Deploy to AWS Beanstalk fails: remove environment from a VPC

danbrellis

I am able to successfully deploy to my dev environment from my CLI with `eb deploy`.

I have set up a pipeline, which successfully zips the app and uploads it, but Beanstalk fails to deploy.

Relevant Beanstalk Events:

ERROR Failed to deploy application.
ERROR Missing value for required parameter: InstanceType
ERROR Missing value for required parameter: InstanceTypeFamily
ERROR You cannot remove an environment from a VPC. Launch a new environment outside the VPC.
ERROR "option_settings" in one of the configuration files failed validation. More details to follow.

In my .ebextensions I have option_settings:

option_settings: 
  aws:autoscaling:launchconfiguration:
    InstanceType: t3.medium
    SecurityGroups: sg-1111111111
    IamInstanceProfile: arn:aws:iam::1111111111:instance-profile/my-profile-IAMEBInstanceProfile-AAAAAAAAAAAAA
  aws:autoscaling:asg:
    MinSize: '2'
    MaxSize: '5'
  aws:ec2:vpc:
    VPCId: "vpc-1111111111"
    Subnets: "subnet-1111111111,subnet-1111111111"
    ELBSubnets: "subnet-1111111111,subnet-1111111111"
  aws:elbv2:listener:443:
    ListenerEnabled: 'true'
    Protocol: HTTPS
    SSLCertificateArns: arn:aws:acm:us-east-1:1111111111:certificate/1111111111

 

Here's the pipeline:

image: atlassian/default-image:2

pipelines:
branches:
master:
      - step:
name: "Zip Project Files"
          script:
            - zip -r application.zip *
          artifacts:
            - application.zip
      - step:
name: "Deploy to Production"
          deployment: production
script:
            - pipe: atlassian/aws-elasticbeanstalk-deploy:0.6.7
variables:
AWS_ACCESS_KEY_ID: $AWS_ACCESS_KEY_ID
AWS_SECRET_ACCESS_KEY: $AWS_SECRET_ACCESS_KEY
AWS_DEFAULT_REGION: $AWS_DEFAULT_REGION
APPLICATION_NAME: $APPLICATION_NAME
ENVIRONMENT_NAME: $ENVIRONMENT_NAME
ZIP_FILE: "application.zip"
                S3_BUCKET: $S3_BUCKET

$APPLICATION_NAME and $ENVIRONMENT_NAME match what's shown in AWS console.

I can provide output from Bitbucket pipeline log if helpful.

If anyone has insight into why the error is occurring/how to fix it. Not sure if I'm missing setting something explicitly. Would think since it works fine from my local cli it should work from the pipe with the same files. Thanks.

danbrellis

I figured this out. In my bitbucket-pipelines.yml in the script that creates the application.zip artifact, I was running a zip command that did not include hidden files/directories.

I replaced 

zip -r application.zip *

with

git archive -o application.zip master 

and everything worked as expected.

Thanks for the help.