Concepts

How convoy works.

• Sources

  How convoy verifies source endpoints and fetches the content that a Release renders.

• Clusters

  How convoy connects to a target cluster, when Releases can use it, and how its metadata reaches templates.

• Rendering

  How convoy turns fetched source files into the objects of a render, and how it identifies a render.

• Mutations

  The ordered pipeline that changes rendered objects before convoy applies them.

• Apply and prune

  How convoy checks, stores and applies a render, and how it deletes the objects that a render no longer contains.

• Sync policy

  When convoy applies a render, how a /sync request works, and when convoy reconciles a Release again.

• Drift

  How the observer compares the live objects with the desired render, and how automatic mode corrects drift.

• Health assessment

  How the observer decides if the resources of a Release are healthy, and how it reports the result.

• Sync waves

  How convoy orders the resources of one Release, and the Releases on one cluster.

• Lifecycle hooks

  How convoy runs Jobs and Pods on the target cluster before and after a sync, on a sync failure, and on deletion.

• Progressive rollout

  How convoy detects a failed rollout and rolls back to the last healthy render.

• Pin and history

  How convoy stores applied renders, keeps a history of them, and holds a Release on one render.

• Deletion

  What convoy does when you delete a Release or a Cluster.

• Status and conditions

  How to read the status of a Release, and which component writes each part.