[BUG] Login succeeds but frontend stays on “Redirecting” and dashboard route renders blank/404

[vleeuwenmenno]

Describe the bug
A valid login succeeds according to the API, but the UI never leaves the login screen. After submitting valid credentials, the login form shows "Success / Login successful! Redirecting..." and then stays there.

If I manually open /web/dashboard/, the page is blank/white.

The backend accepts the login request (POST /api/v1/auth/cookie/login returns 204) and immediately after that /api/v1/users/me can return 200, but frontend routes like /web/ and /web/dashboard/ are served with HTTP 404. Later dashboard API calls begin returning 401, and the UI appears stuck or blank.

To Reproduce
Steps to reproduce the behavior:

1. Deploy MediaManager v1.12.3 with Docker Compose.
2. Start with a Postgres-backed MediaManager setup and a default/admin user.
3. Go to http://server-ip:8000/web/login.
4. Log in with valid credentials.
5. See "Success / Login successful! Redirecting..." remain on the login screen.
6. Manually open http://server-ip:8000/web/dashboard/.
7. See a blank/white page.

Expected behavior
After a successful login, the browser should navigate to the dashboard and render the dashboard UI.

Directly opening /web/dashboard/ should also load the SPA route successfully.

Screenshots
Login page: (No redirect happens!)

After manually navigating to the dashboard page:

Server Logs
Relevant excerpts:

mediamanager | POST /api/v1/auth/cookie/login HTTP/1.1" 204
mediamanager | GET /api/v1/users/me HTTP/1.1" 200
mediamanager | GET /web/dashboard/ HTTP/1.1" 404
mediamanager | GET /web/_app/version.json HTTP/1.1" 304
mediamanager | GET /api/v1/users/me HTTP/1.1" 401
mediamanager | POST /api/v1/auth/cookie/logout HTTP/1.1" 401
mediamanager | GET /api/v1/tv/shows HTTP/1.1" 401
mediamanager | GET /api/v1/movies HTTP/1.1" 401

On startup, Postgres also logs repeated scheduler duplicate key errors:

mediamanager-postgres | ERROR: duplicate key value violates unique constraint "apscheduler_jobs_pkey"
mediamanager-postgres | DETAIL: Key (id)=(import_all_movie_torrents) already exists.
mediamanager-postgres | STATEMENT: INSERT INTO apscheduler_jobs (id, next_run_time, job_state) VALUES ($1::VARCHAR, $2, $3)

mediamanager-postgres | ERROR: duplicate key value violates unique constraint "apscheduler_jobs_pkey"
mediamanager-postgres | DETAIL: Key (id)=(import_all_show_torrents) already exists.
mediamanager-postgres | STATEMENT: INSERT INTO apscheduler_jobs (id, next_run_time, job_state) VALUES ($1::VARCHAR, $2, $3)

mediamanager-postgres | ERROR: duplicate key value violates unique constraint "apscheduler_jobs_pkey"
mediamanager-postgres | DETAIL: Key (id)=(auto_download_all_approved_season_requests) already exists.
mediamanager-postgres | STATEMENT: INSERT INTO apscheduler_jobs (id, next_run_time, job_state) VALUES ($1::VARCHAR, $2, $3)

mediamanager-postgres | ERROR: duplicate key value violates unique constraint "apscheduler_jobs_pkey"
mediamanager-postgres | DETAIL: Key (id)=(auto_download_all_approved_movie_requests) already exists.
mediamanager-postgres | STATEMENT: INSERT INTO apscheduler_jobs (id, next_run_time, job_state) VALUES ($1::VARCHAR, $2, $3)

mediamanager-postgres | ERROR: duplicate key value violates unique constraint "apscheduler_jobs_pkey"
mediamanager-postgres | DETAIL: Key (id)=(update_all_movies_metadata) already exists.
mediamanager-postgres | STATEMENT: INSERT INTO apscheduler_jobs (id, next_run_time, job_state) VALUES ($1::VARCHAR, $2, $3)

mediamanager-postgres | ERROR: duplicate key value violates unique constraint "apscheduler_jobs_pkey"
mediamanager-postgres | DETAIL: Key (id)=(update_all_non_ended_shows_metadata) already exists.
mediamanager-postgres | STATEMENT: INSERT INTO apscheduler_jobs (id, next_run_time, job_state) VALUES ($1::VARCHAR, $2, $3)

Filesystem check warning:

mediamanager | ERROR - media_manager.main - run_filesystem_checks(): Hardlink creation failed, falling back to copying files
mediamanager | Traceback (most recent call last):
mediamanager |   File "/app/media_manager/filesystem_checks.py", line 35, in run_filesystem_checks
mediamanager |     test_hardlink.hardlink_to(test_torrent_file)
mediamanager |   File "/usr/local/lib/python3.13/pathlib/_local.py", line 798, in hardlink_to
mediamanager |     os.link(target, self)
mediamanager | OSError: [Errno 18] Invalid cross-device link: '/downloads/.media_manager_test_dir/.media_manager.test.torrent' -> '/downloads/tvshows/.media_manager_test_dir/.media_manager.test.hardlink'

Earlier, before increasing Postgres/SQLAlchemy pool limits locally, the backend also produced:

sqlalchemy.exc.TimeoutError: QueuePool limit of size 5 overflow 10 reached, connection timed out, timeout 30.00

Browser Logs
Not captured yet.

Network-level observations:

• POST /api/v1/auth/cookie/login returns 204
• /api/v1/users/me returns 200 immediately after login
• /web/dashboard/ returns 404 while serving the SPA fallback HTML
• Later dashboard API calls return 401

Version
v1.12.3

Additional context
The same behavior occurs when accessing MediaManager directly through the published container port, without a Caddy reverse proxy in front of it. This does not appear to be reverse-proxy-specific.

The app appears to serve SvelteKit fallback HTML for /web/ and /web/dashboard/, but with HTTP status 404. That may be related to the blank dashboard and failed client-side routing after successful login.

docker-compose.yaml

services:
  mediamanager:
    image: quay.io/maxdorninger/mediamanager:latest
    container_name: mediamanager
    environment:
      - CONFIG_DIR=/app/config
      - TZ=Europe/Amsterdam
    ports:
      - "8000:8000"
    volumes:
      - /mnt/services/arr-stack/mediamanager-config:/app/config
      - /mnt/services/arr-stack/mediamanager-images:/app/images
      - /mnt/data:/downloads
    depends_on:
      mediamanager-db:
        condition: service_healthy
    restart: unless-stopped

  mediamanager-db:
    image: postgres:17
    container_name: mediamanager-postgres
    environment:
      POSTGRES_USER: MediaManager
      POSTGRES_DB: MediaManager
      POSTGRES_PASSWORD: REDACTED
    volumes:
      - /mnt/services/arr-stack/mediamanager-postgres:/var/lib/postgresql/data
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -d $${POSTGRES_DB} -U $${POSTGRES_USER}"]
      interval: 10s
      timeout: 5s
      retries: 5
    restart: unless-stopped

  prowlarr:
    image: linuxserver/prowlarr:latest
    container_name: prowlarr
    environment:
      - PUID=1000
      - PGID=100
      - TZ=Europe/Amsterdam
    volumes:
      - /mnt/services/arr-stack/prowlarr-config:/config
    ports:
      - "9696:9696"
    restart: unless-stopped

  flaresolverr:
    image: ghcr.io/flaresolverr/flaresolverr:latest
    container_name: flaresolverr
    environment:
      - LOG_LEVEL=info
      - LOG_HTML=false
      - CAPTCHA_SOLVER=none
      - TZ=Europe/Amsterdam
    ports:
      - "8191:8191"
    restart: unless-stopped

config.toml

[misc]
frontend_url = "http://server-ip:8000"
cors_urls = ["http://server-ip:8000"]
image_directory = "/app/images"
tv_directory = "/downloads/tvshows"
movie_directory = "/downloads/movies"
torrent_directory = "/downloads"
development = false

[[misc.tv_libraries]]
name = "TV Shows"
path = "/downloads/tvshows"

[[misc.movie_libraries]]
name = "Movies"
path = "/downloads/movies"

[database]
host = "mediamanager-db"
port = 5432
user = "MediaManager"
password = "REDACTED"
dbname = "MediaManager"

[auth]
email_password_resets = false
token_secret = "REDACTED"
session_lifetime = 86400
admin_emails = ["admin@example.com"]

[auth.openid_connect]
enabled = false
client_id = ""
client_secret = ""
configuration_endpoint = "https://openid.example.com/.well-known/openid-configuration"
name = "OpenID"

[notifications]

[notifications.smtp_config]
smtp_host = "smtp.example.com"
smtp_port = 587
smtp_user = "admin"
smtp_password = "admin"
from_email = "mediamanager@example.com"
use_tls = true

[notifications.email_notifications]
enabled = false
emails = []

[notifications.gotify]
enabled = false
api_key = ""
url = "https://gotify.example.com"

[notifications.ntfy]
enabled = false
url = "https://ntfy.sh/your-topic"

[notifications.pushover]
enabled = false
api_key = ""
user = ""

[torrents]

[torrents.qbittorrent]
enabled = false
host = "http://gluetun"
port = 8085
username = ""
password = ""

[torrents.transmission]
enabled = true
username = "REDACTED"
password = "REDACTED"
https_enabled = false
host = "gluetun"
port = 9091
path = "/transmission/rpc"

[torrents.sabnzbd]
enabled = true
host = "http://sabnzbd"
port = 8080
api_key = "REDACTED"
base_path = "/api"

[indexers]

[indexers.prowlarr]
enabled = true
url = "http://prowlarr:9696"
api_key = "REDACTED"
timeout_seconds = 60

[indexers.jackett]
enabled = false
url = "http://jackett:9117"
api_key = ""
indexers = []
timeout_seconds = 60

[[indexers.title_scoring_rules]]
name = "prefer_h265"
keywords = ["h265", "hevc", "x265", "h.265", "x.265"]
score_modifier = 100
negate = false

[[indexers.title_scoring_rules]]
name = "avoid_cam"
keywords = ["cam", "camrip", "bdscr", "ddc", "dvdscreener", "dvdscr", "hdcam", "hdtc", "hdts", "scr", "screener", "telesync", "ts", "webscreener", "tc", "telecine", "tvrip"]
score_modifier = -10000
negate = false

[[indexers.indexer_flag_scoring_rules]]
name = "prefer_freeleech"
flags = ["freeleech", "freeleech75"]
score_modifier = 100
negate = false

[[indexers.indexer_flag_scoring_rules]]
name = "reject_nuked"
flags = ["nuked"]
score_modifier = -10000
negate = false

[[indexers.scoring_rule_sets]]
name = "default"
libraries = ["ALL_TV", "ALL_MOVIES"]
rule_names = ["prefer_h265", "avoid_cam", "reject_nuked", "prefer_freeleech"]

[metadata]

[metadata.tmdb]
tmdb_relay_url = "https://metadata-relay.dorninger.co/tmdb"
primary_languages = [""]
default_language = "en"

[metadata.tvdb]
tvdb_relay_url = "https://metadata-relay.dorninger.co/tvdb"

• I understand, that without logs and/or screenshots and a detailed description of the problem, it is very hard to fix bugs.
• I have checked the documentation for help.
• I have searched the issues for similar issues and found none.

[coderabbitai]

🔗 Related PRs

#488 - Use openapi schemas directly [closed]
#543 - feat: add auth.registration_enabled flag and admin ui user creation [merged]

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!