, emailVerificationSubject, useDefault, openIdLambdaIAMPolicy, userpoolClientGenerateSecret, mfaConfiguration, userpoolClientLogPolicy, openIdRolePolicy, identityPoolName, openIdLogPolicy, thirdPartyAuth, authSelections, smsAuthenticationMessage, roleExternalId, mfaLambdaLogPolicy, passwordPolicyMinLength, userPoolName, openIdLambdaRoleName, userpoolClientWriteAttributes, policyName, userpoolClientName, userpoolClientLambdaPolicy, resourceName, mfaLambdaIAMPolicy, mfaPassRolePolicy, emailVerificationMessage, userpoolClientRefreshTokenValidity, userpoolClientSetAttributes, unAuthRoleName, authRoleName, requiredAttributes, roleName, passwordPolicyCharacters, lambdaLogPolicy, unAuthRoleArn, userpoolClientLambdaRole, defaultPasswordPolicy, mfaLambdaRole] must have values

I have a parameters.json with filled in values and a .yml created in backend/auth but for some reason I am unable to do amplify push.

[kaustavghosh06]

Could you please check which region you’re deploying to?

[jtan25]

I am deploying to us-east-1 N.Virginia

[UnleashedMind]

You might have reached the limit of certain AWS services:
Please check the number of S3 buckets, Cognito UserPools, Cognito Identify Pools, and Cloudformation stacks.

[haverchuck]

Is the error ouput you posted from the CLI? Is there any additional output in cloudformation?

[jtan25]

Yes the error is from the CLI, nothing additional on cloudformation. I just re-initialized the project and migrated my code over, works now.

[kaustavghosh06]

@jtan25 This is an intermittent Cloudformation issue which we've seen. Closing this issue for now.

[ju-Skinner]

I know this issue is closed but is there a fix on the horizon? I've looked at the cloud formation for the API, I see there are dependencies on the Auth template, but if it's present shouldn't it just read the values from the auth parameters.json; followed up with skipping the attempt to update the auth stack.

I can take the API Cloudformation template that is stored in S3 and pass the necessary params in Cloudformation and have the stack successfully be created. However, doing so I lose the option to update the GraphQL schema by using the amplify push api feature.

[tobinbc]

FWIW, I encountered this problem, I believe after the project directory name was changed and it did not pickup the parameters.json files. Fix was to copy contents of both amplify-meta.json files, do amplify init again, then put original amplify-meta.json files back. It won't remove the existing defined resources.

[babzcraig]

Encountered the same issue.To solve, I just reran amplify-init without copying over anything. Solved the issue and persisted all my data correctly.

[thelinguist]

TL;DR I had the same problem. If anyone else runs into this error, I suggest updating amplify-cli to the latest version.

I just installed amplify and tried it out. I got the same error. I found an error in the cloudFormation console. Under the stack that failed -> events, there's an error that says:

The runtime parameter of nodejs6.10 is no longer supported for creating or updating AWS Lambda functions. We recommend you use the new runtime (nodejs8.10) while creating or updating functions. (Service: AWSLambdaInternal; Status Code: 400; Error Code: InvalidParameterValueException; Request ID: da65986f-7c36-11e9-9bcc-cfe59ec061e8)

So I went into my project, in the file: /amplify/backend/auth/<projectName>/<authID>-cloudformation-template.yml and replaced all instances of nodejs6.10 with nodejs8.10. Now it works!

I'm using amplify-cli v 1.6.10 on mac. Before I used 1.1 I believe. If anyone else runs into this problem, I think the issue is simply upgrading your amplify-cli. At this point in time, the lambdas were using node.js6.10 but has since been updated