NullReferenceException when accessing #User.Identity.Name - asp.net-core

We are running an ASP net core 2.1 website. In the header of the page, we want to display the name of the current user. The variable is retrieved through the windows authentication scheme using the #User.Identity.Name reference and is assembled like this: ZH_MB\rohzeh, where ZH_MB is the domain and rohzeh the windows AD user. Because we just want the username, we split both loike this:
<p class="pull-right navbar-text"><span class="glyphicon glyphicon-user"></span> #User.Identity.Name.Split("\\")[1]</p>
Now when we run this code, we get the following error:
An unhandled exception occurred while processing the request.
NullReferenceException: Object reference not set to an instance of an
object. AspNetCore.Views_Shared__Layout.b__46_1() in
_Layout.cshtml, line 111
Line 111 is the code above. When I run it without the split part like this:
<p class="pull-right navbar-text"><span class="glyphicon glyphicon-user"></span> #User.Identity.Name</p>
It works just fine, except for the domain information that I don't want.
When i surround this code with a try/catch block, it works as expected, showing only the user name:
#try
{
<p class="pull-right navbar-text"><span class="glyphicon glyphicon-user"></span> #User.Identity.Name.Split("\\")[1]</p>
}
catch (NullReferenceException) { }
Any idea what the problem might be? My initial idea was: the information is just not ready when the page is rendered. But in this case, the second line of could should provide the same error.

It seems
your user is not authenticated yet or Name is null
or you didn't install ASP.NET Core/.NET Core: Runtime & Hosting Bundle for IIS
or you didn't configure all needed to use Windows Authentication
or forwarding Windows Authentication is set to false for some reasons
You can set it this way or others:
public void ConfigureServices(IServiceCollection services)
{
services.Configure<IISOptions>(options => options.ForwardWindowsAuthentication = true);
// Add framework services.
services.AddMvc();
}

Thanks, the answer of Chris Pratt did the trick. I could replace the try/catch with the "?" and the username is shown corrently.

Related

Calling WebAPI with authorisation from server side blazor

Apologies if this is a bit of a dumb question - I'm trying to get my head round security setup and don't have much experience in this area. Been reading as much as I can, but can't find a clear example for what I'm trying to do.
I have created the default server-side and wasm blazor projects from the visual studio templates, and have shared the wasm project so I can re-use both client-side and server-side as per Carl Franklin's article:
http://www.appvnext.com/blog/2020/2/2/reuse-blazor-wasm-ui-in-blazor-server
That all works fine.
Next, I repeat, but add the "Individual User Accounts" option to both projects when creating, set the db string to a shared identity database. Both work individually, however, when I share the client code, and call from server-side blazor, the webapi call fails with an "unauthorized" error.
So, in short, I'm logging into the server-side blazor project successfully. The failure happens when I attempt to call the webapi which now sits in a separate project (the WASM project) and so will run in a different domain (don't think I'm hitting cors problems yet). When I attempt to call the webapi I get the unauthorized error. When I run in WASM, it all works as expected.
Can someone please give me a pointer on what steps I need to take to get this working? Full code for the razor component is below...
#page "/fetchdata"
#using Microsoft.AspNetCore.Authorization
#using Microsoft.AspNetCore.Components.WebAssembly.Authentication
#using BlazorWasm.Shared
#attribute [Authorize]
#inject HttpClient Http
<h1>Weather forecast</h1>
<p>This component demonstrates fetching data from the server.</p>
#if (forecasts == null)
{
<p><em>Loading...</em></p>
}
else
{
<table class="table">
<thead>
<tr>
<th>Date</th>
<th>Temp. (C)</th>
<th>Temp. (F)</th>
<th>Summary</th>
</tr>
</thead>
<tbody>
#foreach (var forecast in forecasts)
{
<tr>
<td>#forecast.Date.ToShortDateString()</td>
<td>#forecast.TemperatureC</td>
<td>#forecast.TemperatureF</td>
<td>#forecast.Summary</td>
</tr>
}
</tbody>
</table>
}
#code {
private WeatherForecast[] forecasts;
protected override async Task OnInitializedAsync()
{
try
{
string url = "https://localhost:44378/WeatherForecast";
forecasts = await Http.GetFromJsonAsync<WeatherForecast[]>(url);
}
catch (AccessTokenNotAvailableException exception)
{
exception.Redirect();
}
catch(HttpRequestException exception)
{
string msg = exception.Message;
}
}
}

.NET CORE and ADFS no Claims Available

Migrating to .NET Core 3 from a 4.6 project and I'm not 100% sure I am implementing things properly.
I followed the steps in this article, making sure to configure startup.cs following the code sample under the "Use WS-Federation without ASP.NET Core Identity" section.
https://learn.microsoft.com/en-us/aspnet/core/security/authentication/ws-federation?view=aspnetcore-3.0
Login seems to be working in that I'm redirected to MS login and sent back to my app with the expected cookies, for example MSISAuthenticated. But, user.identity.isauthenticated is always false and I have no claims available. Is this expected behavior? Perhaps I'm not configuring things properly? Ideally I'd like to be able to check if a user is authenticated and access the claims.
I've come across a number of articles about adding policies based on groups, but how would [Authorize (Policy="SomeGroup")] even work if no claims are available?
ConfigureServices Code:
enter image description here
Configure Code:
enter image description here
Controller Action:
public IActionResult Index()
{
var identity = (ClaimsIdentity)User.Identity;
ViewBag.Claims = identity.Claims;
return View();
}
View Code:
#using System.Security.Claims;
#{
ViewBag.Title = "Home Page";
IEnumerable<Claim> claims = (IEnumerable<Claim>)ViewBag.Claims;
}
#if (User.Identity.IsAuthenticated)
{
<div class="jumbotron">
<h1>Successful Sign On!</h1>
</div>
<div class="row">
<div class="col-md-12">
<h2>WS Federation Services Claims</h2>
#foreach (Claim claim in claims)
{
<p>
<b>#(claim.Type.ToString())</b>
<br />
#(claim.Value.ToString()) (type: #(claim.ValueType.ToString()))
<hr />
</p>
}
</div>
</div>
}
else
{
<div class="jumbotron">
<h1>SSO Test</h1>
<p class="lead">To sign in using Microsoft's single sign-on service, click the button below.</p>
<p>Sign in ยป</p>
</div>
}
perhaps the fact is that you are not send the desired ResourceUrl to ADFS. Then ADFS considers the default resource and issues a token without claims.
See more info on 3 step in "High level AD FS authentication flow"
enter link description here
AD FS identifies the resource which the client wants to access through
the resource parameter passed in the auth request. If using MSAL
client library, then resource parameter is not sent. Instead the
resource url is sent as a part of the scope parameter: scope =
[resource url]//[scope values e.g., openid].

Identity.IsAuthenticated always true on azure

I'm currently working on a website with the ASP.net MVC framework for my own amusement and I decided
to stop working with a local database and to publish my web application on azure with it's associated database.
Now I have a strange issue with authentication. Identity.IsAuthenticated is always true.
At first I got the following error :
A claim of type 'http://schemas.xmlsoap.org/ws/2005/05/identity/claims/nameidentifier' or 'http://schemas.microsoft.com/accesscontrolservice/2010/07/claims/identityprovider' was not present on the provided ClaimsIdentity. To enable anti-forgery token support with claims-based authentication, please verify that the configured claims provider is providing both of these claims on the ClaimsIdentity instances it generates. If the configured claims provider instead uses a different claim type as a unique identifier, it can be configured by setting the static property AntiForgeryConfig.UniqueClaimTypeIdentifier.
Without understanding too much I found a solution which was to configure the AntiForgery token in the global.cs file
AntiForgeryConfig.UniqueClaimTypeIdentifier = ClaimTypes.Name;
At the point the error stopped showing but the user identified don't exist in the AspNetUsers table. Of course when I try to get info from the user in a section where authentification is needed the application crashes since no entry exists.
On my layout, I have a section of code to display some data if the user is authenticated
#if (User.Identity.IsAuthenticated)
{
using (Html.BeginForm("LogOff", "Account", FormMethod.Post, new { id = "logoutForm", #class = "navbar-right" }))
{
#Html.AntiForgeryToken()
<ul class="nav navbar-nav navbar-right">
<li>
#Html.ActionLink("Hello " + User.Identity.GetUserName() + "!", "Index", "Manage", routeValues: null, htmlAttributes: new { title = "Manage" })
</li>
<li>Log off</li>
</ul>
}
}
The User.Identity.GetUserName() always returns live.com#[MyAzureAccount]#outlook.com and it's impossible for me to logoff.
This seems like an identity issue but as I didn't touch the code generated when I started the project and as it works in local I would've expected it to work.
The solution was simply to download the publishing files from azure instead of filling by hand the fields for publishing. I don't really know why it worked but it solved the problem.

Supporting Individual User Accounts AND Organizational Accounts in MVC5 / ASP.Net Identity 2

I've created an ASP.Net MVC5 application, in which I have configured (and have working fine) Individual User Accounts via Google, Facebook, etc.
What I'd like to do is also support authentication against Azure Active Directory (Organizational Accounts). This would be for internal staff to be able to logon to the app as administrators.
All existing information/guides/documentation I've found typically deals with using one or the other. How would I enable them both together?
If there needs to be a separate logon form for each type of user, that would not be an issue.
EDIT:
I was looking at the Application configuration within Azure Active Directory portal, and notice that they define an "OAUTH 2.0 AUTHORIZATION ENDPOINT". Can MVC5 be configured within Startup.Auth.cs to use this?
I managed to implement this by doing the following:
First, adding a reference to the Microsoft.Owin.Security.OpenIdConnect Nuget package.
Second, configuring it in my Startup.Auth.cs:
app.UseOpenIdConnectAuthentication(new OpenIdConnectAuthenticationOptions
{
ClientId = "From the Azure Portal (see below)",
Authority = "https://login.windows.net/<domain>.onmicrosoft.com",
Notifications = new OpenIdConnectAuthenticationNotifications
{
RedirectToIdentityProvider = (ctx) =>
{
if (ctx.Request.Path.Value.EndsWith("ExternalLogin"))
{
string appBasePathUrl = ctx.Request.Scheme + "://" + ctx.Request.Host + ctx.Request.PathBase;
ctx.ProtocolMessage.RedirectUri = appBasePathUrl + "/";
ctx.ProtocolMessage.PostLogoutRedirectUri = appBasePathUrl;
}
else
{
ctx.State = NotificationResultState.Skipped;
ctx.HandleResponse();
}
return Task.FromResult(0);
}
},
Description = new AuthenticationDescription
{
AuthenticationType = "OpenIdConnect",
Caption = "SomeNameHere"
}
});
Third, I setup the application in the Azure Portal (classic):
Fourth, I added a separate logon page for admin users:
#using (Html.BeginForm("ExternalLogin", "Home"))
{
#Html.AntiForgeryToken()
<div class="ui basic segment">
<div class="ui list">
<div class="item">
<button type="submit" name="provider" value="OpenIdConnect" class="left floated huge ui button social">
<i class="windows icon"></i>
<span>My Org Name</span>
</button>
</div>
</div>
</div>
}
Fifth, the ExternalLogin action doesn't need to change - we just let OWIN middleware redirect us to the external login page. The flow would then direct the user back to the ExternalLoginCallback action.
Finally, in the ExternalLoginCallback action, I check the incoming claims to determine that the login was via Azure AD, and instead of calling into ASP.NET Identity, I construct my own ClaimsIdentity, which has all my (application specific) claim information which my application recognises as an admin user.
Now, admin users navigate to https://example.com/admin, click the login button, are redirected to the Azure AD login, and windup back at the application as an admin user.
Your best bet would be to leverage Azue AD Access Control Services (ACS) and setup the Identity Providers to include Azure AD, Facebook, et al. See the documentation here: http://azure.microsoft.com/en-us/documentation/articles/active-directory-dotnet-how-to-use-access-control/
ACS can indeed be used, however as you have already implemented Google/Facebook signin I recommend that you directly integrate with Azure AD instead of going through an intermediate STS like ACS/thinktecture.
If your app' signin experience involves the user clicking on "Signin with Google/Signin with Facebook" stickers - you can add "Signin with your company' account." (there's even a recommended branding style: http://msdn.microsoft.com/en-us/library/azure/dn132598.aspx
If you app performs realm discovery and forwards the user to the appropriate IdP (employing a text box saying something like "Enter your email address to sign in") - then you could add matching for your company name email addresses and forward those users to AAD.
In both cases, your application will issue an SSO request to SSO endpoint of AAD: https://login.windows.net/{company domain name/id}/{wsfed/saml/oauth2}. If you're using .Net, WSFed, this should see you through: http://msdn.microsoft.com/en-us/library/azure/dn151789.aspx. Look for the code:
SignInRequestMessage sirm = FederatedAuthentication.WSFederationAuthenticationModule.CreateSignInRequest("", HttpContext.Request.RawUrl, false);
result = Redirect(sirm.RequestUrl.ToString());
There's also an OpenIdConnect sample here: http://msdn.microsoft.com/en-us/library/azure/dn151789.aspx

signalR : /signalr/hubs is not generated

I can get this tutorial to work in a new project, but not in my existing project.
My project is an ASP.Net MVC 4 web application with the following attribute in the web.config file:
<appSettings>
<add key="webpages:Enabled" value="true"/>
</appSettings>
This is because my application is a Single-Page-Application, which uses AngularJS on the client side. The only page in my application is index.cshtml, to which I've added the relevant code for signalR:
<!-- signalR chat -->
<script src="~/Scripts/jquery.signalR-1.0.0.js"></script>
<!--Reference the autogenerated SignalR hub script. -->
<script src="/signalr/hubs"></script>
<!--Add script to update the page and send messages.-->
<script type="text/javascript">
$(function () {
// Declare a proxy to reference the hub.
var chat = $.connection.chatHub;
// Create a function that the hub can call to broadcast messages.
chat.client.broadcastMessage = function (name, message) {
// Html encode display name and message.
var encodedName = $('<div />').text(name).html();
var encodedMsg = $('<div />').text(message).html();
// Add the message to the page.
$('#discussion').append('<li><strong>' + encodedName
+ '</strong>: ' + encodedMsg + '</li>');
};
// Get the user name and store it to prepend to messages.
$('#displayname').val(prompt('Enter your name:', ''));
// Set initial focus to message input box.
$('#message').focus();
// Start the connection.
$.connection.hub.start().done(function () {
$('#sendmessage').click(function () {
// Call the Send method on the hub.
chat.server.send($('#displayname').val(), $('#message').val());
// Clear text box and reset focus for next comment.
$('#message').val('').focus();
});
});
});
</script>
Then I've got the ChatHub.cs file:
public class ChatHub : Hub
{
public void Send(string name, string message)
{
// Call the broadcastMessage method to update clients.
Clients.All.broadcastMessage(name, message);
}
}
And finally in the global.asax:
protected void Application_Start()
{
RouteTable.Routes.MapHubs();
BundleConfig.RegisterBundles(BundleTable.Bundles);
}
When I run the application, the /signalr/hubs file is not generated. I get a 404 when requesting the file, and it crashes on the line:
chat.client.broadcastMessage = function (name, message) { ....
because chat is null as the previous line did not find chatHub:
var chat = $.connection.chatHub;
Does anyone know what's wrong with my code ?
UPDATE
I have solved my problem by changing the line::
<script src="/signalr/hubs"></script>
to
<script src="~/signalr/hubs"></script>
I have solved my problem by changing the line::
<script src="/signalr/hubs"></script>
to
<script src="~/signalr/hubs"></script>
Also, the reason why /signalr/hubs are not generated is forget to Map SignalR in OWIN Startup Configuration.
public class Startup
{
public void Configuration(IAppBuilder appBuilder){
...
appBuilder.MapSignalR();
...
}
...
In my case, it was because my ChatHub class was not marked public.
I had a similar problem where the hubs file wasn't being generated. It looks like the OP was following the steps here. The way I fixed the problem had to do with the jquery includes. The tutorial I linked below was written with jquery 1.6.4 and jquery-signalr version 2.1.0. When Visual Studio generated the Scripts folder for me, it used jquery version 1.10.2 and jquery-signalr version 2.0.2.
The way I fixed this was simply to edit the index.html file. Note that you can use Chrome's javascript console window Ctrl+Shift+J to see errors.
For me the solution was to reinstall all the packages and restore all the dependecies.
Open nuget powershell console and use this command.
Update-Package -Reinstall
I'll like to add that the signalR Readme file have some note about this issue.
And also if your signalR page is in a PartialView some script should be place in the master page.
Please see http://go.microsoft.com/fwlink/?LinkId=272764 for more information on using SignalR.
Upgrading from 1.x to 2.0
-------------------------
Please see http://go.microsoft.com/fwlink/?LinkId=320578 for more information on how to
upgrade your SignalR 1.x application to 2.0.
Mapping the Hubs connection
----------------------------
To enable SignalR in your application, create a class called Startup with the following:
using Microsoft.Owin;
using Owin;
using MyWebApplication;
namespace MyWebApplication
{
public class Startup
{
public void Configuration(IAppBuilder app)
{
app.MapSignalR();
}
}
}
Getting Started
---------------
See http://www.asp.net/signalr/overview/getting-started for more information on how to get started.
Why does ~/signalr/hubs return 404 or Why do I get a JavaScript error: 'myhub is undefined'?
--------------------------------------------------------------------------------------------
This issue is generally due to a missing or invalid script reference to the auto-generated Hub JavaScript proxy at '~/signalr/hubs'.
Please make sure that the Hub route is registered before any other routes in your application.
In ASP.NET MVC 4 you can do the following:
<script src="~/signalr/hubs"></script>
If you're writing an ASP.NET MVC 3 application, make sure that you are using Url.Content for your script references:
<script src="#Url.Content("~/signalr/hubs")"></script>
If you're writing a regular ASP.NET application use ResolveClientUrl for your script references or register them via the ScriptManager
using a app root relative path (starting with a '~/'):
<script src='<%: ResolveClientUrl("~/signalr/hubs") %>'></script>
If the above still doesn't work, you may have an issue with routing and extensionless URLs. To fix this, ensure you have the latest
patches installed for IIS and ASP.NET.
In my case i was missing :
app.MapSignalR();
in public void Configuration(IAppBuilder app) function located in startup.cs
Some advice for those that start scaling out from the get go. In my case I was working on getting a remote client to work and never realized that
A. the tutorial example lists the web app(server) startup in a using statement, after which the web app disposes properly and no long exists.
I got rid of the using statement and keep a reference to the web app
for later disposal
B. the client has a different url than the server. the example relies on them having the same url. the "/signalr/hubs" is an endpoint run by the signalr server, called by the signalr client to get a script of all the hubs the server implements.
You need to include "http://myServerURL/signalr/hubs", rather than making it
relative to the client.
No lies. This tripped me up for a solid 2 weeks, because by some magic the solution worked anyways on my coworker's setup. This caused me to keep looking for IIS settings and firewall settings and CORS settings that must have been blocking the connection on my computer. I scavenged every last stack overflow question I could and the ultimate answer was that I should have just implemented a heartbeat monitor on the web app from the start.
Good luck and hopefully this saves other people some time.
See if Microsoft.Owin.Host.Systemweb nuget package is not installed, and therefore not building the dynamic js lib.
OwinStartup not firing