hello, I am trying to push my code on Gitlab but it constantly fails on the push stage giving me the below error. I would really appreciate any suggestions of how to tackle this issue…

1 Running with gitlab-runner 12.6.0-rc1 (ec299e72)
2   on docker-auto-scale 72989761
3
Using Docker executor with image ruby:2.5 ...
00:33
4 Pulling docker image ruby:2.5 ...
5 Using docker image sha256:54cb86b0bcdc353e103e9dd1bff69fcd01e3235420ed17cbf5efac22010b373e for ruby:2.5 ...
7
Running on runner-72989761-project-16129735-concurrent-0 via runner-72989761-srm-1578007908-e356a4c4...
00:05
9
$ eval "$CI_PRE_CLONE_SCRIPT"
00:01
10 Fetching changes with git depth set to 50...
12 Created fresh repository.
14  * [new ref]         refs/pipelines/106725538 -> refs/pipelines/106725538
15  * [new branch]      master                   -> origin/master
16 Checking out 4620472d as master...
17 Skipping Git submodules setup
21
$ docker login -u "gitlab-ci-token" -p "$CI_BUILD_TOKEN" "$CI_REGISTRY"
00:01
22 /bin/bash: line 89: docker: command not found
26 ERROR: Job failed: exit code 1

Hi,

please post the CI config here, the linked repo is private.
I assume you want to build a Docker image and push it to the registry?

Cheers,
Michael

hello Michael, thank you for your reply.
Please let me know what you think the issue could be, I’m only a beginner, so I’m not sure what may be causing it and I spent a lot of time researching this but no luck (I kept creating and deleting projects to see if sth could work).

current issue in the Push stage:

    1 Running with gitlab-runner 12.6.0-rc1 (ec299e72)
    2   on docker-auto-scale 72989761
    3
    Using Docker executor with image ruby:2.5 ...
    00:37
    4 Pulling docker image ruby:2.5 ...
    5 Using docker image sha256:54cb86b0bcdc353e103e9dd1bff69fcd01e3235420ed17cbf5efac22010b373e for ruby:2.5 ...
    7
    Running on runner-72989761-project-16138356-concurrent-0 via runner-72989761-srm-1578057356-89d9bc73...
    00:05
    9
    $ eval "$CI_PRE_CLONE_SCRIPT"
    00:02
    10 Fetching changes with git depth set to 50...
    11 Initialized empty Git repository in 
    12 Created fresh repository.
    13 From 
    14  * [new ref]         refs/pipelines/106836401 -> refs/pipelines/106836401
    15  * [new branch]      master                   -> origin/master
    16 Checking out 58e0d93f as master...
    17 Skipping Git submodules setup
    21
    $ docker login -u "gitlab-ci-token" -p "$CI_BUILD_TOKEN" "$CI_REGISTRY"
    00:01
    22 /bin/bash: line 89: docker: command not found
    26 ERROR: Job failed: exit code 1

my .gitlab.ci-yml file is as follows:

    stages:
  - pull
  - build
  - lint
  - push
  - deploy
  - cleanup

before_script:
  - docker login -u "gitlab-ci-token" -p "$CI_BUILD_TOKEN" "$CI_REGISTRY"

pull:
  stage: pull
  allow_failure: true
  script:
    - docker pull "$CI_REGISTRY_IMAGE":latest

build:
  stage: build
  script:
    - docker build --tag="$CI_PIPELINE_ID":"$CI_COMMIT_REF_NAME" --cache-from="$CI_REGISTRY_IMAGE":latest .

lint:
  stage: lint
  script:
    - echo $CI_COMMIT_REF_NAME
    - echo $CI_PIPELINE_ID
    - echo $CI_REGISTRY_IMAGE
    - export CI_PIPELINE_ID=$CI_PIPELINE_ID
    - export CI_BUILD_REF_NAME=$CI_COMMIT_REF_NAME
    - docker-compose -p "$CI_PIPELINE_ID" -f docker-compose.ci.yml run app flake8 .

push image:
  stage: push
  only:
    - master
    - tags
  script:
    - docker tag "$CI_PIPELINE_ID":"$CI_COMMIT_REF_NAME" "$CI_REGISTRY_IMAGE":"$CI_COMMIT_REF_NAME"
    - docker push "$CI_REGISTRY_IMAGE":"$CI_COMMIT_REF_NAME"

push latest:
  stage: push
  script:
    - docker tag "$CI_PIPELINE_ID":"$CI_COMMIT_REF_NAME" "$CI_REGISTRY_IMAGE":latest
    - docker push "$CI_REGISTRY_IMAGE":latest

cleanup:
  stage: cleanup
  when: always
  script:
    - docker rmi -f "$CI_PIPELINE_ID":"$CI_COMMIT_REF_NAME"
    - docker-compose -p "$CI_PIPELINE_ID" -f docker-compose.ci.yml down --remove-orphans

deploy:
  stage: deploy
  when: manual
  only:
    - master
  script:
    - docker-compose -f docker-compose.deploy.yml pull
    - docker-compose -f docker-compose.deploy.yml down --remove-orphans
    - docker-compose -f docker-compose.deploy.yml up -d

my docker-compose-deploy:

    version: '3'
services:

  backend:
    image: registry.gitlab.com/monixm/last-propulsion-project:latest
    restart: always
    networks:
      - nginx_network
      - database1_network  # <-- connect to the bridge
    env_file:
      - prod.env
    command: 'sh /scripts/run.sh'
    volumes:
      - frontend:/frontend-build
      - static-files:/static-files
      - media-files:/media-files
    depends_on:
      - database

  database:
    image: postgres:latest
    restart: always
    env_file:
      - prod.env
    volumes:
      - database:/var/lib/postgresql/data
    networks:
      - database1_network

  nginx:
    image: nginx:latest
    volumes:
      - static-files:/static-files
      - media-files:/media-files
      - ./config:/etc/nginx/conf.d
      - frontend:/frontend
      - /etc/letsencrypt:/etc/letsencrypt
    ports:
      - "80:80"
      - "443:443"
    command: /bin/bash -c "nginx -g 'daemon off;'"
    depends_on:
      - backend
    networks:
      - nginx_network

networks:  # <-- and here
  nginx_network:
    driver: bridge
  database1_network:  # <-- add the bridge
    driver: bridge

volumes:
  frontend:
  database:
  static-files:
  media-files:

my docker-compose:

    version: '3'
services:
  database:
    image: postgres:latest
    ports:
      - '5432:5432'
    env_file:
      - dev.env
    volumes:
      - database:/var/lib/postgresql/data

  app:
    image: registry.gitlab.com/monixm/last-propulsion-project:latest
    restart: always
    env_file:
      - dev.env
    command: '/usr/sbin/sshd -D'
    ports:
      - '8003:8000'
      - '4777:22'
    depends_on:
      - database
    volumes:
      - ./frontend:/frontend
      - ./backend:/backend
      - ./media-files:/media-files
      - ./static-files:/static-files
      - ./scripts:/scripts

volumes:
  database:

Hi,

did you build that setup yourself, or where did you copy the parts? From a first peek, it looks like as if the GitLab runner is configured to use the ruby image as default, and none of .gitlab-ci.yml references a different Docker image.

Using Docker executor with image ruby:2.5 ...

This works fine up until you’re calling the build job - which calls the docker CLI command but Docker isn’t installed inside the container.

Funnily enough the push image job works, but push latest fails.

Screen Shot 2020-01-03 at 18.08.522280×438 32.1 KB

Can you modify the before_script section and add a docker info execution? Possibly you are running into a different error here with Docker.

before_script:
  - docker info
  - docker login -u "gitlab-ci-token" -p "$CI_BUILD_TOKEN" "$CI_REGISTRY"

Cheers,
Michael

I built the project with my team mates - it was a final project for a course I attended and it worked just fine but then I wanted to launch it on my own domain & in my private Gitlab (not the school dedicated one) and ever since then I have had this problem. I have just updated default.conf file with my domain and the pipeline broke on the lint stage with the same ‘docker: command not found’ error…

Ok, I managed to solve this issue - it was caused by the fact that the shared runners were activated (hence gitlab was using ruby instead of shell executor which I chose initially when setting up gitlab runners). uff… thank you so much for your willingness to help, Michael, I appreciate it.

Ah. Shell executor. Now that makes sense, it would have been my next guess that the runner setup now differs Glad you’ve solved it yourself, nifty CI pipeline overall

Can you tell me how did you solve it, I am having same issue

Hello, in my case, I went to my project Settings on Gitlab - CI/CD - Runners - deactivate Shared Runners. This solved the issue. I hope it fixes it for you too.

Hi, I am also new to this. And if I disable shared runner then it mentions that no runner found.
My code:

gitlab stages

docker-test:
  stage: docker-test
  script: docker info    # working fine


integration-test:
  image: gradle:alpine
  stage: integration-test
  script:
    - gradle itest --info   # not working

build.gradle code :

task setupTestDB(type :Exec){
    commandLine 'sh','./src/integration/scripts/testDBSetup.sh'
}

task itest(type: Test, dependsOn: "test") {
    description = 'Runs the integration tests.'
    dependsOn setupTestDB
    useTestNG()
    mustRunAfter test
}

testDBSetup.sh :

docker run --rm --name pg_test -d -p 5100:5432 -e POSTGRES_USER=pg -e POSTGRES_PASSWORD=aaaaaa -e POSTGRES_DB=db_test postgres:9.6

And I get the error

Task :setupTestDB FAILED
./src/integration/scripts/testDBSetup.sh: line 5: docker: not found

Hi Michael,now I have the same issue,and I used the Docker executor with image ruby:2.6,the issue comes when I use it in gitlab-ci.yml。。。。May I ask your help?

I encountered a similar error during my CI/CD pipeline execution:
$ echo “Executing CI/CD pipeline”
Executing CI/CD pipeline
$ echo “$CI_REGISTRY_PASSWORD” | docker login -u “$CI_REGISTRY_USER” --password-stdin $CONTAINER_REGISTRY
/usr/bin/bash: line 140: docker: command not found
Cleaning up project directory and file based variables 00:00
ERROR: Job failed: exit code 1

Here is my script:

yaml

stages:

• build
• deploy

variables:
CONTAINER_REGISTRY: docker.io # Docker Hub
CONTAINER_IMAGE_BACKEND: $CONTAINER_REGISTRY/containerqueen/matieres:backend-latest
EKS_CLUSTER_NAME: matiere_cluster
AWS_DEFAULT_REGION: us-east-1

before_script:

• echo “Executing CI/CD pipeline”

MASTER

.server_rules_master: &server_rules_master
rules:

• if: ‘$CI_COMMIT_BRANCH == “master”’
  changes:
• ./**/*
• .gitlab-ci.yml

.deploy_rules_master: &deploy_rules_master
rules:

• if: ‘$CI_COMMIT_BRANCH == “master”’

######################################################### BUILD ###############################################

#Backend
build:
stage: build
image: docker:18.09
services:

• docker:18.09-dind
  variables:
  DOCKER_DRIVER: overlay2
  DOCKER_HOST: tcp://docker:2375
  IMAGE: ${CI_REGISTRY_IMAGE}/backend
  TAG: master
  FOLDER: .
  before_script:
• docker login -u $CI_REGISTRY_USER -p $CI_REGISTRY_PASSWORD $CI_REGISTRY
  script:
• docker build -t $IMAGE:$TAG -f $FOLDER/Dockerfile $FOLDER
• docker push $IMAGE:$TAG

######################################################### DEPLOY ###############################################

Backend

deploy:
stage: deploy
image: docker:latest
variables:
AWS_ACCESS_KEY_ID: $AWS_ACCESS_KEY_ID
AWS_SECRET_ACCESS_KEY: $AWS_SECRET_ACCESS_KEY
AWS_DEFAULT_REGION: us-east-1
script:

• echo “$CI_REGISTRY_PASSWORD” | docker login -u “$CI_REGISTRY_USER” --password-stdin $CONTAINER_REGISTRY
• docker pull $CONTAINER_IMAGE_BACKEND
• docker tag $CONTAINER_IMAGE_BACKEND $CONTAINER_IMAGE_BACKEND

- aws configure set aws_access_key_id $AWS_ACCESS_KEY_ID
- aws configure set aws_secret_access_key $AWS_SECRET_ACCESS_KEY
- aws configure set default.region $AWS_DEFAULT_REGION
- aws eks update-kubeconfig --region $AWS_DEFAULT_REGION --name $EKS_CLUSTER_NAME
- kubectl apply -f deployment.yml
- kubectl rollout restart deployment backend

Can someone help me solve this issue, please?