Use of the Canary version
- Faye Muchery
Definition of the Canary version
The Canary version is a test version of Feedier, first released to a limited group of users to trial new features before a wider rollout deployment. It allows testing new functionalities and detecting potential bugs without affecting all users.
The term "Canary" comes from the canaries once used in coal mines to detect toxic gases. If the bird was affected, it warned miners of danger. Similarly, a Canary version helps find issues before they impact the whole system.
Why use a Canary version ?
✅Risk reduction: Identify issues before a full deployment.
✅Continuous improvement: Gather feedback and adjust the version accordingly.
✅Gradual deployment: Introduce new features in a controlled way.
✅Real-world testing: Validate compatibility with the production environment.
How to use a Canary version ?
🚨 Warning: The Canary version shares the same database as the production version.
When using the Canary version, it is crucial to understand that any data manipulation in this environment will have a direct impact on the production database.
🔹 What does this mean :
Changes are real: Any data added, modified, or deleted in the Canary version is instantly updated in production. There is no isolated environment.
Errors can be irreversible: An accidental deletion, incorrect update, or poorly calibrated test can impact ongoing operations and end users.
Tests must be controlled: It is essential to avoid destructive tests and not inject fake data that could pollute the production database.
Performance may be affected: Certain heavy actions (bulk imports, complex queries) can slow down the production system and impact the user experience.
🚫 What not to do :
❌Test scenarios for deleting reports, surveys, KPIs, or feedback.
❌ Modify critical data (reports, surveys, KPIs, etc.) without validation.
❌ Inject fake data that could distort analyses and KPIs.
❌ Share Canary access with untrained individuals.
✅ Best practices to follow :
✔ Follow the steps of the test according to the checklist established with the Feedier project team.
✔ Verify each action before executing it.
✔ Use filters to minimize the impact of tests.
✔ Communicate any user errors or incorrect actions to the Feedier team.