sourcetoad / aws-codedeploy-action Goto Github PK

View Code? Open in Web Editor NEW

35.0 18.0 13.0 260 KB

AWS CodeDeploy via GitHub Actions

License: MIT License

Shell 97.22% Dockerfile 2.78%

ec2 codedeploy github-action

aws-codedeploy-action's People

Contributors

Stargazers

Watchers

Forkers

robertobermejo grocerkey savroman sosolog bennettp123 d2l-tbecks william-brooks anand-parmar patriotsoftware ardensworkplace buzzlightyear726 corwindesign

aws-codedeploy-action's Issues

Failed to register layer: failed to Lchown "/dev/console"

I have an error using v1.11.0 and v1

Build container for action use: '/opt/actions-runner/_work/_actions/sourcetoad/aws-codedeploy-action/v1.11.0/Dockerfile'.
  /home/ubuntu/bin/docker build -t 456935:06d9fb7a8d1f40d481[2](https://github.com/abaqusinc/allgeo-server/actions/runs/9671528017/job/26682306923#step:3:2)2359352c46268 -f "/opt/actions-runner/_work/_actions/sourcetoad/aws-codedeploy-action/v1.11.0/Dockerfile" "/opt/actions-runner/_work/_actions/sourcetoad/aws-codedeploy-action/v1.11.0"
  DEPRECATED: The legacy builder is deprecated and will be removed in a future release.
              Install the buildx component to build images with BuildKit:
              https://docs.docker.com/go/buildx/
  
  Sending build context to Docker daemon  18[3](https://github.com/abaqusinc/allgeo-server/actions/runs/9671528017/job/26682306923#step:3:3).8kB
  
  Step 1/4 : FROM amazon/aws-cli:2.17.0
  2.17.0: Pulling from amazon/aws-cli
  [4](https://github.com/abaqusinc/allgeo-server/actions/runs/9671528017/job/26682306923#step:3:4)92439933c1f: Pulling fs layer
  659dcd9fe751: Pulling fs layer
  cadf92deee9d: Pulling fs layer
  42aa93e9233a: Pulling fs layer
  87a391e9688b: Pulling fs layer
  42aa93e9233a: Waiting
  87a391e9688b: Waiting
  6[5](https://github.com/abaqusinc/allgeo-server/actions/runs/9671528017/job/26682306923#step:3:5)9dcd9fe751: Verifying Checksum
  659dcd9fe751: Download complete
  42aa93e9233a: Verifying Checksum
  42aa93e9233a: Download complete
  87a391e9688b: Verifying Checksum
  87a391e9688b: Download complete
  492439933c1f: Verifying Checksum
  492439933c1f: Download complete
  cadf92deee9d: Verifying Checksum
  cadf92deee9d: Download complete
  failed to register layer: failed to Lchown "/dev/console" for UID 0, GID 0: lchown /dev/console: no such file or directory
  Warning: Docker build failed with exit code 1, back off 5.794 seconds before retry.
  /home/ubuntu/bin/docker build -t 456935:06d9fb7a8d1f40d48122359352c46268 -f "/opt/actions-runner/_work/_actions/sourcetoad/aws-codedeploy-action/v1.11.0/Dockerfile" "/opt/actions-runner/_work/_actions/sourcetoad/aws-codedeploy-action/v1.11.0"
  DEPRECATED: The legacy builder is deprecated and will be removed in a future release.
              Install the buildx component to build images with BuildKit:
              https://docs.docker.com/go/buildx/
  
  Sending build context to Docker daemon  183.8kB
  
  Step 1/4 : FROM amazon/aws-cli:2.17.0
  2.17.0: Pulling from amazon/aws-cli
  492439933c1f: Pulling fs layer
  [6](https://github.com/abaqusinc/allgeo-server/actions/runs/9671528017/job/26682306923#step:3:6)59dcd9fe[7](https://github.com/abaqusinc/allgeo-server/actions/runs/9671528017/job/26682306923#step:3:7)51: Pulling fs layer
  cadf92deee9d: Pulling fs layer
  42aa93e9233a: Pulling fs layer
  [8](https://github.com/abaqusinc/allgeo-server/actions/runs/9671528017/job/26682306923#step:3:8)7a3[9](https://github.com/abaqusinc/allgeo-server/actions/runs/9671528017/job/26682306923#step:3:9)1e9688b: Pulling fs layer
  42aa93e9233a: Waiting
  87a391e9688b: Waiting
  659dcd9fe751: Verifying Checksum
  659dcd9fe751: Download complete
  42aa93e9233a: Verifying Checksum
  42aa93e9233a: Download complete
  87a391e9688b: Verifying Checksum
  87a391e9688b: Download complete
  cadf92deee9d: Verifying Checksum
  cadf92deee9d: Download complete
  492439933c1f: Verifying Checksum
  492439933c1f: Download complete
  failed to register layer: failed to Lchown "/dev/console" for UID 0, GID 0: lchown /dev/console: no such file or directory
  Warning: Docker build failed with exit code 1, back off 4.595 seconds before retry.
  /home/ubuntu/bin/docker build -t 456935:06d9fb7a8d1f40d48122359352c46268 -f "/opt/actions-runner/_work/_actions/sourcetoad/aws-codedeploy-action/v1.[11](https://github.com/abaqusinc/allgeo-server/actions/runs/9671528017/job/26682306923#step:3:11).0/Dockerfile" "/opt/actions-runner/_work/_actions/sourcetoad/aws-codedeploy-action/v1.11.0"
  DEPRECATED: The legacy builder is deprecated and will be removed in a future release.
              Install the buildx component to build images with BuildKit:
              https://docs.docker.com/go/buildx/
  
  Sending build context to Docker daemon  183.8kB
  
  Step 1/4 : FROM amazon/aws-cli:2.[17](https://github.com/abaqusinc/allgeo-server/actions/runs/9671528017/job/26682306923#step:3:17).0
  2.17.0: Pulling from amazon/aws-cli
  492439933c1f: Pulling fs layer
  659dcd9fe751: Pulling fs layer
  cadf92deee9d: Pulling fs layer
  42aa93e9233a: Pulling fs layer
  87a391e9688b: Pulling fs layer
  42aa93e9[23](https://github.com/abaqusinc/allgeo-server/actions/runs/9671528017/job/26682306923#step:3:23)3a: Waiting
  87a391e9688b: Waiting
  659dcd9fe751: Verifying Checksum
  659dcd9fe751: Download complete
  42aa93e9233a: Verifying Checksum
  42aa93e9233a: Download complete
  87a391e9688b: Verifying Checksum
  87a391e9688b: Download complete
  cadf92deee9d: Verifying Checksum
  cadf92deee9d: Download complete
  49[24](https://github.com/abaqusinc/allgeo-server/actions/runs/9671528017/job/26682306923#step:3:24)39933c1f: Verifying Checksum
  492439933c1f: Download complete
  failed to register layer: failed to Lchown "/dev/console" for UID 0, GID 0: lchown /dev/console: no such file or directory
Error: Docker build failed with exit code 1

Please help me ... It is very urgent

Zip Exclusion not working

Sed command does nothing and we deploy node_modules and a bulky zip file.

Validation

Our properties (required) if missing cause downstream issues. Lets validate out of box.

MIssing "ListDeployments" permission not handled correctly

When "ListDeployments" is missing from the IAM permissions of the assumed role, the action doesn't handle the received error and thinks a deployment is currently running and enters the "wait 15s and try again" loop.

Fix Bash Linting

 ⚠️  Found 2 unscanned files that could potentially be supported: 

	 cleanup.sh 

	 deploy.sh 

 ShellCheck only supports sh/bash/dash/ksh scripts. For supported scripts to be scanned, make sure to add a proper shebang on the first line of the script.

Saw that because I was curious why files were linting so fast.

Audit for Open Source

Take a look to see if this repo can be open-sourced (as-is) or if we need to strip commits and restart it.

Archive is a warning, not info

Error: 937040804-38f1812cbfafe541627f2eae2692acf151c1c1c7.zip being using as zip filename. Skipping generation of ZIP.

Add License (MIT)

directory error

Error: appspec.yml was not located at: ./

what we mention at directory

Add Documentation for "directory" parameter.

Error with Blue Green deployment

Our deployment group is blue/green with copy autoscaling group.
When the deployment starts and the action begins polling aws deploy get-deployment, this does not return the deploymentOverview object until the new ASG and instances are ready.

The action seems to continue fine anyway and once the instances are ready and deployment starts then it works as normal.

The log below is the last poll before instances were running and the first after.

/deploy.sh: line 65: [: null: integer expression expected
Deployment in progress. Sleeping 15 seconds. (Try 18)
Instance Overview: Failed (null), In-Progress (null), Skipped (null), Pending (null), Succeeded (null)
Deployment Status: InProgress
Deployment in progress. Sleeping 15 seconds. (Try 19)
Instance Overview: Failed (0), In-Progress (1), Skipped (0), Pending (1), Succeeded (0)
Deployment Status: InProgress

Documentation

add readme
add info about minimal IAM role needed
give some examples of using workflow (redact project specific examples)

Request: Integrate with configure-aws-credentials

It would be nice if the task could use the credentials set up by the
https://github.com/aws-actions/configure-aws-credentials task.
This would help avoid having to refresh the secret in the repo settings and use AWS roles instead

AppSpec.yml not located at

Unable to fix this issue -

appspec.yml was not located at: ./

Missing CodeDeploy permissions in recommended IAM policy in README (?)

Readme mentions S3 permission, but not exact CodeDeploy permissions.

I know it needs at least:

codedeploy:RegisterApplicationRevision for applcations
codedeploy:CreateDeployment and list deployments (I just used codedeploy:List*) for deployment groups

I'd make a PR, but I'm not sure what the exact policies are, because took a bit of a shortcut and used some * and permissions I used for a previous implementation I was using.

Cut 1.6.0

Release 1.6.0 on Monday - March 27, 2023
Move v1 tag on Tuesday - March 28, 2023

Colors shouldn't force a space.

Using ${ORANGE}Foo will allow no space vs current of $ORANGE Foo which requires the space.

Register Outputs

Turns out you have to register the outputs your script makes

Expose output for common information

We should expose the name of the archive that was created

This will allow a tweak to the job to deploy an existing revision for projects that may generate 1 zip and send to many many servers.

Stopped Deployments should not block forever.

If you stop a deployment its done, over, etc. We shouldn't poll forever.

S3 step

Hi, I have the following question!

From what I understand, this action, in addition to deploying, also compresses the application first and sends it to s3, correct? I'm wanting to use this action, but I already sent my application to s3 in a previous step, is it possible to just use deploy?

Action fails with no output

I've tried steps recommended in #29 , but I still can't tell what's going wrong. I've added debug secret variables and I'm using v1.0.3 of this action.

This is the most verbose output I could get out of the action

...
##[debug]Input variables correctly validated.
##[debug]Zip directory located (with appspec.yml).
##[debug]Exclusion file created for files to ignore in Zip Generation.
##[debug]Zip Archive created.
##[debug]Zip Archived validated.
##[debug]Docker Action run completed with exit code 1
##[debug]Finishing: AWS CodeDeploy

In CodeDeploy panel I see no new releases. Bucket name is correct, policy gives access to it, AWS credentials also seem correct.

Any idea how can I debug this?

Detect Instances during deploment.

While we poll for the instances we should output a list of the servers with status so if we detect a failure we can fail build.

Error: appspec.yml was not located at: ./

appspec.yml is not found, but it exists.

Repo:
https://github.com/econominhas/backend

GithubAction:
https://github.com/econominhas/backend/blob/master/.github/workflows/api-deploy.yml

Usage:

- name: AWS CodeBuild & CodeDeploy
  uses: sourcetoad/aws-codedeploy-action@v1
  with:
    aws_access_key: ${{ secrets.AWS_ACCESS_KEY_ID }}
    aws_secret_key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
    aws_region: ${{ secrets.AWS_REGION }}
    codedeploy_name: Econominhas
    codedeploy_group: development_gropup
    s3_bucket: econominhas-dev-api-deploy
    s3_folder: api
    excluded_files: '.git/* .github/* .husky/* .vscode/* localstack/* openapi/* .dockerignore .editorconfig .eslintrc.js .gitignore .lintstagedrc .nvmrc .prettierrc docker-compose.yml Dockerfile.dev jest.config.json LICENSE README.md redocly.yaml'
    max_polling_iterations: 60

Pipe:
https://github.com/econominhas/backend/actions/runs/7427009109/job/20211875467

# This workflow will build a Java project with Maven
# For more information see: https://help.github.com/actions/language-and-framework-guides/building-and-testing-java-with-maven

name: CI with Maven

on:
# Triggers the workflow on push request event but only for the master branch
  push:
    branches: [ master ]

jobs:
  # This workflow contains a single job called "build"
  build:
    # The type of runner that the job will run on
    runs-on: ubuntu-latest
    
    # Steps represent a sequence of tasks that will be executed as part of the job
    
    steps:
    # Checkout the master branch
    - name: Step 1 - Checkout master branch from GitHub
      uses: actions/checkout@v2
    
    # Ensure the JDK is been setup
    - name: Step 2 - Set up JDK 1.8
      uses: actions/setup-java@v1
      with:
        java-version: 1.8
    
    # Build the project with Maven
    - name: Step 3 - Have GitHub Actions build the project with Maven
      run: mvn -B package --file pom.xml
    
    # Copy the jar to a folder called staging
    - name: Step 4 -Copy the jar file
      run: mkdir staging && cp target/*.jar staging
    
    # Cache all the dependencies to speed up the workflow runs
    - name: Step 5 - Set up a cache for Maven
      uses: actions/cache@v2
      with:
        path: ~/.m2
        key: ${{ runner.os }}-m2-${{ hashFiles('**/pom.xml') }}
        restore-keys: ${{ runner.os }}-m2

    # Package the artificate and allow us to download them. This is useful for re-testing in local environment
    - name: Step 6 - Persist workflow data as astifacts
      uses: actions/upload-artifact@v2
      with:
        name: github-actions-artifact
        path: staging

    - name: Step 7 - Sourcetoad - AWS CodeDeploy for GitHub Actions
      uses: sourcetoad/[email protected]
      with:
        # AWS Access Key
        aws_access_key: ***
        # AWS Secret Key
        aws_secret_key: ${{ secrets.AWS_SECRET_KEY }}
        # AWS Region
        aws_region: ap-southeast-1
        # S3 Bucket for CodeDeploy Assets
        s3_bucket: ***
        # S3 Folder for ZIP.
        s3_folder: /
        # AWS CodeDeploy Application Name
        codedeploy_name: ***
        # AWS CodeDeploy Application Group
        codedeploy_group: ***

And the output:

Register vs Deploy

Need to support a flag to deploy a revision or register one.