Keystone server crashes with too many requests

Discussed in https://github.com/keystonejs/keystone/discussions/7753

<div type='discussions-op-text'>

^{Originally posted by IamMattL August 1, 2022} Hi,

I find that the keystone server dies when you refresh the admin page a few times, which then i have to start up the server again. This happens in dev and production, is this a bug or a configuration issue? I’ve looked for this issue on github and the Docs but cant find anything about it.

This seems to be a linux issue as it doesn’t happen on windows.

Running on Ubuntu 20.04

Here is the console output

$ yarn dev
yarn run v1.22.19
$ keystone dev
✨ Starting Keystone
⭐️ Dev Server Starting on http://localhost:3000
⭐️ GraphQL API Starting on http://localhost:3000/api/graphql
✨ Generating GraphQL and Prisma schemas
✨ The database is already in sync with the Prisma schema.
✨ Connecting to the database
✨ Creating server
✅ GraphQL API ready
✨ Generating Admin UI code
✨ Preparing Admin UI app
event - compiled client and server successfully in 14.1s (1315 modules)
✅ Admin UI ready
wait  - compiling /api/__keystone_api_build...
event - compiled successfully in 285 ms (196 modules)
wait  - compiling /init (client and server)...
event - compiled client and server successfully in 581 ms (1341 modules)
Error: aborted
    at connResetException (node:internal/errors:691:14)
    at abortIncoming (node:_http_server:597:17)
    at socketOnClose (node:_http_server:591:3)
    at Socket.emit (node:events:532:35)
    at TCP.<anonymous> (node:net:687:12)
error Command failed with exit code 1.
info Visit https://yarnpkg.com/en/docs/cli/run for documentation about this command.

[dcousens]

Hi @jamie-lord and @IamMattL,

The following CVE and security advisory was issued for this problem upstream in nextjs, and was resolved as part of Keystone’s recent update to nextjs@v12.2.4.

• https://github.com/vercel/next.js/security/advisories/GHSA-wff4-fpwg-qqv3

Thanks for your patience,

[dcousens]

@jamie-lord @IamMattL, sorry for the vague update, we’ll elaborate on the problem soon. We intend to make a release today in which this problem should be fixed.