• View jobs in a pipeline
  • View all jobs in a project
• See why a job failed
  • Troubleshoot a failed job with root cause analysis
• The order of jobs in a pipeline
• Job name limitations
• Group jobs in a pipeline
• Hide jobs
• Control the inheritance of default keywords and global variables
• Specifying variables when running manual jobs
• Delay a job
• View job logs in full screen mode
• Deployment jobs
• Troubleshooting
  • Job log slow to update
  • get_sources job section fails because of an HTTP/2 problem
  • Job using resource_group gets stuck

Jobs

Jobs are fundamental elements of a GitLab CI/CD pipeline. Jobs are configured in the .gitlab-ci.yml file with a list of commands to run to accomplish tasks like building, testing, or deploying code.

Jobs are:

• Defined with constraints stating under what conditions they should be executed.
• Top-level elements with an arbitrary name and must contain at least the script clause.
• Not limited in how many can be defined.

For example:

job1:
  script: "execute-script-for-job1"

job2:
  script: "execute-script-for-job2"

The above example is the simplest possible CI/CD configuration with two separate jobs, where each of the jobs executes a different command. Of course a command can execute code directly (./configure;make;make install) or run a script (test.sh) in the repository.

Jobs are picked up by runners and executed in the environment of the runner. What is important is that each job is run independently from each other.

View jobs in a pipeline

When you access a pipeline, you can see the related jobs for that pipeline.

Selecting an individual job shows you its job log, and allows you to:

• Cancel the job.
• Retry the job, if it failed.
• Run the job again, if it passed.
• Erase the job log.

View all jobs in a project

To view the full list of jobs that ran in a project:

1. On the left sidebar, select Search or go to and find your project.
2. Select Build > Jobs.

You can filter the list by job status and job name.

See why a job failed

When a pipeline fails or is allowed to fail, there are several places where you can find the reason:

• In the pipeline graph, in the pipeline details view.
• In the pipeline widgets, in the merge requests and commit pages.
• In the job views, in the global and detailed views of a job.

In each place, if you hover over the failed job you can see the reason it failed.

You can also see the reason it failed on the Job detail page.

Troubleshoot a failed job with root cause analysis

You can use root cause analysis in GitLab Duo Chat to troubleshoot failed CI/CD jobs.

The order of jobs in a pipeline

The order of jobs in a pipeline depends on the type of pipeline graph.

• For full pipeline graphs, jobs are sorted by name.
• For pipeline mini graphs, jobs are sorted by status, and then by name.

The job status order is:

• failed
• warning
• pending
• running
• manual
• scheduled
• canceled
• success
• skipped
• created

Job name limitations

You can’t use these keywords as job names:

• image
• services
• stages
• before_script
• after_script
• variables
• cache
• include
• pages:deploy configured for a deploy stage

Additionally, these names are valid when quoted, but are not recommended as they can make pipeline configuration unclear:

• "true":
• "false":
• "nil":

Job names must be 255 characters or fewer.

Use unique names for your jobs. If multiple jobs have the same name in a file, only one is added to the pipeline, and it’s difficult to predict which one is chosen. If the same job name is used in one or more included files, parameters are merged.

Group jobs in a pipeline

If you have many similar jobs, your pipeline graph becomes long and hard to read.

You can automatically group similar jobs together. If the job names are formatted in a certain way, they are collapsed into a single group in regular pipeline graphs (not the mini graphs).

You can recognize when a pipeline has grouped jobs if you don’t see the retry or cancel button inside them. Hovering over them shows the number of grouped jobs. Select to expand them.

To create a group of jobs, in the .gitlab-ci.yml file, separate each job name with a number and one of the following:

• A slash (/), for example, slash-test 1/3, slash-test 2/3, slash-test 3/3.
• A colon (:), for example, colon-test 1:3, colon-test 2:3, colon-test 3:3.
• A space, for example space-test 0 3, space-test 1 3, space-test 2 3.

You can use these symbols interchangeably.

In the example below, these three jobs are in a group named build ruby:

build ruby 1/3:
  stage: build
  script:
    - echo "ruby1"

build ruby 2/3:
  stage: build
  script:
    - echo "ruby2"

build ruby 3/3:
  stage: build
  script:
    - echo "ruby3"

The pipeline graph displays a group named build ruby with three jobs.

The jobs are ordered by comparing the numbers from left to right. You usually want the first number to be the index and the second number to be the total.

This regular expression evaluates the job names: ([\b\s:]+((\[.*\])|(\d+[\s:\/\\]+\d+))){1,3}\s*\z. One or more : [...], X Y, X/Y, or X\Y sequences are removed from the end of job names only. Matching substrings found at the beginning or in the middle of job names are not removed.

Hide jobs

To temporarily disable a job without deleting it from the configuration file:

• Comment out the job’s configuration:

  # hidden_job:
  #   script:
  #     - run test
  

• Start the job name with a dot (.) and it is not processed by GitLab CI/CD:

  .hidden_job:
    script:
      - run test
  

You can use hidden jobs that start with . as templates for reusable configuration with:

• The extends keyword.
• YAML anchors.

Control the inheritance of default keywords and global variables

You can control the inheritance of:

• default keywords with inherit:default.
• global variables with inherit:variables.

For example:

default:
  image: 'ruby:2.4'
  before_script:
    - echo Hello World

variables:
  DOMAIN: example.com
  WEBHOOK_URL: https://my-webhook.example.com

rubocop:
  inherit:
    default: false
    variables: false
  script: bundle exec rubocop

rspec:
  inherit:
    default: [image]
    variables: [WEBHOOK_URL]
  script: bundle exec rspec

capybara:
  inherit:
    variables: false
  script: bundle exec capybara

karma:
  inherit:
    default: true
    variables: [DOMAIN]
  script: karma

In this example:

• rubocop:
  • inherits: Nothing.
• rspec:
  • inherits: the default image and the WEBHOOK_URL variable.
  • does not inherit: the default before_script and the DOMAIN variable.
• capybara:
  • inherits: the default before_script and image.
  • does not inherit: the DOMAIN and WEBHOOK_URL variables.
• karma:
  • inherits: the default image and before_script, and the DOMAIN variable.
  • does not inherit: WEBHOOK_URL variable.

Specifying variables when running manual jobs

When running manual jobs you can supply additional job specific variables.

You can do this from the job page of the manual job you want to run with additional variables. To access this page, select the name of the manual job in the pipeline view, not Run ().

Define CI/CD variables here when you want to alter the execution of a job that uses CI/CD variables.

If you add a variable that is already defined in the CI/CD settings or .gitlab-ci.yml file, the variable is overridden with the new value. Any variables overridden by using this process are expanded and not masked.

Delay a job

When you do not want to run a job immediately, you can use the when:delayed keyword to delay a job’s execution for a certain period.

This is especially useful for timed incremental rollout where new code is rolled out gradually.

For example, if you start rolling out new code and:

• Users do not experience trouble, GitLab can automatically complete the deployment from 0% to 100%.
• Users experience trouble with the new code, you can stop the timed incremental rollout by canceling the pipeline and rolling back to the last stable version.

View job logs in full screen mode

You can view the contents of a job log in full screen mode by clicking Show full screen.

To use full screen mode, your web browser must also support it. If your web browser does not support full screen mode, then the option is not available.

Deployment jobs

Deployment jobs are a specific kind of CI job in that they deploy code to environments. A deployment job is any job that uses the environment keyword and the start environment action. Deployment jobs do not need to be in the deploy stage. The following deploy me job is an example of a deployment job. action: start is the default behavior and is defined for the sake of the example, but you can omit it:

deploy me:
  script:
    - deploy-to-cats.sh
  environment:
    name: production
    url: https://cats.example.com
    action: start

The behavior of deployment jobs can be controlled with deployment safety settings like preventing outdated deployment jobs and ensuring only one deployment job runs at a time.

Troubleshooting

Job log slow to update

When you visit the job log page for a running job, there could be a delay of up to 60 seconds before a log update. The default refresh time is 60 seconds, but after the log is viewed in the UI one time, log updates should occur every 3 seconds.

get_sources job section fails because of an HTTP/2 problem

Sometimes, jobs fail with the following cURL error:

++ git -c 'http.userAgent=gitlab-runner <version>' fetch origin +refs/pipelines/<id>:refs/pipelines/<id> ...
error: RPC failed; curl 16 HTTP/2 send again with decreased length
fatal: ...

You can work around this problem by configuring Git and libcurl to use HTTP/1.1. The configuration can be added to:

• A job’s pre_get_sources_script:

  job_name:
    hooks:
      pre_get_sources_script:
        - git config --global http.version "HTTP/1.1"
  

• The runner’s config.toml with Git configuration environment variables:

  [[runners]]
  ...
  environment = [
    "GIT_CONFIG_COUNT=1",
    "GIT_CONFIG_KEY_0=http.version",
    "GIT_CONFIG_VALUE_0=HTTP/1.1"
  ]
  

Job using resource_group gets stuck

If a job using resource_group gets stuck, a GitLab administrator can try run the following commands from the rails console:

# find resource group by name
resource_group = Project.find_by_full_path('...').resource_groups.find_by(key: 'the-group-name')
busy_resources = resource_group.resources.where('build_id IS NOT NULL')

# identify which builds are occupying the resource
# (I think it should be 1 as of today)
busy_resources.pluck(:build_id)

# it's good to check why this build is holding the resource.
# Is it stuck? Has it been forcefully dropped by the system?
# free up busy resources
busy_resources.update_all(build_id: nil)