debian-mirror-gitlab/.gitlab/issue_templates/Experiment Rollout.md

Summary

This issue tracks the rollout and status of an experiment through to removal.

1. Experiment key / feature flag name: <experiment-key>
2. Epic or issue link: <issue or epic link>

This is an experiment rollout issue using the scoped experiment label. As well as defining the experiment rollout and cleanup, this issue incorporates the relevant Feature Flag Roll Out steps.

Owners

• Team: group::TEAM_NAME
• Most appropriate slack channel to reach out to: #g_TEAM_NAME
• Best individual to reach out to: NAME
• Product manager (PM): NAME

Stakeholders

Expectations

What are we expecting to happen?

What might happen if this goes wrong?

What can we monitor to detect problems with this?

Tracked data

Note: you can use the CXL calculator to determine if your experiment has reached significance. The calculator includes an estimate for how much longer an experiment must run for before reaching significance.

Rollout plan

• Runtime in days, or until we expect to reach statistical significance: 30
• We will roll this out behind a feature flag and expose this to <rollout-percentage>% of actors to start then ramp it up from there.

/chatops run feature set <experiment-key> <rollout-percentage> --actors

Status

Preferred workflow

The issue should be assigned to the Product manager (PM) or Engineer (Eng) as follows:

1. PM determines and manages the status of the experiment (assign this issue to the PM)
2. PM asks for initial rollout on production, or changes to the status (assign to an Eng)
3. Eng changes the status using chatops (reassign to the PM)
4. When concluded, PM updates the 'Roll Out Steps' and adds a milestone (assigns to an Eng)

The current status and history can be viewed using the:

• API (GitLab team members)
• Feature flag log (GitLab team members)
• Experiment rollout board

In this rollout issue, ensure the scoped experiment:: label is kept accurate.

Experiment Results

Roll Out Steps

• Confirm that end-to-end tests pass with the feature flag enabled. If there are failing tests, contact the relevant stable counterpart in the Quality department to collaborate in updating the tests or confirming that the failing tests are not caused by the changes behind the enabled feature flag.
• Enable on staging (/chatops run feature set <experiment-key> true --staging)
• Test on staging
• Ensure that documentation has been updated
• Enable on GitLab.com for individual groups/projects listed above and verify behaviour (/chatops run feature set --project=gitlab-org/gitlab feature_name true)
• Coordinate a time to enable the flag with the SRE oncall and release managers
  • In #production mention @sre-oncall and @release-managers. Once an SRE on call and Release Manager on call confirm, you can proceed with the rollout
• Announce on the issue an estimated time this will be enabled on GitLab.com
• Enable on GitLab.com by running chatops command in #production (/chatops run feature set feature_name true)
• Cross post chatops Slack command to #support_gitlab-com (more guidance when this is necessary in the dev docs) and in your team channel
• Announce on the issue that the flag has been enabled
• Remove experiment code and feature flag and add changelog entry - a separate cleanup issue might be required
• After the flag removal is deployed, clean up the feature flag by running chatops command in #production channel
• Assign to the product manager to update the knowledge base (if applicable)

Rollback Steps

• This feature can be disabled by running the following Chatops command:

/chatops run feature set <experiment-key> false

Experiment Successful Cleanup Concerns

Items to be considered if candidate experience is to become a permanent part of GitLab

/label ~"feature flag" ~"devops::growth" ~"growth experiment" ~"experiment-rollout" ~Engineering ~"workflow::scheduling" ~"experiment::pending" /milestone %"Next 1-3 releases"