Why:

With the current documentation a failover will be completed and access to the GitHub Enterprise Server will be restored, however a configuration cleanup will not be performed.

This means that Spokes, DPages, and Storage will maintain references to the old (potentially inaccessible) servers yet will be unable to replicate any data to them. This can and will eventually lead to clone/push failures and full repository outages should those servers not be re-added as replicas.

Solves github/docs-content#2564

What's being changed:

Our current documentation outlines the high-level steps to perform a failover, however more detail is needed clarifying follow-up actions in the event that that the primary server and/or other replicas will not be re-added to the GitHub Enterprise Server configuration.

I added a new (optional) step following the current step 7 noting the UUIDs of any servers which will not be re-added to the environment should be cleaned up by running ghe-repl-teardown -u <uuid_here>.

Check off the following:

• All of the tests are passing.
• I have reviewed my changes in staging.
• For content changes, I have reviewed the localization checklist
• For content changes, I have reviewed the Content style guide for GitHub Docs.

@maclarel
@myarb