How to set OpenID url parameter in OWIN - asp.net-mvc

I have this basic setup:
Http Request -> Hardware OpenWRT Router -> Apache -> IIS. Between Apache and IIS there's a simple mod_proxy config like:
<VirtualHost *:80>
ServerAdmin me#mail.com
ServerName steam.domain.com
ProxyPass / http://192.168.1.9/ timeout=600 Keepalive=On
ProxyPassReverse / http://192.168.1.9/
</VirtualHost *:80>
The application itself hosted on 192.168.1.9 IIS works flawlessly except for one thing - when I've followed the guide from website then I'm having my Steam OpenID return url set to 192.168.1.9 instead of steam.domain.com. Host is in the OpenID query string, so I suppose that it can be changed somehow?
Edit: The issue is when I click Steam button on login view I end up with SteamCommunity login screen with Sign In through Steam at 192.168.1.69 instead of Sign In through Steam at steam.domain.com as per picture:

The return url is build on the request launched when the user click on the link to authenticate himself via Steam:
private string BuildReturnTo(string state)
{
return Request.Scheme + "://" + Request.Host +
RequestPathBase + Options.CallbackPath +
"?state=" + Uri.EscapeDataString(state);
}
In OpenIDAuthenticationHandler.cs.
So I think that the problem is just in dev (I assume the website is still in dev cause of the ip address 192.168, but I know nothing about Hardware OpenWRT Router and mod_proxy).

Related

How to use cookies between two docker containers (docker compose)

Graph of network structure
I am following the flows inspired by these Ben Awad videos: https://www.youtube.com/watch?v=iD49_NIQ-R4 https://www.youtube.com/watch?v=25GS0MLT8JU.
The general pattern is access-token in memory, refresh token as httponly cookie*. This seems pretty secure and dev friendly.
However since both my node frontend and my api backend are dockerized: during SSR I want to use the local connection to the backend, not through the DNS. By default this is a bridge network. This comes with a problem. Since the internal uri of the backend is http://backend, not http://localhost:8000 (or DNS name in production), the cookie does not apply to that domain, even though it really is the same app as we got the cookie from.
So: what is the best solution, and how do I implement it?
Ideas for solutions:
To not use local connection and let the frontend container use host network
To "rename" the local connection from http://backend to http://localhost
To somehow set two cookies, one for http://backend and one for localhost
Store the refresh token somewhere thats not a cookie
You can use Nginx to solve this problem, since the header of your frontend will include a cookie for verification when a request is sent to your backend server, you can bind each server to a different port on the host machine, then add a CNAME record on your domain control panel to direct all request sent to (let's as) api.mydomain.com, to serve mydomain.com then in your Nginx config, you can do something like this
Nginx Config
server {
server_name mydomain.com;
listen 80;
location '/' {
proxy_pass 'http://localhost:8000/';
}
}
server {
listen 80;
server_name api.mydomain.com;
location '/' {
proxy_pass 'http://localhost:7000/';
}
}
then you can use the svelte externalFetch to change the path on the server side, so when the request hit the server, instead of fetching the specified URL, you can override it with the local host URL, like this:
src/hooks.ts
export async function externalFetch(request) {
if (request.url.includes('api.mydomain.com')) {
const localPath = new URL(request.url.pathname, 'http://localhost:7000')
request = new Request(localPath.href, request);
}
}

Haproxy remove port number from URL

I have a backend trying to route traffic to a specific IP address and a port. The first hit on the url doesn't contain the port number. But the subsequesnt requests from within the website is not redirected properly.
Example: http://test.com has server 123.45.67.89:9080
When I try the URL http://test.com/login --> It is redirected correctly and I get the login page.
But once I give the login details and press OK. It is redirected to http://test.com:9080/loginSuccess.
The page is available under http://test.com/loginsuccess
This is the backend code and all the ways I have tried from different solution is commented.
backend lf_was_9080
acl auth_lf_was http_auth(lf_was_auth_list)
http-request auth realm lf_was_auth_list if !auth_lf_was
mode http
#TRY 1
#http-request redirect prefix http://test.com/login if { hdr(host) -i test.com:9080/login }
#TRY 2
# Clean the request and remove any existing header named X-Rewrite
http-request del-header X-REWRITE
# Copy the full request URL into X-Rewrite unchanged
http-request add-header X-REWRITE %[url] if { path_sub 9080/login }
# Change the X-REWRITE header to contain out new path
http-request replace-header X-REWRITE ^:9080/login(/.*)?$ /login\1 if { hdr_cnt(X-REWRITE) gt 0 }
# Perform the 301 redirect
http-request redirect code 301 location http://%[hdr(host)]%[hdr(X-REWRITE)] if { hdr_cnt(X-REWRITE) gt 0 }
#TRY 3
#reqrep ^([^\ :]*)\ /login/(.*) \1\ /login/\2
#TRY 4
#http-request redirect prefix https://test.com if { hdr(host) -i test.com:9080 }
server lf_was_9080 10.85.200.158:9080 check
I have also https redirect which works perfectly.
How can i rewrite the url without the port number in Haproxy?

Dynamic configuration inside initializer (devise by example)

I'm developing an Rails 4 app. Auth is against an MS ADFS.
This is my config/initializers/devise.rb (Only omniauth part)
config.omniauth :saml,
issuer: "https://#{ENV['APP_HOST_AND_PORT']}",
idp_sso_target_url: "https://yyy.yyy.yyy/adfs/ls",
assertion_consumer_service_url: "https://#{ENV['APP_HOST_AND_PORT']}/auth/saml/callback",
name_identifier_format: "urn:oasis:names:tc:SAML:1.1:nameid-format:emailAddress",
idp_cert: "xxxxxxxxxx"
This works fine but APP_HOST_AND_PORT need to be setted (of course.)
In this case, APP_HOST_AND_PORT match with HTTP host header + port, DNS domain + port... So I don't really need set this value. The app knows ( or will know ) this value.
But request.host (and request.port) or whatever aren't accessible yet.
There's an elegant way to do this?

Ajax call triggering base URL in Apache + Passenger integration mode

I am trying to configure multiple RAILS apps through httpd configuration file. Everything is working fine but the AJAX calls are triggering the wrong URL, for example if the application is configured as
http://localhost/helloapp/
and it has AJAX call as get "/say_hello"
it is trying to get "localhost/say_hello" instead of "localhost/helloapp/say_hello".
Below is my httpd configuration file located at '/etc/httpd/conf/httpd.conf
'. I am using centOS.
<VirtualHost *:80>
ServerName localhost
<Directory /var/www/html >
Allow from all
Options -MultiViews
# Uncomment this if you're on Apache >= 2.4:
#Require all granted
</Directory>
Alias /helloapp /var/www/html/hello_application/public
<Location /helloapp>
PassengerBaseURI /helloapp
PassengerAppRoot /var/www/html/hello_application
</Location>
<Directory /var/www/html/hello_application/public>
# MultiViews must be turned off.
Allow from all
Options -MultiViews
</Directory>
</VirtualHost>
Any time you make an HTTP request to a url beginning with /, for example /say_hello, it is implied that you are sending a request to the path you specify from the root URL. The root URL is always your domain, which in your case is http://localhost/.
So no matter how you configure your Apache server, when you make a request to /say_hello, your browser will direct it to http://localhost/say_hello. If you want to make a request to http://localhost/helloapp/say_hello, you will have to tell ajax to go to /helloapp/say_hello in your website code.
If your rails app is named "helloapp", you could get this in one of the ways described here to access your application name in your rails code. Once you have it you could store it in an instance variable, like #app_name, for your template or javascript (or wherever you are making the ajax request from). I think what you are looking for is something along these lines:
var url = "/<%= #app_name %>/say_hello";
// do your ajax request with your new url variable
One way is to modify the urls for all ajax call in all js files and adding #app_name in the urls.
If you don't want to edit all ajax urls in all js file.Write the following piece of code in a separate js file or in application.js file:
$.ajaxSetup({
beforeSend: function(data, settings) {
var url = "/#app_name"+ settings.url;
settings.url = url;
}
});

Laravel generate secure https URL from route

Edit
I can't really find a way to generate a secure URL from route name.
To get a full URL, I use
echo route('my_route_name');
But what to do, if I want a URL with https?
UPDATE: As pointed out in the comments, a simpler way of doing this would be adding URL::forceSchema('https'); for Laravel version between 4.2-5.3 or URL::forceScheme('https'); for version 5.4+ in the boot method of your AppServiceProvider file.
Old answer:
It's actually entirely possible and there's only one line of code needed to accomplish that.
Laravel doesn't check the presence of SSL by itself, it depends on Symfony. And there goes our key to making it believe that the current request is secure.
The thing is, we have to set the HTTPS server param to true and the easiest method is to paste the following code in the boot method of your AppServiceProvider:
$this->app['request']->server->set('HTTPS', true);
In my very case, I only need to force SSL in production, the local env should still work on http. This is how I force SSL only on production:
$this->app['request']->server->set('HTTPS', $this->app->environment() != 'local');
By the way, mind those terms, you may need them in the future.
Laravel 8
I recently resolved this by modifying this file:
app/Providers/AppServiceProvider.php
in the method boot() add the following:
URL::forceScheme('https');
Add the use in the top:
use Illuminate\Support\Facades\URL;
to work in your local environment you can leave it like this:
public function boot()
{
if(env('APP_ENV') !== 'local') {
URL::forceScheme('https');
}
}
Note: Don't forget to set your env variable APP_ENV with prod for the production file.
APP_ENV=prod
Actually turns out, that laravel doesn't care if url is secure or not, because it generates based on the current url. If you're on https page, route() will return secure url. If on http, then http:// url
The problem was, that Laravel didn't detect that https was enabled, which was due to faulty server configuration.
You can check if Laravel sees the current connection as https by calling Request::isSecure();
As I mentioned in a relevant question, I found 5 ways of how to generate secure URLs.
Configure your web server to redirect all non-secure requests to https. Example of a nginx config:
server {
listen 80 default_server;
listen [::]:80 default_server;
server_name example.com www.example.com;
return 301 https://example.com$request_uri;
}
Set your environment variable APP_URL using https:
APP_URL=https://example.com
Use helper secure_url() (Laravel5.6)
Add following string to AppServiceProvider::boot() method (for version 5.4+):
\Illuminate\Support\Facades\URL::forceScheme('https');
Implicitly set scheme for route group (Laravel5.6):
Route::group(['scheme' => 'https'], function () {
// Route::get(...)->name(...);
});
At the moment this way is not documented, but it works well.
I think there is only one way to do this.
To generate the secure URL to your named routes, you might want to pass in your route into the secure_url helper function.
secure_url(URL::route('your_route_name', [], false));
You can't really use the route helper function because it generates absolute URL (with http://) by default and it's http not the https version that you wanted
Laravel 5.x will generate secure URL via route() helper if it detects the incoming connection is secure. Problem usually happen if the app is hidden behind load balancer or proxy (e.g. Cloudflare) since the connection between app server and load balancer/proxy might not be secure.
I am using Laravel Forge + Cloudflare now and this is the easiest way I could find to enable app thinking incoming connection is secure (not sure about other proxy).
Generate self signed certificate (see https://www.digitalocean.com/community/tutorials/openssl-essentials-working-with-ssl-certificates-private-keys-and-csrs or http://www.selfsignedcertificate.com/)
In Forge panel, insert your private key and cert via Sites > your-site > SSL Certificates > Install Existing Certificate.
Activate
In CloudFlare panel, Crypto > SSL, choose “Full” (not strict)
Done (it will take few minutes for the change to get propagated)
In short, connection between client and Cloudflare is secured by Cloudflare's own SSL. Connection between app server and Cloudflare is protected via your generated cert (thus the app is seeing 'connection' as secure.
You can apply the same principle with other stacks.
Use secure_url:
secure_url(URL::route('your_route_name', [], false));
You will need to set URL::route to false in order to not return a full URL. Then use secure_url function generates a fully qualified HTTPS URL to the given path.
From the UrlGenerator interface you can use URL::route
string route(string $name, mixed $parameters = array(), bool $absolute = true)
Get the URL to a named route.
Parameters
string $name
mixed $parameters
bool $absolute
Return Value
string
https://laravel.com/api/5.4/Illuminate/Contracts/Routing/UrlGenerator.html
In most cases routes should be generated with the same scheme your site was loaded with. Laravel automatically detects if request has X-Forwarded-Proto header and uses it to decide which scheme to use in generated route URLs. If your site is behind reverse proxy then you should add reverse proxy IP address to list of trusted proxies. https://github.com/fideloper/TrustedProxy package helps to do this. It's included in Laravel 5.5. For example, my config/trustedproxy.php looks like:
<?php
return [
'proxies' => '*',
'headers' => [
]
];
I use it with nginx reverse proxy that has the following configuration:
server {
listen 80;
server_name example.com;
access_log /var/log/nginx/example.com_access.log;
error_log /var/log/nginx/example.com_error.log;
client_max_body_size 50m;
location / {
proxy_pass http://localhost:8002;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
}
listen 443 ssl;
ssl_certificate /etc/letsencrypt/live/example.com/fullchain.pem;
ssl_certificate_key /etc/letsencrypt/live/example.com/privkey.pem;
include /etc/letsencrypt/options-ssl-nginx.conf;
ssl_dhparam /etc/letsencrypt/ssl-dhparams.pem;
if ($scheme != "https") {
return 301 https://$host$request_uri;
}
}
Replace example.com with your domain. SSL certificates was provided by Let's Encrypt with certbot.
On laravel 5.5.*
You only need to add https on your .env file
as AppServiceProvider already had function that checks if your APP_URL or app.url on your config has https on it.
class AppServiceProvider extends ServiceProvider
{
public function boot()
{
\URL::forceRootUrl(\Config::get('app.url'));
if (str_contains(\Config::get('app.url'), 'https://')) {
\URL::forceScheme('https');
}
}
This is certainly old, but someone like me will dump over here one day.
In your .env file define the APP_URL to use https instead of using http. Because all laravel url are generated based on this variable.
APP_URL=https://example.com
and wherever you want you can just say
{{ URL::route('my.route', params) }}
Or
{{ route('my.route', params) }}
With make sure all the routes are generated with secure protocol, add in the boot method of AppServiceProvider class:
<?php
namespace App\Providers;
use Illuminate\Routing\UrlGenerator;
use Illuminate\Support\ServiceProvider;
class AppServiceProvider extends ServiceProvider
{
/**
* Bootstrap any application services.
*
* #return void
*/
public function boot(UrlGenerator $url)
{
if (config('app.production')) {
$url->forceScheme('https');
}
}
Just add your application domain with the https protocol in the APP_URL of your .env file.
APP_URL=https://example.com
Then run route:cache
For reference of future visitors:
The secure_url function doesn't correctly handle GET parameters. So, for example, if you want to convert the url that the user has visited into a secure url while retaining the GET fields, you need to use this:
secure_url(Request::path()).'?'.http_build_query(Input::all());
Particularly note the use of path() rather than url() - if you give it a full url, it doesn't replace the http at the start, making it efectively useless.
I came across this issue while trying to generate a route as form action in Blade using Laravel 5.4.
Then I hit upon secure_url(), so I tried
{{ secure_url(route('routename', $args)) }}
This still returned a non-secure URL. :-(
After digging through the code and adding some debug logs, I finally figured out that secure_url does not change the incoming url argument, if it's already an absolute URL (including the scheme).
Fortunately route has an absolute flag as the third argument, and returns a relative URL if $absolute is passed as false.
Assuming /a/{id}/b is a named route "a.b"
route('a.b', 1) : will return http://[domain]/a/1/b
route('a.b', 1, false) : will return /a/1/b
Joining the two I arrived at :
{{ secure_url(route('routename', $args, false)) }}
As expected it generated https://[domain]/routeXXX
:-)
I had a problem with redirect trailing slashes after 2 hours of looking for a bug, just need to remove
.htaccess
<IfModule mod_rewrite.c>
<IfModule mod_negotiation.c>
Options -MultiViews
</IfModule>
RewriteEngine On
# Redirect Trailing Slashes If Not A Folder...
RewriteCond %{REQUEST_FILENAME} !-d
RewriteRule ^(.*)/$ /$1 [L,R=301]
# Handle Front Controller...
RewriteCond %{REQUEST_FILENAME} !-d
RewriteCond %{REQUEST_FILENAME} !-f
RewriteRule ^ index.php [L]
</IfModule>
to
<IfModule mod_rewrite.c>
<IfModule mod_negotiation.c>
Options -MultiViews
</IfModule>
RewriteEngine On
# Handle Front Controller...
RewriteCond %{REQUEST_FILENAME} !-d
RewriteCond %{REQUEST_FILENAME} !-f
RewriteRule ^ index.php [L]
</IfModule>
If you are using Load Balancer, Laravel will never have the actual schema available.
So use https://stackoverflow.com/a/65691937/6489768. Working with Laravel - 9.x
Place this in your filters.php file and everywhere will be forced to https while retaining URL parameters:
//force ssl
App::before(function () {
if(!Request::secure() && App::environment() != 'local')
{
$baseHost = Request::getHttpHost();
$requestUri = Request::getRequestUri();
$newLink = 'https://'.$baseHost.$requestUri;
return Redirect::to($newLink);
}});
According to the laravel documentation on the url() helper method.
If no path is provided, a Illuminate\Routing\UrlGenerator instance is
returned
So you can use the secure method of the UrlGenerator class in the following way:
echo url()->secure('my_route_name');
To generate a secure (https) route use the following built-in 'before' filter called 'auth':
For example:
Route::get('your-route', ['before' => 'auth', 'uses' => YourController#yourAction']);
Now when you output your link it will be prepended with 'https'

Resources