Skip to main content
Archiving a service takes it out of the project without erasing that it was there. The service moves to the project’s Archived list, where its metrics, deploy history and logs stay readable. The Archived entry appears under Monitor in a project’s sidebar once the project has at least one.

​
What lands there

Both are the same thing from the archive’s point of view: a service that used to run and does not any more.

​
What is kept

An archived service has three tabs. Latency and request charts do not appear. Those are measured at the public edge against a hostname that no longer resolves. CPU and memory are recorded against the project and survive the service.

​
What is not kept

Archiving a service still:
  • brings its containers down and stops billing for them,
  • releases its public hostname for anyone to claim,
  • deletes its object-storage buckets and their contents.
Disks, project variables and project secrets are left alone, because they belong to the project rather than the service. Metrics and logs age out on their normal schedule. 30 days of history is stored, and your plan sets how far back you can read. An archived service older than that still lists, with empty charts.

​
Reusing a name

A name frees up the moment its service is archived, so you can create a new service with the same name immediately. The two do not get confused. Each keeps its own identity, so the new one’s charts start when it started and the archived one’s keep showing what it did.