App Runner Deployment Race Condition When Using CloudFormation - No Outbound Connectivity

0

I've noticed that App Runner instances can get into a weird state where they have no outbound connectivity (Not just DNS as mentioned in other questions, even a ping of 1.1.1.1 or a HTTP request to an AWS service such as DynamoDB fails) but still retains inbound connectivity.

This seems to happen sometimes in a specific deployment scenario when using CloudFormation, but not always.

Scenario

  • An App Runner service has been created via CloudFormation
  • The App Runner service has configured to auto-deploy from Elastic Container Registry
  • App Runner is configured with a status check which does not rely on outbound connectivity
  • Push a new container image to ECR (Or tag an existing image with a tag which will trigger the automatic deployment)
  • Immediately start a CloudFormation deployment (There don't have to be any changes to the AppRunner configuration)

Sometimes this will cause App Runner to create an instance with no outbound connectivity. I've tested this with a .NET 6 container image but I suspect this will affect all images as even pings were failing.

Work-Around

  • Disable auto-deployment of App Runner
  • Manually trigger a deployment of App Runner via the AWS API as part of the deployment pipeline, after the CloudFormation deployment

This has worked every time for me to date.

  • Thanks. Are you in a position to open a support request with AWS to follow up on this? Let me know.

yngndrw
질문됨 2년 전309회 조회
1개 답변
0

I am not sure if this is the same issue but we are indeed deploying a fix for a similar problem. Please open a ticket if you are still experiencing the issue.

AWS
전문가
답변함 2년 전

로그인하지 않았습니다. 로그인해야 답변을 게시할 수 있습니다.

좋은 답변은 질문에 명확하게 답하고 건설적인 피드백을 제공하며 질문자의 전문적인 성장을 장려합니다.

질문 답변하기에 대한 가이드라인

관련 콘텐츠