0
我正在尝试使用GitLab管道将Docker容器部署到EC2容器服务中。我已经在AWS中创建了群集,并且还运行了2个实例。问题是,一旦执行通过GitLab将Docker容器部署到EC2容器服务时出错管道
命令我收到以下错误:
WARN[0000] Skipping unsupported YAML option for service... option name=build service name=testContainer
ERRO[0001] Error registering task definition error="ClientException: Container.image should not be null or empty.\n\tstatus code: 400, request id: 27277917-8d8e-11e7-8cb2-8d5ad96f4568" family=ecscompose-workillence
ERRO[0001] Create task definition failed error="ClientException: Container.image should not be null or empty.\n\tstatus code: 400, request id: 27277917-8d8e-11e7-8cb2-8d5ad96f4568"
FATA[0001] ClientException: Container.image should not be null or empty.
这是一个截图从我的AWS回购:
图像不似乎不是空的。
这是我的CI文件看起来像:
image: docker:latest
services:
- docker:dind
stages:
- build
- deploy
testBuild:
stage: build
script:
- docker version
- docker build -t 084622260700.dkr.ecr.us-east-2.amazonaws.com/testrepo:latest app/
- docker images
- docker login -u AWS -p <password> 084622260700.dkr.ecr.us-east-2.amazonaws.com
- docker push 084622260700.dkr.ecr.us-east-2.amazonaws.com/testrepo:latest
testDeploy:
stage: deploy
image: python:3-alpine
variables:
AWS_REGION: "us-east-2"
AWS_ACCESS_KEY_ID: "key"
AWS_SECRET_ACCESS_KEY: "key"
PROD_TARGET_GROUP_ARN: "arn:aws:elasticloadbalancing:us-east-2:084622260700:targetgroup/TestTG/af0fbec5b9cc869f"
ECS_CLUSTER: "testCluster"
AWS_KEY_PAIR: "testKP.pem"
AWS_VPC: "vpc-0419416d"
AWS_SUBNETS: "subnet-ad6cdfe0,subnet-ad6cdfe0,subnet-0fc1aa74"
AWS_INSTANCE_TYPE: "t2.micro"
AWS_INSTANCE_COUNT: "2" #Number of instances to scale to
AWS_ROLE: "arn:aws:iam::aws:policy/service-role/AmazonEC2ContainerServiceRole"
before_script:
##### Install AWS ECS-CLI #####
- apk add --update curl
- curl -o /usr/local/bin/ecs-cli https://s3.amazonaws.com/amazon-ecs-cli/ecs-cli-linux-amd64-latest
- chmod +x /usr/local/bin/ecs-cli
script:
##### Configure ECS-CLI, run the container and scale to 2 instances #####
- ecs-cli configure --region $AWS_REGION --access-key $AWS_ACCESS_KEY_ID --secret-key $AWS_SECRET_ACCESS_KEY --cluster $ECS_CLUSTER
- ecs-cli up --keypair $AWS_KEY_PAIR --capability-iam -c $ECS_CLUSTER --instance-type t2.micro --size 2 --vpc $AWS_VPC --subnets $AWS_SUBNETS --force
##### This docker-compose.yml is the one described above #####
- ecs-cli compose --file docker-compose.yml service up --target-group-arn $PROD_TARGET_GROUP_ARN --container-name app --container-port 8080 --role $AWS_ROLE
- ecs-cli compose --file docker-compose.yml service scale 2
environment:
name: ci
##### This is the URL seen under 'DNS name' when the LB was created #####
url: TestLB-1717492587.us-east-2.elb.amazonaws.com
only:
- aws
这是撰写文件的样子:
app:
image: 084622260700.dkr.ecr.us-east-2.amazonaws.com/testrepo:latest
command: yarn run start-dev
links:
- mongodb
- redis
- elasticsearch
ports:
- '8000:8000'
- '5858:5858'
- '9229:9229'
- '80:8080'
environment:
NODE_ENV: local
volumes:
- ./testrepo/.:/opt/app
mongodb:
build: docker/definitions/mongo
ports:
- "27017:27017"
redis:
build: docker/definitions/redis
ports:
- "6379:6379"
elasticsearch:
build: docker/definitions/elasticsearch
ports:
- "9200:9200"
- "9300:9300"
你曾经面临相同或相似的问题,可能是如何做你解决了吗?
我没有猜测那些服务将在此错误中扮演任何角色,我没有描述,但我将使用外部提供程序来执行mongo,redis和elasticsearch。这意味着它们不会在Docker包内部构建,而且我已经在本地测试了环境并且它可以工作,我可以连接它们中的三个。 有什么办法可以在撰写文件中声明它们是外部的吗? –
如果它们是外部的,则将它们从撰写中删除。保留两个组合'-test.yml'和'-prod.yml'。使用'prod'进行AWS部署 –
好的,我测试了它,并再次发生了相同的错误。 我用下面的代码: 'testContainer: 编译:084622260700.dkr.ecr.us-east-2.amazonaws.com/testrepo:latest 命令:丝线行进启动开发 端口: - “8000 :8000' - '5858:5858' - '9229:9229' - '80:8080' 环境: NODE_ENV:本地 卷: - ./testrepo /.:/选择/ app' –