Gitlab CI Failed: NPM command not found

Question

I have been playing around Gitlabl CI but for some reason I can't get my tests to "passed". It always says npm: command not found

My Gitlab CI config looks like this:

image: node:latest
# This folder is cached between builds
# http://docs.gitlab.com/ce/ci/yaml/README.html#cache
cache:
  paths:
    - node_modules/

before_script:
  - npm install
  - npm install eslint -g
  - npm install eslint-plugin-react -g
  - npm install babel-eslint -g

test:lint:
  script:
    - eslint src/*

I keep getting the error below and I have No Idea why:

By the way, Im NOT using the gitlab shared runner. Not sure if that contributes to the problem but just to make sure, the machine that has my gitlab runner has all the necessary packages to run nodejs.

Your help is greatly appreciated

Best regards,

Flare

A naive question, but do you have node.js installed on your gitlab runner? – Selfinterest

Can you figure out under which OS account gitlab runner executes tests? After that you could try to login to that account and verify if npm is available. – Selfinterest

@Selfinterest Yes- as I mentioned the machine that has my gitlab runner has everything to run nodejs. That means I have nodejs installed. Thank you f – Flare

@Selfinterest it's under root account. On my runner.ssh I specify user as the root and Yes- it has nodejs installed too. – Flare

Maybe npm is not in in the PATH of root user? – Selfinterest

@Selfinterest it does. Just to make sure I also installed nodejs under gitlab-runner and added it to sudo group – Flare

Answer 1

The image tag specifies a docker image, hence you must specify the executor of your runner to be docker. Did you perhaps set it to be something else, like shell, when you created the runner?

Achilles

In my case I was using ssh instead of docker. – Message

Answer 2

You can use like below:-

stages:
          - build
          - deploy

deploy-prod:
          image: node:12.13.0-alpine
          stage: deploy
          script:
            - npm i -g firebase-tools

Marsala

Answer 3

I have same problem.

Gitlab-runner use user of 'gitlab-runner' default when it starts. So the user have not root access.

ps aux|grep gitlab-runner

1. copy the shell of running

2. change user: run /usr/bin/gitlab-runner run --working-directory /home/gitlab-runner --config /etc/gitlab-runner/config.toml --service gitlab-runner --syslog --user root in bash.

3. gitlab-ci runner test

4. pass!

5. kill old pid

6. nohup /usr/bin/gitlab-runner run --working-directory /home/gitlab-runner --config /etc/gitlab-runner/config.toml --service gitlab-runner --syslog --user root

ps: gitlab-runner -u root also change user.

Puri

Answer 4

This helped me:

n=$(which node);n=${n%/bin/node}; chmod -R 755 $n/bin/*; sudo cp -r $n/{bin,lib,share} /usr/local

Module

Answer 5

Your cli is taking ssh executer by default. You probably need docker. Try adding tag in your yml script.

 tags: [ <your docker image name> ]

Elwoodelwyn

Answer 6

If you are using sshpass to pass some command to your server, you'll have to update your .bashrc file (file path: ~/.bashrc).

In .bashrc file you see following code uncommented, comment it and it resolve your issue.

case $- in
    *i*) ;;
      *) return;;
esac

Nahuatl

Answer 7

This worked greatly for me

    before_script:
    - apt-get update -qy
    - apt-get install --no-install-recommends -y npm
    - npm --version
    - npx --version
    - gem install dpl
    - npm install

Source: https://forum.gitlab.com/t/bin-bash-line-131-npm-command-not-found/70632/3

Blaine

Answer 8

In my case, there are two gitlab-runner. There is no node enviroment in the specific Runner.Then I stopped the wrong one

Noellenoellyn