React Developer
Since moving my repo from GitHub to Codeberg I deploy by uploading code, and every upload fails immediately with “Internal server error. Please try again.” The build never starts (buildSize 0, buildDuration 0). It happens via the CLI, the dashboard, and a raw API call, so I believe it is server-side.
Environment: Appwrite Cloud, CLI 27.2.1, node-22, entrypoint src/main.js
Project ID:694749fe0023ceb1b093
Functions: 6947eb01002458a585ec (original), main2 (CLI-created), 6acce9a6000ac77292aa (dashboard-created)
What works:
Redeploying an existing deployment (69bd0cf741d8f1d03ff3)
Creating a new function from the dashboard
What fails (always instantly):
CLI create-deployment on main2: 6accdd3803708b0a3edc (11,609 B)
Dashboard manual upload (tar.gz): 6acce1ca703476161990
Raw API upload: 6acceba284a63c00d826 (11,633 B)
Raw API upload, hello-world: 6accedd2f24dc8b045c1 (240 B)
The raw API calls return status “waiting” and then flip to “failed”.
Minimal reproduction:
src/main.js contains: export default async ({ res }) => res.text(“ok”);
package.json contains: {“type”:”module”}
Archived with tar -czf hello.tar.gz . and uploaded via POST /functions/{id}/deployments (multipart: entrypoint, activate, code).
Probably ruled out: archive size and contents, function state (fails on new functions), CLI bugs (raw API fails too), quota or outage (usage page clean, status page green), my code (redeploys of it build fine).
Could someone check the upload or build worker logs for deployment? My deploy pipeline is blocked, so any workaround would help too.
Thanks!
For transparency: I have used AI in the troubleshooting process. At this point I don't have any ideas anymore.