Fresh Docker Install - Can not log in // Backend error missing IX email

[Muttles]

HI All,

I am running a fresh install of Bracket via docker. I am unable to login to the front end with the test account and when I look at the logs from the backend I am seeing issues.
Backend Logs:

sqlalchemy.exc.ProgrammingError: (psycopg2.errors.UndefinedObject) index "ix_users_email" does not exist
[SQL:
DROP INDEX ix_users_email]
(Background on this error at: https://sqlalche.me/e/20/f405)
[2026-05-28 16:21:53 +0000] [10] [ERROR] Application startup failed. Exiting.
[2026-05-28 16:21:53 +0000] [10] [INFO] Worker exiting (pid: 10)
[2026-05-28 16:21:53 +0000] [9] [ERROR] Worker (pid:10) exited with code 3
[2026-05-28 16:21:53 +0000] [9] [ERROR] Shutting down: Master
[2026-05-28 16:21:53 +0000] [9] [ERROR] Reason: Worker failed to boot.

ProgressDB Logs:
PostgreSQL init process complete; ready for start up.
2026-05-28 16:18:23.116 UTC [1] LOG: starting PostgreSQL 18.4 (Debian 18.4-1.pgdg13+1) on x86_64-pc-linux-gnu, compiled by gcc (Debian 14.2.0-19) 14.2.0, 64-bit
2026-05-28 16:18:23.116 UTC [1] LOG: listening on IPv4 address "0.0.0.0", port 5432
2026-05-28 16:18:23.116 UTC [1] LOG: listening on IPv6 address "::", port 5432
2026-05-28 16:18:23.164 UTC [1] LOG: listening on Unix socket "/var/run/postgresql/.s.PGSQL.5432"
2026-05-28 16:18:23.267 UTC [75] LOG: database system was shut down at 2026-05-28 16:18:22 UTC
2026-05-28 16:18:23.308 UTC [1] LOG: database system is ready to accept connections
2026-05-28 16:18:27.510 UTC [89] ERROR: index "ix_users_email" does not exist
2026-05-28 16:18:27.510 UTC [89] STATEMENT:
DROP INDEX ix_users_email
2026-05-28 16:18:31.680 UTC [100] ERROR: index "ix_users_email" does not exist
2026-05-28 16:18:31.680 UTC [100] STATEMENT:
DROP INDEX ix_users_email
2026-05-28 16:18:35.859 UTC [111] ERROR: index "ix_users_email" does not exist
2026-05-28 16:18:35.859 UTC [111] STATEMENT:
DROP INDEX ix_users_email
2026-05-28 16:18:40.192 UTC [122] ERROR: index "ix_users_email" does not exist
2026-05-28 16:18:40.192 UTC [122] STATEMENT:

Here is the compose file I have used:

`services:

bracket-frontend:
    image: ghcr.io/evroon/bracket-frontend
    container_name: bracket-frontend
    ports:
        - "3001:3000"
    environment:
        NODE_ENV: "production"
        VITE_API_BASE_URL: "http://192.168.5.136:8400"
        VITE_HCAPTCHA_SITE_KEY: "10000000-ffff-ffff-ffff-000000000001"
    restart: unless-stopped

bracket-backend:
    image: ghcr.io/evroon/bracket-backend
    container_name: bracket-backend
    ports:
        - "8400:8400"
    environment:
        ENVIRONMENT: "PRODUCTION"
        PG_DSN: "postgresql://bracket_prod:password@postgres:5432/bracket_prod"
        CORS_ORIGINS: http://192.168.5.136:3001
        JWT_SECRET: SECRETCODESTUFF
    volumes:
        - /shared/docker/bracket/backend/static:/app/static
    restart: unless-stopped
    depends_on:
      - postgres

postgres:
    image: postgres
    restart: always
    environment:
      POSTGRES_DB: bracket_prod
      POSTGRES_USER: bracket_prod
      POSTGRES_PASSWORD: password
    volumes:
      - /shared/docker/postgres-bracket:/var/lib/postgresql`

Any suggestions?

[sueun-dev]

fwiw, I reproduced this with the backend image and a fresh Postgres DB.

Without ADMIN_EMAIL / ADMIN_PASSWORD, the backend exits during startup with the same DROP INDEX ix_users_email / index does not exist migration error. With those two env vars set on a fresh DB, it creates the tables, stamps alembic, creates the admin user, and the container goes healthy.

So for a fresh install, I’d add ADMIN_EMAIL and ADMIN_PASSWORD, then recreate only the empty Postgres volume/db. Don’t wipe it if there’s data you care about.

[Muttles]

Ok Thank you, this fixed the ix errors however I can still not authenticate.

Now when trying to login with the admin email I get the follow errors:
Console error:

[2026-05-28 20:49:35 +0000] [12] [WARNING] Invalid HTTP request received.

[sueun-dev]

I spun this up again with a fresh Postgres DB + ADMIN_EMAIL / ADMIN_PASSWORD. POST http://127.0.0.1:18400/token returned 200 with a bearer token, so the admin login path itself works after the DB init fix.

I can reproduce your exact backend warning by sending HTTPS to the plain HTTP backend port:

curl -k https://127.0.0.1:8400/token
# backend logs: Invalid HTTP request received.

So I’d check the browser Network tab for the login request URL. If it is https://...:8400/token, or a reverse proxy is forwarding TLS directly to port 8400, that’s the issue. Port 8400 is plain HTTP unless TLS is terminated before it.

One more thing I found: ghcr.io/evroon/bracket-frontend:latest has http://localhost:8400 baked into the static JS. Setting VITE_API_BASE_URL under compose environment: won’t rewrite an already-built Vite bundle. If you need http://192.168.5.136:8400, rebuild the frontend with that value at build time, or put the frontend/backend behind a proxy so the browser hits the right backend URL.

[Muttles]

I wish I understood how do rebuild the frontend for the docker container but that's well beyond what I understand. What seems to be the case is if I want users to use this on a browser thats not on my docker stack I can't as it can't resolve localhost and the vite enviorment issue is what I was trying to solve for.

Maybe you can explain the steps to build via docker so I can access from another computer? What I can't seem to understand is the step by step process to deploy via docker to a fresh docker environment and have it work when accessing from an external PC. If you have discord I'd be happy to screen share to show you from my end but i'm at a lose. I really want to use this to build out local competitive wargames as there are no other good free solutions but i'm at a real loss here on what to do next.