Unable to Deploy Redis Containers Following Instructions

[yuta519]

Context

• Following this Defang documentation to deploy to AWS as BYOC.

  https://docs.defang.io/docs/tutorials/deploy-to-aws

• The documentation states that compose.yaml needs to be created by following this

  https://docs.docker.com/compose/gettingstarted/

• Encountered an error where the Redis container could not be connected.

Steps Taken

• Created the compose.yaml file with the following content:

  services:
    web:
      build: .
      ports:
        - "8000:5000"
    redis:
      image: "redis:alpine"

• Ran the command: defang compose up --project-name=yutadeploy --provider=aws

Issue Encountered

• Received the following error:

redis.exceptions.ConnectionError: Error -2 connecting to redis:6379. Name does not resolve

2025-02-04T18:49:39.424-08:00 web 2c8cfa86f988457eaaf0bf842340dac6 redis.exceptions.ConnectionError: Error -2 connecting to redis:6379. Name does not resolve.
2025-02-04T18:49:39.518-08:00 web 2c8cfa86f988457eaaf0bf842340dac6 10.0.165.192 - - [05/Feb/2025 02:49:39] "GET / HTTP/1.1" 500 -
2025-02-04T18:49:39.520-08:00 web 2c8cfa86f988457eaaf0bf842340dac6 Traceback (most recent call last):
2025-02-04T18:49:39.520-08:00 web 2c8cfa86f988457eaaf0bf842340dac6   File "/usr/local/lib/python3.10/site-packages/redis/connection.py", line 357, in connect
2025-02-04T18:49:39.520-08:00 web 2c8cfa86f988457eaaf0bf842340dac6     sock = self.retry.call_with_retry(
2025-02-04T18:49:39.520-08:00 web 2c8cfa86f988457eaaf0bf842340dac6   File "/usr/local/lib/python3.10/site-packages/redis/retry.py", line 62, in call_with_retry
2025-02-04T18:49:39.520-08:00 web 2c8cfa86f988457eaaf0bf842340dac6     return do()
2025-02-04T18:49:39.520-08:00 web 2c8cfa86f988457eaaf0bf842340dac6   File "/usr/local/lib/python3.10/site-packages/redis/connection.py", line 358, in <lambda>
2025-02-04T18:49:39.520-08:00 web 2c8cfa86f988457eaaf0bf842340dac6     lambda: self._connect(), lambda error: self.disconnect(error)
2025-02-04T18:49:39.520-08:00 web 2c8cfa86f988457eaaf0bf842340dac6   File "/usr/local/lib/python3.10/site-packages/redis/connection.py", line 698, in _connect
2025-02-04T18:49:39.520-08:00 web 2c8cfa86f988457eaaf0bf842340dac6     for res in socket.getaddrinfo(
2025-02-04T18:49:39.520-08:00 web 2c8cfa86f988457eaaf0bf842340dac6   File "/usr/local/lib/python3.10/socket.py", line 967, in getaddrinfo
2025-02-04T18:49:39.520-08:00 web 2c8cfa86f988457eaaf0bf842340dac6     for res in _socket.getaddrinfo(host, port, family, type, proto, flags):
2025-02-04T18:49:39.520-08:00 web 2c8cfa86f988457eaaf0bf842340dac6 socket.gaierror: [Errno -2] Name does not resolve

Expected Behavior

• The Redis container should be accessible from the web service.

Modifications Made

• Updated compose.yaml with the following changes:

  services:
    web:
      build: .
      networks:
        - default
      ports:
        - "8000:5000"
      depends_on:
        - redis
      environment:
        REDIS_HOST: redis
        REDIS_PORT: 6379
  
    redis:
      image: "redis:6.2"
      x-defang-redis: true
      networks:
        - default
      healthcheck:
        test: ["CMD", "redis-cli", "ping"]
        interval: 10s
        timeout: 5s
        retries: 3
  
  networks:
    default:
      driver: bridge

  • Specified Redis version 6.2 instead of alpine.
  • Added networking configurations.
  • Introduced a health check for Redis.

Outcome

• Although the CLI did not return a successful result, an new ElastiCache instance was successfully created.

- After Redis cache instance’s status get “Available”, I ran the defang command again and it worked finally. - This is because Redis cache takes time to be available and web container can’t establish connection during redis bootstrap.

Needed Steps

• Fix the Redis deployment
  • Need to add some workflow to wait for Redis cache up
• Fix the document

[yuta519]

defang compose down waits for termination of the managed service. But it took about 10 mins to be completed.

Starting: /Users/yuta/go/bin/dlv dap --listen=127.0.0.1:56385 --log-dest=3 from /Users/yuta/work/defang/src/cmd/cli
DAP server listening at: 127.0.0.1:56385
Type 'dlv help' for list of commands.
 - Reading access token from file /Users/yuta/.local/state/defang/fabric-prod1.defang.dev
 - Using tenant default for cluster fabric-prod1.defang.dev:443
 - fabricClient /io.defang.v1.FabricController/WhoAmI null
 - fabricClient /io.defang.v1.FabricController/GetVersion null
 - Fabric: v0.6.0-299-ge52e2129 CLI: development CLI-Min: v1.0.0
 - fabricClient /io.defang.v1.FabricController/CheckToS null
 - tracking event "COMPOSE-DOWN INVOKED": [{args []} {err <nil>} {non-interactive true} {provider aws} {CalledAs   down} {version development} {debug true} {provider aws}]
 - fabricClient /io.defang.v1.FabricController/Track {"anon_id":"16f77a06-f29e-464e-b7c2-e1862a6d9c49","event":"COMPOSE-DOWN INVOKED","properties":{"CalledAs":"  down","args":"[]","debug":"true","err":"\u003cnil\u003e","non-interactive":"true","provider":"aws","version":"development"},"os":"darwin","arch":"arm64"}
 * Using AWS provider from command line flag
 - Creating aws provider
name: deploy-to-lab
services:
    redis:
        healthcheck:
            test:
                - CMD
                - redis-cli
                - ping
            timeout: 5s
            interval: 10s
            retries: 3
        image: redis:6.2
        networks:
            default: null
        x-defang-redis: true
    web:
        build:
            context: /Users/yuta/work/deploy-to-lab
            dockerfile: Dockerfile
        depends_on:
            redis:
                condition: service_started
                required: true
        environment:
            REDIS_HOST: redis
            REDIS_PORT: "6379"
        networks:
            default: null
        ports:
            - mode: ingress
              target: 5000
              published: "8000"
              protocol: tcp
networks:
    default:
        name: deploy-to-lab_default
        driver: bridge

 - fabricClient /io.defang.v1.FabricController/CanIUse {"project":"deploy-to-lab","provider":2}
 - Destroying project "deploy-to-lab" []
 - Using CD image: "public.ecr.aws/defang-io/cd:public-v0.6.0-299-ge52e2129@sha256:bd6a9b1aa57263f6d384f2de8f9ac241e347071916019b63b553d8794c6c703a"
Waiting for CloudFormation stack defang-cd to be updated...
 - AWS_REGION=us-west-2 AWS_PROFILE=defang-sandbox DEFANG_MODE=unspecified_mode NPM_CONFIG_UPDATE_NOTIFIER=false DEFANG_DEBUG= PROJECT=deploy-to-lab PULUMI_BACKEND_URL=s3://defang-cd-bucket-ren6kjmnovwj?region=us-west-2&awssdk=v2 NO_COLOR=1 DOMAIN=dummy.domain PULUMI_SKIP_UPDATE_CHECK=true DEFANG_ORG=yuta519 DEFANG_PREFIX=Defang PRIVATE_DOMAIN=deploy-to-lab.internal PULUMI_CONFIG_PASSPHRASE=asdf STACK=beta npm run dev down
 - fabricClient /io.defang.v1.FabricController/PutDeployment {"deployment":{"id":"e17920e2691f46faa93196660130cef9","project":"deploy-to-lab","provider":"aws","provider_account_id":"532501343364","timestamp":{"seconds":1739216441,"nanos":486916000},"action":2}}
 * Deleted services, deployment ID e17920e2691f46faa93196660130cef9
 - Tailing RUN,BUILD logs in project "deploy-to-lab"
 - Tailing task arn:aws:ecs:us-west-2:532501343364:task/defang-cd-Cluster-J7Tr6gUfmrEC/e17920e2691f46faa93196660130cef9
2025-02-10T11:41:50.076-08:00 cd  ** Destroy started for stack beta
2025-02-10T11:41:56.999-08:00 cd  ** Refresh succeeded in 3.0s

 * Done.  -> finished around 11:53