1

Our CI/CD pipeline was working perfectly fine for years and since yesterday I stumbled upon this error, without explaination:

GitLab Runner - There has been a runner system failure, please try again

I don't know a lot about GitLab runner and I am not the one who configured it years ago.

To add more information, this is our .gitlab-ci.yml file:

image:
  name: clevercloud/clever-tools:latest
  entrypoint: ["/bin/sh", "-c"]

variables:
  CLEVER_KIND_APPLICATION: "node"
  CLEVER_ORGANIZATION_ID: ${CLEVER_ORGANIZATION_ID}
  CLEVER_SECRET: ${CLEVER_SECRET}
  CLEVER_TOKEN: ${CLEVER_TOKEN}
  STAGING_APP_NAME: $STAGING_APP_NAME
  STAGING_APP_ID: $STAGING_APP_ID

stages:
  - test_and_lint
  - deploy

test_and_lint:
  image: node:14
  stage: test_and_lint
  script:
    - yarn install
    - yarn lint
    - yarn test

deploy:staging:clevercloud:
  stage: deploy
  only:
    - staging
  environment:
    name: production-clever-cloud/${STAGING_APP_NAME}
    url: https://${STAGING_APP_NAME}.cleverapps.io
  script: |
    CREATE_CC_APP=false
    clever link $STAGING_APP_ID -o ${CLEVER_ORGANIZATION_ID}  || CREATE_CC_APP=true
    if ${CREATE_CC_APP}; then clever create --type ${CLEVER_KIND_APPLICATION} -o ${CLEVER_ORGANIZATION_ID} $STAGING_APP_NAME; sleep 10; fi
    clever domain add ${STAGING_APP_NAME}.cleverapps.io || true
    clever deploy -f

The runner configuration:

enter image description here

I tried various thing, including:

  • Creating a new runner, registering it and execute the job on it => I think I failed at configuring it, I got another error about docker not being launched
  • Clearing runner caches in "Pipelines" tab
  • Looking at various SO posts about this error => none of theme solved my issue.
  • Looking at this post mentionning a disk space issue.

For the last one, I couldn't loggin into the runner using SSH. It asks me a password and I don't know where to find it.

Clément Roig
  • 21
  • 3
  • 1
    If you can't login to the server, this is probably something you need to surface to the administrator of the runner/server who does have the ability to login and troubleshoot from that end. For a regular docker executor that was previously configured and working correctly, you would want to verify that you don't have any full disk partitions (especially the partition used for docker) and that the docker daemon is up and running and that you can successfully run `docker` commands. Check for orphaned containers/volumes that may need to be cleaned up manually. But it could be _anything_, really. – sytech Aug 15 '23 at 21:01

0 Answers0