From the workshop
How symflip works, why it was built, and notes on shipping to a Linux box you own — without the downtime.
-
Anatomy of a release, step by step
What actually happens between clicking deploy and your new release serving traffic. A walk through every step symflip runs, and the one line that divides the safe half from the live half.
Read post -
How symflip handles your SSH keys and secrets
A deploy tool asks for the keys to your production box. Here's exactly what symflip does with your SSH keys and secrets, where they live, and the honest limits of the design.
Read post -
The one thing symflip can't undo: database migrations
symflip rolls back a bad release in one atomic step. But a symlink flip reverts your code, never your database schema. Here's the honest caveat, and the migration pattern that makes it a non-issue.
Read post -
Why I built symflip
Shipping to a single Linux box shouldn't need a Kubernetes cluster. Here's the story behind symflip, and how a zero-downtime release actually works.
Read post