We are switching from the Docker Hub to ECR, and I'm curious how to structure the Dockerrun.aws.json file to use this image. I tried to change the name as <my_ECR_URL>/<repo_name>:<image_tag>, but it failed. I also saw the details of private registries using the authentication file in S3, but this does not seem to be the correct route when it aws ecr get-loginis the recommended authentication method using ECR.
Can someone tell me how I can use the ECR image in the Beanstalk Dockerrun.aws.json file?
If I look at the definition of an ECS task, an attribute is needed there com.amazonaws.ecs.capability.ecr-auth, but I do not set it anywhere in my file Dockerrun.aws.json, and I'm not sure what should be there. Perhaps this is an S3 bucket? Something is needed, since every time I try to run tasks created by the Elastic Beanstalk from ECS, I get:
Run tasks failed
Reasons : ATTRIBUTE
Any ideas are greatly appreciated.
Refresh I can see from some other threads that this happened with earlier versions of the ECS agent, but I am currently running Agent version 1.6.0and Docker version 1.7.1which I consider are recommended versions. Perhaps this is a problem with the Docker version?
source
share