Questions tagged with AWS CodePipeline

Hi AWS, I am following this blog post https://aws.amazon.com/blogs/containers/ci-cd-pipeline-for-testing-containers-on-aws-fargate-with-scaling-to-zero/ to learn about building CI/CD pipelines for containers on ECS Fargate but while accessing the CloudFormation Template and Dockerfile stored in the zipped format I am getting the following error. Here is the URL for the files stored: https://d268s23yov0ww.cloudfront.net/aws-autostart-pipeline.zip and here is the error screenshot:  Please acknowledge this as I am stucked because of the same. Thank youlg...

I'd like to build a multi-architecture Docker image, push it to the default CDK ECR repo, and then push it to different deployment stages (stacks in separate accounts) using CDK Pipelines. I create the image using something like the following: ``` IMAGE_TAG=${AWS_ACCOUNT}.dkr.ecr.${REGION}.amazonaws.com/cdk-hnb659fds-container-assets-${AWS_ACCOUNT}-${REGION}:myTag docker buildx build --progress=plain \ --platform linux/amd64,linux/arm64 --push \ --tag ${IMAGE_TAG} \ myDir/ ``` This results in three things pushed to ECR, two images and an image index (manifest). I'm then attempting to use the [cdk-ecr-deployment](https://github.com/cdklabs/cdk-ecr-deployment) to copy the image to a specific stack, for example: ``` cdk_ecr_deployment.ECRDeployment( self, "MultiArchImage", src=cdk_ecr_deployment.DockerImageName(f"{cdk_registry}:myTag"), dest=cdk_ecr_deployment.DockerImageName(f"{stack_registry}:myTag"), ) ``` However, this ends up copying only the image corresponding to the platform running the CDK deployment instead of the 2 images plus manifest. There's a [feature request](https://github.com/cdklabs/cdk-ecr-deployment/issues/192) open on `cdk-ecr-deployment` to support multi-arch images. I'm hoping someone might be able to suggest a modification to the above or some alternative that achieves the same goal, which is to deploy the image to multiple environments using CDK Pipelines. I also tried building the images + manifest into a tarball locally and then using the `aws_ecr_assets.TarballImageAsset` construct, but I encountered this [open issue ](https://github.com/aws/aws-cdk/issues/18044) when attempting the deployment locally. I'm not sure if the `TarballImageAsset` supports a multi-arch image, as it seems like the `DockerImageAsset` doesn't. Any ideas?lg...

I have a bunch of resources that get stood up as part of a regular deployment, i.e. worker EC2's. I need their IP's and to populate them into a config (But this is a post-deployment process) and it can become long/monotonous as its trivial work. Is there any api calls or any features in AWS I can use to automate this? I'm fine with shell/bash/python. Something in my head is to create a pipeline job that runs after the deployment pipeline, it queries or "scans" for the scope-named EC2's and then retrieves the IP's. Stores them for reference and then appends this to a config file I store in code commit. Possible?lg...

All of our codebuild jobs have been stuck in Submitted phase (the Queued phase is shown with no status). They worked earlier in the day and we haven't changed anything about the configuration I've checked and there's no outage reported on the AWS health page. Is there an unreported codebuild outage?lg...

Are there any plans to allow users to specify the timeout on approval stages on pipelines? Currently, I don’t believe it is possible.lg...

Hi, As I updated AMI ID in parameter store to the latest untested version to try to test it on our Integration and UAT environments on some infrastructure codepipeline, then it turned out that we needed to change the AMI back to the previous tested version to do another codepipeline deployment on integration, UAT and production, so the deployment was successful. I then had to change the AMI ID back to the latest version in the parameter store so that we can test another infrastructure codepipeline for Integration and UAT, I then got an error in one of the ECS cloudformation stack during the deployment on integration as "UPDATE_ROLLBACK_COMPLETE" Can anyone help where I should check to find the root cause please? Many thanks in advance Dellg...

I have an automated Codebuild project that is triggered by a Github Enterprise Pull Request Webhook. The problem is that I cannot see the information related inside the PR comments. Maybe there is a way from buildspec.yml to push certain logging information back to the PR comments so certain non-privileged users can see the Codebuild results.lg...

 I trying tutorial on "MLOps template for model building, training, and deployment with third-party Git repositories using CodePipeline". But I am getting error as shown in imagelg...

Currently facing an issue with creating a notification rule for code pipeline using CDK with Pipeline or CodePipeline constructs ``` Resource handler returned message: "Invalid request provided: AWS::CodeStarNotifications::NotificationRule" (RequestToken: <GUID>, HandlerErrorCode: InvalidRequest) ``` The topic and chatbot are already defined in a different stack. The topic already has configured an access policy. However when I attempt to create a rule for my pipelines, it fails. Every time. ``` { "Version": "2008-10-17", "Statement": [ { "Sid": "AWSCodeStarNotifications_publish", "Effect": "Allow", "Principal": { "Service": "codestar-notifications.amazonaws.com" }, "Action": "SNS:Publish", "Resource": "arn:aws:sns:us-east-1:123456789:PipelineNotifications" } ] } ``` Referencing topic: ``` const pipelineNotificationsTopic = sns.Topic.fromTopicArn( this, "PipelineNotifications", `arn:aws:sns:${props?.env.region}:${props?.env.account}:PipelineNotifications ` ); ``` Pipeline code: ``` this.pipeline = new Pipeline( this, `${props?.appName}-MyPipeline`, { pipelineName: "MyPipeline", crossAccountKeys: true, restartExecutionOnUpdate: true, } ); const rule = this.pipeline.notifyOn( "CodeStarRule", pipelineNotificationsTopic, { events: [ PipelineNotificationEvents.PIPELINE_EXECUTION_SUCCEEDED, PipelineNotificationEvents.PIPELINE_EXECUTION_FAILED, PipelineNotificationEvents.ACTION_EXECUTION_FAILED, ], detailType: DetailType.FULL, } ); ``` or ``` this.pipeline = new Pipeline( this, `${props?.appName}-MyPipeline`, { pipelineName: "MyPipeline", crossAccountKeys: true, restartExecutionOnUpdate: true, } ); const rule =new NotificationRule(this, "Notification", { detailType: DetailType.FULL, events: [ "codepipeline-pipeline-pipeline-execution-failed", "codepipeline-pipeline-action-execution-failed", "codepipeline-pipeline-stage-execution-failed", "codepipeline-pipeline-manual-approval-failed", "codepipeline-pipeline-manual-approval-needed", ], source: pipeline, targets: [pipelineNotificationsTopic], }); ```lg...

We have been trying to launch App Runner Service using CDK (Java) but keep getting the following error. > Resource handler returned message: "null" (RequestToken: 565eeb6a-8af1-af59-5612-12ef28c1380e, HandlerErrorCode: InternalFailure) To provide details, our CDK code does the following: 1. Creates an AWS Code Pipeline for building our code (CDK + Spring Boot Service) from GitHub 2. Builds a Docker image for a Spring Boot Service. 3. Upload the Docker image to AWS ECR. 4. Launch AWS App Runner --> This step fails We were able to comment out step 4 in our CDK code and then manually launch AWS App runner from console using the Docker image built in Step 3. This makes us believe that something is wrong with either our CDK code or AWS Cloud Formation which runs the CDK. Thank you for looking at our issue.lg...

Hello, I have a question about one of the Workshops published in the AWS Workshop studio. I am going through the "MLOps and integrations" hands-on, but it is not working well during the process. Here is the [URL](https://catalog.us-east-1.prod.workshops.aws/workshops/50716cb8-6a42-427f-9eeb-0465dea6e95b/en-US) for the hands-on. Specifically, the build process performed when deploying the provided CloudFormation template will generate an error occurs. I think that the [URL](https://aws-mlops-samples.s3-eu-west-1.amazonaws.com/mnist_output_10.png) for S3 listed in the distributed [source](https://static.us-east-1.prod.workshops.aws/public/3d5c4066-8fc6-4f66-ba73-d0cfe5e3c927/static/code/ml-ops.zip)(source/test.py) in the workshop is probably not valid. Do any of you know how to solve this problem? ----Excerpts from the CodeBuild error log----- Traceback (most recent call last): File "source/test.py", line 20, in <module> test_file, File "/usr/local/lib/python3.6/site-packages/wget.py", line 526, in download (tmpfile, headers) = ulib.urlretrieve(binurl, tmpfile, callback) File "/usr/local/lib/python3.6/urllib/request.py", line 248, in urlretrieve with contextlib.closing(urlopen(url, data)) as fp: File "/usr/local/lib/python3.6/urllib/request.py", line 223, in urlopen return opener.open(url, data, timeout) File "/usr/local/lib/python3.6/urllib/request.py", line 532, in open response = meth(req, response) File "/usr/local/lib/python3.6/urllib/request.py", line 642, in http_response 'http', request, response, code, msg, hdrs) File "/usr/local/lib/python3.6/urllib/request.py", line 570, in error return self._call_chain(*args) File "/usr/local/lib/python3.6/urllib/request.py", line 504, in _call_chain result = func(*args) File "/usr/local/lib/python3.6/urllib/request.py", line 650, in http_error_default raise HTTPError(req.full_url, code, msg, hdrs, fp) urllib.error.HTTPError: HTTP Error 403: Forbiddenlg...

We just started to replace our bitbucket pipelines for AWS Codebuild. One feature common to Bitbucket Pipeline and GitHub Actions, is there's an API to trigger a build. This is quite useful when outbound webhooks fire from external system events not coming from the source control provider. As an example, we are using pact.io for contract testing. The pact broker will fire a webhook to trigger a build when a certain event occurs. Pact.io has lots of documentation for CI build tools and this is easily replicated when using bitbucket pipelines or github actions. However, I cannot seem to locate any documentation on how to trigger a Codebuild from an API Request. See pact webhook templates for trigger_build examples of many CI tools, https://docs.pact.io/pact_broker/webhooks/template_library. How can triggering a build be easily done with AWS Codebuild?lg...