This list has things I agree with and things I don’t. You could probably make another similar list that didn’t overlap at all. That’s because the real way to b

This list has things I agree with and things I don’t. You could probably make another similar list that didn’t overlap at all. That’s because the real way to build software that’s reliable has very little to do with a list of features/tools it has to do with designing for the sad paths at the same time as the happy paths. For each failure modality the app should have a conscious decision about what the app will do when the bad thing happens. This can be everything from “this will cause an outage” to “this will route to an alternate system”. What’s important is figuring out up front what those decisions mean for your data, your operations, your business and your users. Without that all the structured logs and version controlled configs in the world are just window dressing.