Running tests that require special setup

Jenkins spec

The jenkins_build_status_spec spins up a Jenkins instance in a Docker container based on an image stored in the GitLab-QA container registry. The Docker image it uses is preconfigured with some base data and plugins. The test then configures the GitLab plugin in Jenkins with a URL of the GitLab instance that will be used to run the tests. Unfortunately, the GitLab Jenkins plugin does not accept ports so http://localhost:3000 would not be accepted. Therefore, this requires us to run GitLab on port 80 or inside a Docker container.

To start a Docker container for GitLab based on the nightly image:

docker run \
  --publish 80:80 \
  --name gitlab \
  --hostname localhost \

To run the tests from the /qa directory:

CHROME_HEADLESS=false bin/qa Test::Instance::All http://localhost -- qa/specs/features/ee/browser_ui/3_create/jenkins/jenkins_build_status_spec.rb

The test will automatically spin up a Docker container for Jenkins and tear down once the test completes.

However, if you need to run Jenkins manually outside of the tests, use this command:

docker run \
  --hostname localhost \
  --name jenkins-server \
  --env JENKINS_HOME=jenkins_home \
  --publish 8080:8080 \

Jenkins will be available on http://localhost:8080.

Admin username is admin and password is password.

It is worth noting that this is not an orchestrated test. It is tagged with the :orchestrated meta only to prevent it from running in the pipelines for live environments such as Staging.


If Jenkins Docker container exits without providing any information in the logs, try increasing the memory used by the Docker Engine.

Gitaly Cluster tests

The tests tagged :gitaly_ha are orchestrated tests that can only be run against a set of Docker containers as configured and started by the Test::Integration::GitalyCluster GitLab QA scenario.

As described in the documentation about the scenario noted above, the following command will run the tests:

gitlab-qa Test::Integration::GitalyCluster EE

However, that will remove the containers after it finishes running the tests. If you would like to do further testing, for example, if you would like to run a single test via a debugger, you can use the --no-tests option to make gitlab-qa skip running the tests, and to leave the containers running so that you can continue to use them.

gitlab-qa Test::Integration::GitalyCluster EE --no-tests

When all the containers are running you will see the output of the docker ps command, showing on which ports the GitLab container can be accessed. For example:

CONTAINER ID   ...     PORTS                                    NAMES
d15d3386a0a8   ...     22/tcp, 443/tcp,>80/tcp   gitlab-gitaly-ha

That shows that the GitLab instance running in the gitlab-gitaly-ha container can be reached via http://localhost:32772. However, Git operations like cloning and pushing are performed against the URL revealed via the UI as the clone URL. It uses the hostname configured for the GitLab instance, which in this case matches the Docker container name and network, gitlab-gitaly-ha.test. Before you can run the tests you need to configure your computer to access the container via that address. One option is to use caddyserver as described for running tests against GDK.

Another option is to use NGINX.

In both cases you will need to configure your machine to translate gitlab-gitlab-ha.test into an appropriate IP address:

echo ' gitlab-gitaly-ha.test' | sudo tee -a /etc/hosts

Then install NGINX:

# on macOS
brew install nginx

# on Debian/Ubuntu
apt install nginx

# on Fedora
yum install nginx

Finally, configure NGINX to pass requests for gitlab-gitaly-ha.test to the GitLab instance:

# On Debian/Ubuntu, in /etc/nginx/sites-enabled/gitlab-cluster
# On macOS, in /usr/local/etc/nginx/nginx.conf

server {
  server_name gitlab-gitaly-ha.test;
  client_max_body_size 500m;

  location / {
    proxy_set_header Host gitlab-gitaly-ha.test;

Restart NGINX for the configuration to take effect. For example:

# On Debian/Ubuntu
sudo systemctl restart nginx

# on macOS
sudo nginx -s reload

You could then run the tests from the /qa directory:

CHROME_HEADLESS=false bin/qa Test::Instance::All http://gitlab-gitaly-ha.test -- --tag gitaly_ha

Once you have finished testing you can stop and remove the Docker containers:

docker stop gitlab-gitaly-ha praefect postgres gitaly3 gitaly2 gitaly1
docker rm gitlab-gitaly-ha praefect postgres gitaly3 gitaly2 gitaly1

Guide to run and debug Monitor tests

How to set up

To run the Monitor tests locally, against the GDK, please follow the preparation steps below:

  1. Complete the Prerequisites, at least through step 5. Note that the monitor tests do not require permissions to work with GKE because they use k3s as a Kubernetes cluster provider.
  2. The test setup deploys the app in a Kubernetes cluster, using the Auto DevOps deployment strategy. To enable Auto DevOps in GDK, follow the associated setup instructions. If you have problems, review the troubleshooting guide or reach out to the #gdk channel in the internal GitLab Slack.
  3. Do secure your GitLab instance since it is now publicly accessible on https://[YOUR-PORT]
  4. Install the Kubernetes command line tool known as kubectl. Use the official installation instructions.

You might see NGINX issues when you run gdk start or gdk restart. In that case, run sft login to revalidate your credentials and regain access the QA Tunnel.

How to run

Navigate to the folder in /your-gdk/gitlab/qa and issue the command:

QA_DEBUG=true CHROME_HEADLESS=false GITLAB_ADMIN_USERNAME=rootusername GITLAB_ADMIN_PASSWORD=rootpassword GITLAB_QA_ACCESS_TOKEN=your_token_here GITLAB_QA_ADMIN_ACCESS_TOKEN=your_token_here CLUSTER_API_URL=https://kubernetes.docker.internal:6443 bundle exec bin/qa Test::Instance::All https://[YOUR-PORT] -- qa/specs/features/browser_ui/8_monitor/all_monitor_core_features_spec.rb --tag kubernetes --tag orchestrated --tag requires_admin

The following includes more information on the command:

-QA_DEBUG - Set to true to verbosely log page object actions. -CHROME_HEADLESS - When running locally, set to false to allow Chrome tests to be visible - watch your tests being run. -GITLAB_ADMIN_USERNAME - Admin username to use when adding a license. -GITLAB_ADMIN_PASSWORD - Admin password to use when adding a license. -GITLAB_QA_ACCESS_TOKEN and GITLAB_QA_ADMIN_ACCESS_TOKEN - A valid personal access token with the api scope. This is used for API access during tests, and is used in the version that staging is currently running. The ADMIN_ACCESS_TOKEN is from a user with admin access. Used for API access as an admin during tests. -CLUSTER_API_URL - Use the address https://kubernetes.docker.internal:6443 . This address is used to enable the cluster to be network accessible while deploying using Auto DevOps. -https://[YOUR-PORT] - The address of your local GDK -qa/specs/features/browser_ui/8_monitor/all_monitor_core_features_spec.rb - The path to the monitor core specs ---tag - the meta-tags used to filter the specs correctly

At the moment of this writing, there are two specs which run monitor tests:

-qa/specs/features/browser_ui/8_monitor/all_monitor_core_features_spec.rb - has the specs of features in GitLab Core -qa/specs/features/ee/browser_ui/8_monitor/all_monitor_features_spec.rb - has the specs of features for paid GitLab (Enterprise Edition)

How to debug

The monitor tests follow this setup flow:

  1. Creates a k3s cluster on your local machine.
  2. Creates a project that has Auto DevOps enabled and uses an Express template (NodeJS) for the app to be deployed.
  3. Associates the created cluster to the project and installs GitLab Runner, Prometheus and Ingress which are the needed components for a successful deployment.
  4. Creates a CI pipeline with 2 jobs (build and production) to deploy the app on the Kubernetes cluster.
  5. Goes to Operation > Metrics menu to verify data is being received and the app is being monitored successfully.

The test requires a number of components. The setup requires time to collect the metrics of a real deployment. The complexity of the setup may lead to problems unrelated to the app. The following sections include common strategies to debug possible issues.

Deployment with Auto DevOps

When debugging issues in the CI or locally in the CLI, open the Kubernetes job in the pipeline. In the job log window, click on the top right icon labeled as “Show complete raw” to reveal raw job logs. You can now search through the logs for Job log, which matches delimited sections like this one:

------- Job log: -------

A Job log is a subsection within these logs, related to app deployment. We use two jobs: build and production. You can find the root causes of deployment failures in these logs, which can compromise the entire test. If a build job fails, the production job won’t run, and the test fails.

The long test setup does not take screenshots of failures, which is a known issue. However, if the spec fails (after a successful deployment) then you should be able to find screenshots which display the feature failure. To access them in CI, go to the main job log window, look on the left side panel’s Job artifacts section, and click Browse.

Common issues

Container Registry

When enabling Auto DevOps in the GDK, you may see issues with the Container Registry, which stores images of the app to be deployed.

You can access the Registry is available by opening an existing project. On the left hand menu, select Packages & Registries > Container Registries. If the Registry is available, this page should load normally.

Also, the Registry should be running in Docker:

$ docker ps

CONTAINER ID        IMAGE                                                                              COMMAND                  CREATED             STATUS              PORTS                    NAMES
f035f339506c   "/bin/sh -c 'exec /b…"   3 hours ago         Up 3 hours>5000/tcp   jovial_proskuriakova

The gdk status command shows if the registry is running:

run: ./services/registry: (pid 2662) 10875s, normally down; run: log: (pid 65148) 177993s
run: ./services/tunnel_gitlab: (pid 2650) 10875s, normally down; run: log: (pid 65154) 177993s
run: ./services/tunnel_registry: (pid 2651) 10875s, normally down; run: log: (pid 65155) 177993s

Also, restarting Docker and then, on the Terminal, issue the command docker login https://[YOUR-REGISTRY-PORT] and use the GDK credentials to login. Note that the Registry port and GDK port are not the same. When configuring Auto DevOps in GDK, the gdk reconfigure command outputs the port of the Registry:

Tunnel URLs

GitLab: https://[PORT]
Registry: https://[PORT]

These Tunnel URLs are used by the QA SSH Tunnel generated when enabling Auto DevOps on the GDK.

Pod Eviction

Pod eviction happens when a node in a Kubernetes cluster is running out of memory or disk. After many local deployments this issue can happen. The UI shows that installing Prometheus, GitLab Runner and Ingress failed. How to be sure it is an Eviction? While the test is running, open another Terminal window and debug the current Kubernetes cluster by kubectl get pods --all-namespaces. If you observe that Pods have Evicted status such as the install-runner here:

$ kubectl get pods --all-namespaces

NAMESPACE             NAME                                      READY   STATUS    RESTARTS   AGE
gitlab-managed-apps   install-ingress                           0/1     Pending   0          25s
gitlab-managed-apps   install-prometheus                        0/1     Pending   0          12s
gitlab-managed-apps   install-runner                            0/1     Evicted   0          75s

You can free some memory with either of the following commands: docker prune system or docker prune volume.

Geo tests

Geo end-to-end tests can run locally against a Geo GDK setup or on Geo spun up in Docker containers.

Using Geo GDK

Run from the qa/ directory with both GDK Geo primary and Geo secondary instances running:

CHROME_HEADLESS=false bundle exec bin/qa QA::EE::Scenario::Test::Geo --primary-address http://localhost:3001 --secondary-address http://localhost:3002 --without-setup

Using Geo in Docker

You can use GitLab-QA Orchestrator to orchestrate two GitLab containers and configure them as a Geo setup.

Geo requires an EE license. To visit the Geo sites in your browser, you will need a reverse proxy server (for example, NGINX).

  1. Export your EE license

    export EE_LICENSE=$(cat <path/to/your/gitlab_license>)
  2. (Optional) Pull the GitLab image

    This step is optional because pulling the Docker image is part of the Test::Integration::Geo orchestrated scenario. However, it’s easier to monitor the download progress if you pull the image first, and the scenario will skip this step after checking that the image is up to date.

    # For the most recent nightly image
    docker pull gitlab/gitlab-ee:nightly
    # For a specific release
    docker pull gitlab/gitlab-ee:13.0.10-ee.0
    # For a specific image
    docker pull
  3. Run the Test::Integration::Geo orchestrated scenario with the --no-teardown option to build the GitLab containers, configure the Geo setup, and run Geo end-to-end tests. Running the tests after the Geo setup is complete is optional; the containers will keep running after you stop the tests.

    # Using the most recent nightly image
    gitlab-qa Test::Integration::Geo EE --no-teardown
    # Using a specific GitLab release
    gitlab-qa Test::Integration::Geo EE:13.0.10-ee.0 --no-teardown
    # Using a full image address
    GITLAB_QA_ACCESS_TOKEN=your-token-here gitlab-qa Test::Integration::Geo --no-teardown

    You can use the --no-tests option to build the containers only, and then run the EE::Scenario::Test::Geo scenario from your GDK to complete setup and run tests. However, there might be configuration issues if your GDK and the containers are based on different GitLab versions. With the --no-teardown option, GitLab-QA uses the same GitLab version for the GitLab containers and the GitLab QA container used to configure the Geo setup.

  4. To visit the Geo sites in your browser, proxy requests to the hostnames used inside the containers. NGINX is used as the reverse proxy server for this example.

    Map the hostnames to the local IP in /etc/hosts file on your machine: gitlab-primary.geo gitlab-secondary.geo 

    Note the assigned ports:

    $ docker port gitlab-primary
    80/tcp ->
    $ docker port gitlab-secondary
    80/tcp ->

    Configure the reverse proxy server with the assigned ports in nginx.conf file (usually found in /usr/local/etc/nginx on a Mac):

    server {
      server_name gitlab-primary.geo;
      location / {
        proxy_pass http://localhost:32768; # Change port to your assigned port
        proxy_set_header Host gitlab-primary.geo;
    server {
      server_name gitlab-secondary.geo;
      location / {
        proxy_pass http://localhost:32769; # Change port to your assigned port
        proxy_set_header Host gitlab-secondary.geo;

    Start or reload the reverse proxy server:

    sudo nginx
    # or
    sudo nginx -s reload
  5. To run end-to-end tests from your local GDK, run the EE::Scenario::Test::Geo scenario from the gitlab/qa/ directory. Include --without-setup to skip the Geo configuration steps.

    QA_DEBUG=true GITLAB_QA_ACCESS_TOKEN=[add token here] GITLAB_QA_ADMIN_ACCESS_TOKEN=[add token here] bundle exec bin/qa QA::EE::Scenario::Test::Geo \
    --primary-address http://gitlab-primary.geo \
    --secondary-address http://gitlab-secondary.geo \

    If the containers need to be configured first (for example, if you used the --no-tests option in the previous step), run the QA::EE::Scenario::Test::Geo scenario as shown below to first do the Geo configuration steps, and then run Geo end-to-end tests. Make sure that EE_LICENSE is (still) defined in your shell session.

    QA_DEBUG=true bundle exec bin/qa QA::EE::Scenario::Test::Geo \
    --primary-address http://gitlab-primary.geo \
    --primary-name gitlab-primary \
    --secondary-address http://gitlab-secondary.geo \
    --secondary-name gitlab-secondary
  6. Stop and remove containers

    docker stop gitlab-primary gitlab-secondary
    docker rm gitlab-primary gitlab-secondary


  • You can find the full image address from a pipeline by following these instructions. You might be prompted to set the GITLAB_QA_ACCESS_TOKEN variable if you specify the full image address.
  • You can increase the wait time for replication by setting GEO_MAX_FILE_REPLICATION_TIME and GEO_MAX_DB_REPLICATION_TIME. The default is 120 seconds.
  • To save time during tests, create a Personal Access Token with API access on the Geo primary node, and pass that value in as GITLAB_QA_ACCESS_TOKEN and GITLAB_QA_ADMIN_ACCESS_TOKEN.