Research
Security News
Malicious npm Packages Inject SSH Backdoors via Typosquatted Libraries
Socket’s threat research team has detected six malicious npm packages typosquatting popular libraries to insert SSH backdoors.
Probably the easiest multi-factor for Django. Ships with standalone views, opinionated defaults and a very simple integration pathway to retrofit onto mature sites. Supports FIDO2/WebAuthn and TOTP authenticators, with removable fallback options for email, SMS, carrier pigeon, or whatever other token exchange you can think of. U2F has been removed in 0.6.
This is not a passwordless authentication system. django-multifactor is a second layer of defence.
FIDO2/WebAuthn is the big-ticket item for MFA. It allows the browser to interface with a myriad of biometric and secondary authentication factors.
This project targets modern stacks, officially supporting Python 3.8+ and Django 3.2+.
Python/Django | 2.2 | 3.2 | 4.0 | 4.1 | 4.2 | 5.0 | 5.1 |
---|---|---|---|---|---|---|---|
3.8 | Y | Y | Y | Y | N/A | N/A | N/A |
3.9 | Y | Y | Y | Y | N/A | N/A | N/A |
3.10 | N | Y | Y | Y | N/A | Y | Y |
3.11 | N | N | N | Y | Y | Y | Y |
3.12 | N | N | N | N | Y | Y | Y |
Install the package:
pip install django-multifactor
Add multifactor
to settings.INSTALLED_APPS
and override whichever setting you need.
MULTIFACTOR = {
'LOGIN_CALLBACK': False, # False, or dotted import path to function to process after successful authentication
'RECHECK': True, # Invalidate previous authorisations at random intervals
'RECHECK_MIN': 60 * 60 * 3, # No rechecks before 3 hours
'RECHECK_MAX': 60 * 60 * 6, # But within 6 hours
'FIDO_SERVER_ID': 'example.com', # Server ID for FIDO request
'FIDO_SERVER_NAME': 'Django App', # Human-readable name for FIDO request
'TOKEN_ISSUER_NAME': 'Django App', # TOTP token issuing name (to be shown in authenticator)
# Optional Keys - Only include these keys if you wish to deviate from the default actions
'LOGIN_MESSAGE': '<a href="{}">Manage multifactor settings</a>.', # {OPTIONAL} When set overloads the default post-login message.
'SHOW_LOGIN_MESSAGE': False, # {OPTIONAL} <bool> Set to False to not create a post-login message
}
Ensure that django.contrib.messages
is installed.
Include multifactor.urls
in your URLs. You can do this anywhere but I suggest somewhere similar to your login URLs, or underneath them, eg:
urlpatterns = [
path('admin/multifactor/', include('multifactor.urls')),
path('admin/', admin.site.urls),
...
]
And don't forget to run a ./manage.py collectstatic
before restarting Django.
At this stage any authenticated user can add a secondary factor to their account by visiting (eg) /admin/multifactor/
, but no view will require secondary authentication. django-multifactor gives you granular control to conditionally require certain users need a secondary factor on certain views. This is accomplished through the multifactor.decorators.multifactor_protected
decorator.
from multifactor.decorators import multifactor_protected
@multifactor_protected(factors=0, user_filter=None, max_age=0, advertise=False)
def my_view(request):
...
factors
is the minimum number of active, authenticated secondary factors. 0 will mean users will only be prompted if they have keys. It can also accept a lambda/function with one request argument that returns a number. This allows you to tune whether factors are required based on custom logic (eg if local IP return 0 else return 1)user_filter
can be a dictionary to be passed to User.objects.filter()
to see if the current user matches these conditions. If empty or None, it will match all users.max_age=600
will ensure the the user has authenticated with their secondary factor within 10 minutes. You can tweak this for higher security at the cost of inconvenience.advertise=True
will send an info-level message via django.contrib.messages with a link to the main django-multifactor page that allows them to add factors for future use. This is useful to increase optional uptake when introducing multifactor to an organisation.You can also wrap entire branches of your URLs using django-decorator-include
:
from decorator_include import decorator_include
from multifactor.decorators import multifactor_protected
urlpatterns = [
path('admin/multifactor/', include('multifactor.urls')),
path('admin/', decorator_include(multifactor_protected(factors=1), admin.site.urls)),
...
]
You can control the factors users can pick from in settings.MULTIFACTOR
:
MULTIFACTOR = {
# ...
'FACTORS': ['FIDO2', 'TOTP'], # <- this is the default
}
django-multifactor
has a fallback system that allows the user to be contacted via a number of sub-secure methods simultaneously. The rationale is that if somebody hacks their email account, they'll still know something is going on when they get an SMS. Providing sane options for your users is critical to security here. A poor fallback can undermine otherwise solid factors.
The included fallback uses user.email
to send an email. This now sends as plain+HTML. You can send just plain by setting settings.MULTIFACTOR.HTML_EMAIL
to False
.
You can plumb in additional functions to carry the OTP message over any other system you like. The function should look something like:
def send_carrier_pigeon(user, message):
bird = find_bird()
bird.attach(message)
bird.send(user.address)
return True # to indicate it sent
Then hook that into settings.MULTIFACTOR
:
MULTIFACTOR = {
# ...
'FALLBACKS': {
'email': (lambda user: user, 'multifactor.factors.fallback.send_email'),
'pigeon': (lambda user: user.address, 'path.to.send_carrier_pigeon'),
}
}
Now if the user selects the fallback option, they will receive an email and a pigeon. You can remove email by omitting that line. You can disable fallback entirely by setting FALLBACKS to an empty dict.
It's sometimes useful to be able to be able to conditionally bypass multifactor requirements. You might be in local testing, you might be in automated testing or impersonating other users. Deactivating a security layer has obvious risks but that's between you and your gods.
settings.MULTIFACTOR.BYPASS
accepts a single path to a function accepting a request. If that returns True, multifactor will bypass its normal checks on a page.
MULTIFACTOR = {
# ...
'BYPASS': 'path.to.bypass_when_impersonating'
}
These are relatively easy to implement:
def bypass_when_impersonating(request):
from loginas.utils import is_impersonated_session
if is_impersonated_session(request):
return True
def bypass_when_debug(request):
from django.config import settings
return settings.DEBUG
It's often useful to monitor which of your users is using django-multifactor and, in emergencies, critical to be able to turn their secondary factors off. We ship a opinionated mixin class that you can add to your existing UserAdmin definition.
from multifactor.admin import MultifactorUserAdmin
@admin.register(User)
class StaffAdmin(UserAdmin, MultifactorUserAdmin):
...
It adds a column to show if that user has active factors, a filter to just show those with or without, and an inline to allow admins to turn certain keys off for their users.
If you want to use the styles and form that django-multifactor supplies, your users may think they're on another site. To help there is an empty placeholder template multifactor/brand.html
that you can override in your project. This slots in just before the h1 title tag and has text-align: centre
as standard.
If you use HTML emails for your email fallback, you can create a multifactor/email.html
template (accepting user, message context variables).
You can use this to include your product logo, or an explanation.
FAQs
Drop-in multifactor authentication subsystem for Django.
We found that django-multifactor demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Research
Security News
Socket’s threat research team has detected six malicious npm packages typosquatting popular libraries to insert SSH backdoors.
Security News
MITRE's 2024 CWE Top 25 highlights critical software vulnerabilities like XSS, SQL Injection, and CSRF, reflecting shifts due to a refined ranking methodology.
Security News
In this segment of the Risky Business podcast, Feross Aboukhadijeh and Patrick Gray discuss the challenges of tracking malware discovered in open source softare.