potentially dumb question: do I have to run

yarn deploy

after stopping the

yarn start

process? any changes I’ve made during the debug process are already live anyway.. functions, at least. or does

yarn deploy

unhook the stubs & other magic needed to run stuff locally?

It's the latter, yarn deploy is "real" deploy and will deploy the real code instead of stubs

got it. so if I ran

yarn start

and then nothing after that.. people using the application would probably run into issues since it’d be trying to run code on my machine?

yep haha

I wonder if either the docs should explain the bit about needing to run a deploy after live debug mode. or when you hit ^C, it makes it clear that you really need to deploy

Ah yeah that's a good point we can clarify that. Working on a process so we get around to these doc updates so we'll be able to address these things soon

I’m sure you guys have your hands full 🙂

fyi I had this question at first as well.

Curious about your workflow for this, what kind of environment do you need to switch between local debugging and normal? I actually go to the other extreme and have 2 environments even for my personal stacks, one always for local debugging and one always for a normal deployment

yeah, it may be a non issue really. I went with the 1-developer-per-AWS-account setup. so devs are always just using the “dev” stage, 100% isolated.. so I suppose they don’t really need to deploy

yeah that's what I have as well, the reason I ended up with 2 stages per developer was to allow sharing a link that other people can look at without their computer being up

but if you had just one stage, dev, and did

yarn deploy

, wouldn’t that allow others to look at things? since deploying unhooks the stubs/websocket connections to the dev’s local machine?

Yeah it would, the reason I like having them seperate is it's slower to unhook/rehook when you want to switch back and forth

that’s true. although I noticed the API URL doesn’t change between start mode & deploys

Yeah it'll stay consistent if you go between start/deploy

I see. I’m still trying to figure out how our setup is gonna work. obviously for the front end, devs will run that locally, and can set the API URL using .env.development.local or whatever, and point that to the dev SST in their AWS account. the tough part is, when they open a PR for the front end, it auto-deploys that to a temporary URL,

<http://pr-xxx.app.com|pr-xxx.app.com>

. I haven’t worked out how that will point to their dev setup yet

Yeah this is a classic problem when be/fe is decoupled, where to point FE for PR branches