[capableweb]

The part about "untimely downtime" is about that CircleCI decides themselves when to push updates (which is the most common reason services has downtime), instead of you deciding when to upgrade/push updates. If you have a big migration/change coming up, you'd put pause on upgrading the CI/CD service as you don't want to muck with it while pushing out other organization-wide changes.

Granted, self-hosting comes with it's own share of problems too, no solution is a silver-bullet without any issues, but being able to "freeze" things to a stable mode helps to stabilize other processes.