Opened 5 years ago
Last modified 5 years ago
#30634 closed Bug
"manage.py runserver" does not respect SCRIPT_NAME header — at Initial Version
Reported by: | Oleg Kainov | Owned by: | nobody |
---|---|---|---|
Component: | HTTP handling | Version: | dev |
Severity: | Normal | Keywords: | |
Cc: | Triage Stage: | Unreviewed | |
Has patch: | no | Needs documentation: | no |
Needs tests: | no | Patch needs improvement: | no |
Easy pickings: | no | UI/UX: | no |
Description
- Configure local nginx to emulate prod-like environment (i.e. when Django is running in Docker) and app should be in subfolder
location /myapp/ { proxy_pass http://127.0.0.1:8000/; #proxy_set_header X-Script-Name /myapp; proxy_set_header SCRIPT_NAME /myapp; include proxy_params; }
- Run default Django app locally
manage.py runserver
- Debug any request and inspect request.META
- Make sure request.METASCRIPT_NAME is empty
In the same time seems that running with uWSGI or something similar works. SO I assume the error is somewhere in manage.py runserver
behavior
Background:
While I'm aware that runserver
considered "insecure" and "slow", it doesn't really matter for me when I'm deploying small internal apps in my company (imagine ~10 users-app with <5 views). Since with manage.py
I can run my app in "one click" and avoid hassle of separating static files serving and the main app, I would really like to keep it this way.