This documentation is for Octopus Deploy Version 3.11. View other versions

Guided failures

Last updated

In an ideal world, every deployment would be successful. Sadly, things sometimes go wrong. Sometimes the cause of a failure is as simple as a missing dependency, Windows feature or configuration setting on a machine, a locked file or read-only path. The fastest way to go from broken to deployed is to let a human get involved, and when it makes sense, either ignore or retry the failing activity.

Guided failure mode tells Octopus that if something goes wrong during the deployment, instead of failing immediately, Octopus should ask for a human to intervene.

Enabling guided failure mode

Guided failure mode is off by default. You can enable it when deploying a release:

See the section on managing environments to enable guided failure mode by default for all deployments to a specific environment.

What happens

If something goes wrong during the deployment, Octopus will interrupt the deployment, and request guidance for how to handle the failure. This uses the same user experience that is used for manual steps (internally, requests for failure guidance, and manual steps, use the same implementation: we call them Interruptions in the REST API).

Anyone who has permission can handle the request for failure guidance by clicking assign to me. When assigned, they can choose how to handle the error.

You can choose to:

  • Retry: try the operation again. If it fails, another interruption will be raised.
  • Fail: mark the deployment as failed, don't try anything else.
  • Ignore: skip the operation, but keep going with the deployment.