API Server cycling up and down after upgrading

I upgraded from 6.1.0 to 6.2.0 today and started noticing something strange. When I start the app with
yarn rw dev
I immediately get this error in the console

 ➜  Local:   http://localhost:8910/
web |   ➜  Network: http://192.168.7.155:8910/
web |   ➜  Network: http://192.168.7.164:8910/
gen | Generating full TypeScript definitions and GraphQL schemas
web | 2:09:58 PM [vite] http proxy error at /auth?method=getToken:
web | ⌛ API Server launching, please refresh your page...
web | 2:09:58 PM [vite] http proxy error at /graphql:
web | ⌛ API Server launching, please refresh your page...
gen | Done.

The app continues and loads. I was using the graphQL playground and a query would work then it wouldn’t work, then it would work again. I looked in the console and noticed the API was starting up, crashing, and then restarting. Here is what it is returning.

...Done importing in 319 ms
api | Took 338 ms
api | API listening on http://localhost:8911/
api | GraphQL endpoint at /graphql
api | 14:19:43 🌲 Server listening at http://[::]:8911
api | [addDir] 
api | Building... Took 109 ms
api | Debugger listening on ws://127.0.0.1:18911/d48c82da-3f9a-4bad-91ab-9bc6a75d5d7b
api | For help, see: https://nodejs.org/en/docs/inspector
api | Starting API Server...
api | Loading server config from xxxxxx/app/api/server.config.js 
api | 
api | Importing Server Functions... 
api | 14:19:50 🐛 GraphiQL and Introspection Config 
api | 🗒 Custom
api | {
api |   "defaultAllowedOperations": [
api |     "query",
api |     "mutation"
api |   ],
api |   "defaultError": "Something went wrong.",
api |   "disableIntrospection": false,
api |   "graphiQLEndpoint": "/graphql"
api | }
api | /graphql 315 ms
api | /auth 1 ms

I don’t know if I have two unrelated issues going on

  1. [vite] http proxy errors
  2. And the graphQL error

My gut tells me the proxy error is causing the graphQL issue. I did see one thing that said to add this to my .env file

RWJS_DEV_API_URL=“http://localhost”

I tried that and it didn’t have any effect.

I will keep searching and hopefully figure it out, but hopefully some of you that understand this a lot more than me can point me in the right direction.

I don’t know what the problem is here but I had some kind of the same problem after upgrading from 6.1 to 7.1. I added host = "0.0.0.0" to [web] and [api] in redwood.toml file and it will connect to graphql after a reload.