suggestions, but I still receive a 404 Not Found when trying to access http://localhost:5001/v1/ and the frontend still fails with "Failed to fetch plan info".

Current Config:

Local Frontend (.env.local): NEXT_PUBLIC_API_PREFIX=http://localhost:5001/console/api NEXT_PUBLIC_PUBLIC_API_PREFIX=http://localhost:5001/api

Docker Backend (.env): CONSOLE_API_URL=http://localhost:5001 SERVICE_API_URL=http://localhost:5001

Observations:

Even with the trailing slash, http://localhost:5001/v1/ returns a 404/Not Found in the browser.

The api container is running and port 5001 is mapped.

My worker CPU usage was extremely high (200%+) earlier but has stabilized now.

I have attached my redacted .env settings and the last 50 lines of the backend API logs. Could there be an issue with how the Gunicorn/Flask server is routing requests inside the container?

.env

CONSOLE_API_URL=http://localhost:5001
CONSOLE_WEB_URL=http://localhost:3000
SERVICE_API_URL=http://localhost:5001
APP_API_URL=http://localhost:5001
APP_WEB_URL=http://localhost:3000
FILES_URL=http://localhost:5001
CONSOLE_CORS_ALLOW_ORIGINS=*
WEB_API_CORS_ALLOW_ORIGINS=*

.env.local

NEXT_PUBLIC_API_PREFIX=http://localhost:5001/console/api
NEXT_PUBLIC_PUBLIC_API_PREFIX=http://localhost:5001/api

File "/app/api/.venv/lib/python3.12/site-packages/sqlalchemy/util/langhelpers.py", line 224, in exit
raise exc_value.with_traceback(exc_tb)
File "/app/api/.venv/lib/python3.12/site-packages/sqlalchemy/pool/base.py", line 895, in __connect
self.dbapi_connection = connection = pool._invoke_creator(self)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/api/.venv/lib/python3.12/site-packages/sqlalchemy/engine/create.py", line 661, in connect
return dialect.connect(*cargs, **cparams)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/api/.venv/lib/python3.12/site-packages/sqlalchemy/engine/default.py", line 629, in connect
return self.loaded_dbapi.connect(*cargs, **cparams) # type: ignore[no-any-return] # NOQA: E501
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/api/.venv/lib/python3.12/site-packages/psycopg2/init.py", line 122, in connect
conn = _connect(dsn, connection_factory=connection_factory, **kwasync)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
sqlalchemy.exc.OperationalError: (psycopg2.OperationalError) connection to server at "db_postgres" (172.19.0.2), port 5432 failed: FATAL: no pg_hba.conf entry for host "172.19.0.4", user "postgres", database "dify", no encryption

(Background on this error at: https://sqlalche.me/e/20/e3q8)
2025-12-29 17:32:00.304 INFO [Thread-293 (process_request_thread)] [_internal.py:97] - 172.19.0.1 - - [29/Dec/2025 17:32:00] "GET /console/api/workspaces HTTP/1.1" 500 -
2025-12-29 17:32:04.126 INFO [Thread-294 (process_request_thread)] [_internal.py:97] - 172.19.0.1 - - [29/Dec/2025 17:32:04] "GET /v1/ HTTP/1.1" 404 -

The /v1/ endpoint is always registered in the Dify backend and should return JSON if the API container is running properly—404 errors here almost always mean the request isn't reaching the backend API as expected, usually due to Docker or environment variable misconfiguration, not a missing route in the codebase source.

Your backend .env is using localhost for CONSOLE_API_URL, SERVICE_API_URL, APP_API_URL, and FILES_URL. In Docker, localhost refers to the container itself, not your host or other containers, so API requests will fail or misroute source. Update these variables to use either your host's LAN IP (e.g., http://192.168.x.x:5001) or the Docker service name (e.g., http://api:5001). Also, add DIFY_BIND_ADDRESS=0.0.0.0 to your backend .env so the API is accessible from outside the container source.

Make sure you don't include duplicate sub-paths (like /console/api) in both backend and frontend configs—these should only be in the frontend API prefix variables, not in the backend URLs source.

After updating your .env files, fully restart all Docker containers (docker-compose down && docker-compose up -d) and your frontend dev server. Then test http://localhost:5001/v1/ in your browser—it should return JSON, not HTML. If you still see 404s, check your Docker Compose and any proxy (nginx) configs to ensure requests to /v1/ and /console/api are routed to the backend API container, not the frontend source.

The database connection error (pg_hba.conf) is separate—it will cause 500 errors on endpoints that need the database, but it won't prevent /v1/ from being registered or responding source.

If you make changes to API prefix variables and previously built the frontend as a Docker image, rebuild the frontend image to avoid stale config source.

If issues persist, please share your updated .env files, docker-compose.yml, proxy configs, and full backend logs from startup for further review.

To reply, just mention @dosu.

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}