Test-reports cause tests to fail

#1

Hello, I’m dealing with this weird issue when I do store_test_reports that make some tests fail. Said tests run and pass if I remove this step.
Here is my full config file:

aliases:
    # Environments
    - &test_environment
        working_directory: ~/alp
        docker:
            - image: circleci/ruby:2.3.4
              environment:
                RAILS_ENV: test
            - image: circleci/mysql:5.6-ram
              environment:
                MYSQL_ALLOW_EMPTY_PASSWORD: 'yes'
                MYSQL_ROOT_PASSWORD: ''
                MYSQL_DATABASE: circle_test
    # Steps
    - &restore_gem_cache
        keys: 
            - v1-gemfile-deps-{{ checksum "Gemfile.lock" }}
            - v1-gemfile-deps-

    - &save_gem_cache
        key: v1-gemfile-deps-{{ checksum "Gemfile.lock" }}
        paths: 
            - ~/alp/vendor/bundle
    
    - &bundle_install
        name: Install Gems
        command: bundle install --path=vendor/bundle --jobs=4 --retry=3  
    
    - &attach_workspace
        attach_workspace:
                at: ~/alp

    - &db_setup
        name: Setup DB
        command: bundle exec rake db:setup

    - &run_test
        name: Run Tests
        command: bundle exec rails test
    
    - &run_linter
        name: Run Rubocop Linter
        command: bundle exec rubocop --force-exclusion

    # Jobs
    - &dependencies
        - add_ssh_keys:
            fingerprints:
              [fingerprint]
        - checkout
        - *attach_workspace
        - restore_cache: *restore_gem_cache
        - run: *bundle_install
        - save_cache: *save_gem_cache
        - persist_to_workspace:
            root: .
            paths:
                - vendor/bundle

    - &database_and_run_tests
        - checkout
        - *attach_workspace
        - run: *bundle_install
        - run: cp ~/alp/config/database.yml.example ~/alp/config/database.yml
        - run: *db_setup
        - run: *run_linter
        - run: *run_test
        - store_test_results:
            path: test
        - store_artifacts:
            path: coverage

        
    

version: 2
workflows:
  version: 2
  build-and-test:
    jobs:
      - "Dependencies"
      - "Database and Test":
          requires:
            - "Dependencies"
          context: TestAlpenrose

jobs:
    "Dependencies":
        <<: *test_environment
        steps: *dependencies

    "Database and Test":
        <<: *test_environment
        steps: *database_and_run_tests

Any help would be amazing.
Thank you

#2

Hello. What I suspect is happening is your test runner exits with a code of 0 even when there are failed tests. To confirm this, can you expand the green test step and see if the same tests are failing in that step?

#3

Turns out the reason was unrelated to test-reports. This topic can be closed. Thanks

1 Like