WSGI/Django: pass username back to Apache for access log - python

My Django app, deployed in mod_wsgi under Apache using Django's standard WSGIHandler, authenticates users via form login on the Django side. So to Apache, the user is anonymous. This makes the Apache access log less useful.
Is there a way to pass the username back through the WSGI wrapper to Apache after handling the request, so that it appears in the Apache access log?
(Versions: Django 1.1.1, mod_wsgi 2.5, Apache 2.2.9)

You can only do it if using embedded mode and only if you use a separate package called apswigpy, which provides a Python binding for original Apache request object. The mod_wsgi package provides an optional mechanism for allowing original Apache request object to be passed as Python CObject reference in WSGI environment. You use that in conjunction with apswigpy something like:
from apache.httpd import request_rec
r = request_rec(environ['apache.request_rec'])
r.user = user
At least I think that will setup the appropriate information which access logging can then use.
You should really take this discussion over to the mod_wsgi mailing list.

You could use mod_auth_tkt. An auth_tkt is a signed cookie with the user id that Apache can understand. Your web application would have to set the cookie when the user logs in and out. Apache can derive a REMOTE_USER from the cookie, pass it to your web app or a non-Django web application running on the same server, include it in logs, whatever.

This probably isn't what you're expecting, but you could use the username in your URL scheme. That way the user will be in the path section of your apache logs.
You'd need to modify your authentication so that auth-required responses are obvious in the apache logs, otherwise when viewing the logs you may attribute unauthenticated requests to authenticated users. E.g. return a temporary redirect to the login page if the request isn't authenticated.

Correct me if I'm wrong, but what's stopping you from creating some custom middleware that sets a cookie equal to the display name of the current user logged in. This middleware will run on every view, so even though technically the user could spoof his username to display whatever he wants it to display, it'll just be reset anyway and it's not like its a security risk because the username itself is just for log purposes, not at all related to the actual user logged in. This seems like a simple enough solution, and then Apache log can access cookies so that gives you easiest access. I know some people wouldn't like the idea of a given user spoofing his own username, but i think this is the most trivial solution that gets the job done. Especially, in my case, when it's an iPhone app and the user doesn't have any direct access to a javascript console or the cookies itself.

for latest (Django 2.x, Apache 2.4) Tested
source https://www.django-rest-framework.org/api-guide/authentication/#apache-mod_wsgi-specific-configuration
you need to add WSGIPassAuthorization on in either server config, virtual host, directory or .htaccess

Related

SAML / Shibb authentication in Django

I am newbie to Django, but I know how to create a simple application in python-Django how to add new page , how to link it into url file etc.
Now what I am trying to do, I am trying to create a very simple webapp where On the landing page I will have a login link.
When the user clicks on this link it should go to george washington universities authentication window and then I can enter my university's credential and it should authenticate and come back to a page stating ** Login Successful**
I have gone through many tutorials, but all looks very confusing.
I have installed xmlsec1, pysaml2, djangosaml2 modules but even after that I was clueless what to do next. I never felt so much clueless like I am feeling for this authentication module.
It will be great if anyone can guide me with the process.
You didn't say what web server you were using but, on Apache, I'd recommend you use the mod_shib Apache module in conjunction with the Django authentication middleware.
In broad strokes, you are going to let Apache/mod_shib do the SAML heavy lifting and interact with the IdP and you are going to let Django manage users for you. You are going to connect the two by using a piece of Django authentication middleware that authenticates users using the REMOTE_USER environment variable to communicate between Apache and the Django app.
So, first setup Django using Django authentication as described in the Django documentation. Validate that you can create a user using the admin tools and that you can login and establish a session using the Django authentication methods.
Once you have simple, local login working, install the RemoteUser middleware and validate that, by setting the REMOTE_USER environment variable, you can cause your Django app to log a user in (you can do all this testing using the development server locally on your development machine).
Once you have demonstrated that you can log a user in by having the REMOTE_USER environment set, you need to install the Apache shibboleth module, mod_shib and use it to protect the root of your application.
Assuming your application is located at /mysite the config in your virtualhost section would include:
<Location /mysite>
ShibRequestSetting redirectToSSL 443
AuthType shibboleth
ShibRequestSetting requireSession 1
Require valid-user
</Location>
That configuration will tell apache that the /mysite path requires mod_shib to get involved and forces the communication over ssl/tls.
I will not go through all the configuration steps needed to install and configure shibboleth but, basically, you want to set your application default (shibboleth2.xml file) with REMOTE_USER=eppn (if you want to use another attribute like eptid you would specify that); this tells the module which attribute to stuff in the REMOTE_USER environment variable. Again, the shib doc is pretty clear here so I won't go into detail about how to redirect to your university IdP but, basically, you will create an entry in your Sessions section of the form:
<SSO entityID="https://idp.testshib.org/idp/shibboleth">
Where you would substitute your IdP location for the testshib URL shown above.
Note that we are setting REMOTE_USER to the eppn value and that that value will be interpreted by the Django auth middleware as the user's username; you will need to create Django users with usernames that are the same as their eppn for this to work. You can also allow Django to auto-provision new accounts if, for instance, you deem IdP authentication sufficient evidence to create a new user account but, with auto-provisioning, only the minimal bits get setup; you would still need to go into that account and set first name, last name, phone, etc.
The net effect is that, whenever an unauthenticated user tries to visit a location under /mysite, they will be redirected to your university IdP, they will logon there and be redirected back. The mod_shib module (in conjunction with the shibd daemon running in the background) will handle the attribute unpacking and the session state with the IdP and will set the eppn value in the REMOTE _USER environment variable. Assuming that your Django application was setup correctly with apache, it will be invoked and the RemoteUser middleware will use the eppn value set in the REMOTE_USER environment variable to lookup the user with that username in the authentication database. If it finds a user, it will complete the Django login process (i.e. set the user object in the request, set the session state, etc.)
One more thing. To talk to you university IdP and have it release attributes to your application (i.e. eppn), you will need to do three things:
Import their IdP metadata
Export your SP metadata and have your university identity folks import it
Get your university identity team to release eppn to you
Just be aware that those three steps can be a challenge and may take non-trivial time and homework.
One more one more thing: I would recommend verifying the SAML setup separate from your Django app/middleware integration. Using the simplest mechanism you are comfortable with (simple wsgi app, php script, whatever) create a page that will simply dump the REMOTE_USER environment variable when browsed then protect that first. Once you have that page redirecting to your IdP and dumping the correct eppn in REMOTE_USER on return, then you can move on to the Django bits.

How do you make a Redirect URI for a Slack App?

I'm trying to create a Slack App (see here), but I'm having incredible difficulty with how to create a Redirect URI.
Slack states the following:
You must specify at least one redirect URL for OAuth to work. If you
pass a URL in an OAuth request, it must (at least partially) match one
of the URLs you enter here. Learn more
I have a rudimentary understanding of a Redirect URI conceptually, but I have no idea how to go about actually getting this Redirect URI that Slack requires.
I've successfully used all of Slacks Integrations with Python including Real Time Messaging, but setting up a Redirect URI seems to require a special server or a website.
As already mentioned in the comments you will need a publicly reachable webserver to host your script for installing the Slack app. So the redirect URL is the URL to your installation script.
Basically any webserver or script hosting service that runs your favorite script flavor (e.g. PHP or Python) will work. See also this answer on how the OAUTH process can be implemented.
The redirect URL works without SSL, but for security reasons SSL is strongly recommended. Also many other features of Slack requires you to run SSL on your webserver (e.g. Interactive Buttons)
Another option is to run a webserver on your local machine (e.g. WAMP for windows) and open it to the Internet through a secure tunnel (e.g. ngrok). For developing and testing this is actually the better alternative, since you can test and fix your Slack app locally without having to deploy every change on a public server.
However for running a public Slack app (e.g. one that is listed on the Slack App Directory) I would strongly recommend to put the production version of your App on a public webserver.
If you're just trying to get it up so that you can authorize another workspace you can always use 'http://localhost' after authorizing it will try to redirect you there and you wont be able to see anything useful, but the authorization should still have taken place I believe.
of course if you're looking for the api code, you will have to pull it directly from the browser url. ... it's very manual.

How to wrap Plone authentication around a third-party servlet?

We're using Plone to serve up some third-party middle-ware.
Unfortunately the middle-ware has a particular servlet that gets invoked from a Java applet and doesn't do any kind of authentication. I would like to firewall this off and somehow wrap authentication around it, preferably using the existing session that users will have on Plone.
My first idea was to configure nginx (which we're using as the reverse proxy) to check the cookie and only proxy if the user has a valid session (along the lines of this example). However, how to check the session ID against Plone, since it's all stored in the Zope database?
Alternatively we could have a Plone python script that basically passes everything along to the back-end after authenticating, but I'm not sure how to do that.
Any suggestions? Or alternative ideas?
Plone uses a auth_tkt session cookie; any system that supports such session cookies and that is configured with the same secret as Plone can validate the cookies.
Unfortunately, I haven't been able to locate a nginx implementation of auth_tkt; the standard originates with the Apache mod_auth_tkt module, so if Apache is an option you could use that. See Plone & CGI single-sign-on using mod_auth_tkt for details on how to configure such a setup.
You can write your own Plone custom add-on which provides a browser view doing the session check for you inside Plone process. How to involve this from Nginx is a mystery for me.
http://collective-docs.readthedocs.org/en/latest/views/browserviews.html
Checking if the authentication credentials are valid
http://collective-docs.readthedocs.org/en/latest/sessions/login.html?highlight=login#extracting-credentials
More about Plone authentication and cookies
http://collective-docs.readthedocs.org/en/latest/sessions/cookies.html?highlight=cookies#default-plone-cookies

maintaining user authentication if i have some web pages on mod_python and some on mod_wsgi

I have a web application written in raw python and hosted on apache using mod_python. I am building another web application which is django based and will be hosted on same server using mod_wsgi.
Now, the scenerio is such that user will login from the web page which is using mod_python and a link will send him to my application which will be using mod_wsgi. My question is how can I maintain session? I need the same authentication to work for my application.
Thanks in advance.
If you're using django with mod_wsgi and a raw python page which only serve a link to django application, you don't need to maintain session on both page. When user click on first link and reach the django application, just check session there.
Django have session_db which use memcache. More information can be found here:
Django Sessions
SSO across web applications is poorly supported. One thing you can look at is:
http://www.openfusion.com.au/labs/mod_auth_tkt/
What you can do is really going to depend though on what authentication database you are currently using in the mod_python application and how you are remembering that someone is logged in. If you can provide that information, may be able to suggest other things.
Conceptually: store a cookie using your raw python web page that you process in a "welcome" view or custom middleware class in Django, and insert them into the sessions db. This is basically what hungnv suggests.
The most ridiculous way to do this would be to figure out how Django deals with sessions and session cookies, insert the correct row into Django's session database from your raw python app, and then custom-set the session cookie using Django's auth functions.

Django remote user authentication and security

I am using Django remote user authentication in a project. What I am actually using is just django.contrib.auth.RemoteUserBackend without the middleware, and manually calling authenticate after having checked with the backend that the user is legitimate.
Reading the source of the middleware, it seems that it just takes the username from a header in the request and then authenticates the user against the backend passing this username. The remote user backend, in turn, just merrily logs the user in with whatever username was passed. The user has then access to every area that requires a valid login.
Isn't this just a huge security flaw? How is this meant to be used?
In my case I should be safe, since the only call to authenticate comes after a successful remote identity verification, but I am wondering the reason why the middleware was introduced.
Let me turn this around on you: if you think this is a security flaw, then try writing an exploit that sets the REMOTE_USER header in a request to your app and see what happens.
REMOTE_USER dates back to the early days of the web when CGI pages were executed locally as the user you were hitting the web page with. REMOTE_USER is actually the name of a unix environment variable that denotes the active user. As security models for web servers changed, this scheme was preserved for compatibility. Now even IIS supports it to transparently handle Active Directory logins.
All user-passed headers begin with HTTP_. Otherwise, you couldn't trust on any header information, like SERVER_NAME, which would be an enormous mess.
Django 'merrily logs the user in' because your webserver has checked that the visitor has valid credentials for that username, and set the header accordingly.
If you trust your webserver (e.g. Apache) to set the REMOTE_USER (or other) header correctly, then it's not a security flaw.
You can see the documentation here. A user can't send a request with customer header for REMOTE_USER.
Warning
Be very careful if using a RemoteUserMiddleware subclass with a custom HTTP header. You must be sure that your front-end web server always sets or strips that header based on the appropriate authentication checks, never permitting an end-user to submit a fake (or “spoofed”) header value. Since the HTTP headers X-Auth-User and X-Auth_User (for example) both normalize to the HTTP_X_AUTH_USER key in request.META, you must also check that your web server doesn’t allow a spoofed header using underscores in place of dashes.
This warning doesn’t apply to RemoteUserMiddleware in its default configuration with header = 'REMOTE_USER', since a key that doesn’t start with HTTP_ in request.META can only be set by your WSGI server, not directly from an HTTP request header.

Categories