ECS "InternalError: failed to normalize image reference"

Question

We have a docker images being stored in ECR, some have come from a TC build agent and some have come from builds on local developer machines. All images can be pulled an run locally but when trying to start the images built on the build TC build agents we get the following error when the ECS cluster starts running the task.

InternalError: failed to normalize image reference "http://xxx.eu-west-1.amazonaws.com/events/name-of-contianer:tag"

I can't seem to find any additional info in the logs, or mention of an error for normalize in any of the ECS docs. Has anyone experienced this error, or even know what it means!

ECS Task Definition:

{
  "ipcMode": null,
  "executionRoleArn": "arn:aws:iam::xxxxxxxxxxxx:role/ecsTaskExecutionRole",
  "containerDefinitions": [
    {
      "dnsSearchDomains": null,
      "environmentFiles": null,
      "logConfiguration": null,
      "entryPoint": null,
      "portMappings": [
        {
          "hostPort": 80,
          "protocol": "tcp",
          "containerPort": 80
        }
      ],
      "command": null,
      "linuxParameters": null,
      "cpu": 0,
      "environment": [
        {
          "name": "ServiceConfiguration__BasePath",
          "value": "/mobile-read-model"
        },
        {
          "name": "ServiceConfiguration__Version",
          "value": "0.1.0"
        }
      ],
      "resourceRequirements": null,
      "ulimits": null,
      "dnsServers": null,
      "mountPoints": [],
      "workingDirectory": null,
      "secrets": null,
      "dockerSecurityOptions": null,
      "memory": null,
      "memoryReservation": null,
      "volumesFrom": [],
      "stopTimeout": null,
      "image": "http://xxxxxxxxxxxx.dkr.ecr.eu-west-1.amazonaws.com/events/mobile-app-read-model:0.1.0-blank-read-model0011",
      "startTimeout": null,
      "firelensConfiguration": null,
      "dependsOn": null,
      "disableNetworking": null,
      "interactive": null,
      "healthCheck": {
        "retries": 1,
        "command": [
          "CMD-SHELL",
          "curl -f http://localhost/health || exit 1"
        ],
        "timeout": 5,
        "interval": 30,
        "startPeriod": 60
      },
      "essential": true,
      "links": null,
      "hostname": null,
      "extraHosts": null,
      "pseudoTerminal": null,
      "user": null,
      "readonlyRootFilesystem": false,
      "dockerLabels": null,
      "systemControls": null,
      "privileged": null,
      "name": "mobile-read-model"
    }
  ],
  "placementConstraints": [],
  "memory": "512",
  "taskRoleArn": "arn:aws:iam::xxxxxxxxxxxx:role/ecsTaskExecutionRole",
  "compatibilities": [
    "EC2",
    "FARGATE"
  ],
  "taskDefinitionArn": "arn:aws:ecs:eu-west-1:xxxxxxxxxxxx:task-definition/mobile-app-readmodel:55",
  "family": "mobile-app-readmodel",
  "requiresAttributes": [
    {
      "targetId": null,
      "targetType": null,
      "value": null,
      "name": "com.amazonaws.ecs.capability.task-iam-role"
    },
    {
      "targetId": null,
      "targetType": null,
      "value": null,
      "name": "ecs.capability.container-health-check"
    },
    {
      "targetId": null,
      "targetType": null,
      "value": null,
      "name": "com.amazonaws.ecs.capability.docker-remote-api.1.18"
    },
    {
      "targetId": null,
      "targetType": null,
      "value": null,
      "name": "ecs.capability.task-eni"
    },
    {
      "targetId": null,
      "targetType": null,
      "value": null,
      "name": "com.amazonaws.ecs.capability.docker-remote-api.1.29"
    }
  ],
  "pidMode": null,
  "requiresCompatibilities": [
    "FARGATE"
  ],
  "networkMode": "awsvpc",
  "cpu": "256",
  "revision": 55,
  "status": "ACTIVE",
  "inferenceAccelerators": null,
  "proxyConfiguration": null,
  "volumes": []
}

That's not a valid ECR image ID. What does the ECS task definition look like? Specifically, what is the value for the `image` attribute of the task definition? — Mark B, Feb 26 '21 at 14:11
@MarkB, thanks for the response. You're absolutely right, in checking for the info you asked for we realised the problem! In the deployment pipeline we've incorrectly added the image with a URL in front instead of just the repo name. Updated the pipeline and all working now! — Dan Gardner, Feb 26 '21 at 14:32

score 11 · Accepted Answer · answered Feb 26 '21 at 14:40

11

The image attribute should be: "image": "xxxxxxxxxxxx.dkr.ecr.eu-west-1.amazonaws.com/events/mobile-app-read-model:0.1.0-blank-read-model0011", without the http://.

answered Feb 26 '21 at 14:40

Mark B

183,023
24
297
295

score 8 · Answer 2 · answered Feb 26 '21 at 14:37

8

You need to remove http:// from the image name.

answered Feb 26 '21 at 14:37

Baron

323
4
10

ECS "InternalError: failed to normalize image reference"

2 Answers2