This site is the archived OWASP Foundation Wiki and is no longer accepting Account Requests.
To view the new OWASP Foundation website, please visit https://owasp.org
.NET Security Cheat Sheet
|
Last revision (mm/dd/yy): 09/20/2017 IntroductionThis page intends to provide quick basic .NET security tips for developers. The .NET FrameworkThe .NET Framework is Microsoft's principal platform for enterprise development. It is the supporting API for ASP.NET, Windows Desktop applications, Windows Communication Foundation services, SharePoint, Visual Studio Tools for Office and other technologies. Updating the FrameworkThe .NET Framework is kept up-to-date by Microsoft with the Windows Update service. Developers do not normally need to run seperate updates to the Framework. Windows update can be accessed at Windows Update or from the Windows Update program on a Windows computer. Individual frameworks can be kept up to date using NuGet. As Visual Studio prompts for updates, build it into your lifecycle. Remember that third party libraries have to be updated separately and not all of them use Nuget. ELMAH for instance, requires a separate update effort. .NET Framework GuidanceThe .NET Framework is the set of APIs that support an advanced type system, data, graphics, network, file handling and most of the rest of what is needed to write enterprise apps in the Microsoft ecosystem. It is a nearly ubiquitous library that is strong named and versioned at the assembly level. Data Access
Encryption
General
ASP.NET Web Forms GuidanceASP.NET Web Forms is the original browser-based application development API for the .NET framework, and is still the most common enterprise platform for web application development.
protected override OnInit(EventArgs e) {
base.OnInit(e);
ViewStateUserKey = Session.SessionID;
}
If you don't use Viewstate, then look to the default master page of the ASP.NET Web Forms default template for a manual anti-CSRF token using a double-submit cookie. private const string AntiXsrfTokenKey = "__AntiXsrfToken";
private const string AntiXsrfUserNameKey = "__AntiXsrfUserName";
private string _antiXsrfTokenValue;
protected void Page_Init(object sender, EventArgs e)
{
// The code below helps to protect against XSRF attacks
var requestCookie = Request.Cookies[AntiXsrfTokenKey];
Guid requestCookieGuidValue;
if (requestCookie != null && Guid.TryParse(requestCookie.Value, out requestCookieGuidValue))
{
// Use the Anti-XSRF token from the cookie
_antiXsrfTokenValue = requestCookie.Value;
Page.ViewStateUserKey = _antiXsrfTokenValue;
}
else
{
// Generate a new Anti-XSRF token and save to the cookie
_antiXsrfTokenValue = Guid.NewGuid().ToString("N");
Page.ViewStateUserKey = _antiXsrfTokenValue;
var responseCookie = new HttpCookie(AntiXsrfTokenKey)
{
HttpOnly = true,
Value = _antiXsrfTokenValue
};
if (FormsAuthentication.RequireSSL && Request.IsSecureConnection)
{
responseCookie.Secure = true;
}
Response.Cookies.Set(responseCookie);
}
Page.PreLoad += master_Page_PreLoad;
}
protected void master_Page_PreLoad(object sender, EventArgs e)
{
if (!IsPostBack)
{
// Set Anti-XSRF token
ViewState[AntiXsrfTokenKey] = Page.ViewStateUserKey;
ViewState[AntiXsrfUserNameKey] = Context.User.Identity.Name ?? String.Empty;
}
else
{
// Validate the Anti-XSRF token
if ((string)ViewState[AntiXsrfTokenKey] != _antiXsrfTokenValue ||
(string)ViewState[AntiXsrfUserNameKey] != (Context.User.Identity.Name ?? String.Empty))
{
throw new InvalidOperationException("Validation of Anti-XSRF token failed.");
}
}
}
<?xml version="1.0" encoding="UTF-8"?> <configuration>
<system.web>
<httpRuntime enableVersionHeader="false"/>
</system.web>
<system.webServer>
<security>
<requestFiltering removeServerHeader="true" />
</security>
<staticContent>
<clientCache cacheControlCustom="public" cacheControlMode="UseMaxAge" cacheControlMaxAge="1.00:00:00" setEtag="true" />
</staticContent>
<httpProtocol>
<customHeaders>
<add name="Content-Security-Policy" value="default-src 'none'; style-src 'self'; img-src 'self'; font-src 'self'" />
<add name="X-Content-Type-Options" value="NOSNIFF" />
<add name="X-Frame-Options" value="DENY" />
<add name="X-Permitted-Cross-Domain-Policies" value="master-only"/>
<add name="X-XSS-Protection" value="1; mode=block"/>
<remove name="X-Powered-By"/>
</customHeaders>
</httpProtocol>
<rewrite>
<rules>
<rule name="Redirect to https">
<match url="(.*)"/>
<conditions>
<add input="{HTTPS}" pattern="Off"/>
<add input="{REQUEST_METHOD}" pattern="^get$|^head$" />
</conditions>
<action type="Redirect" url="https://{HTTP_HOST}/{R:1}" redirectType="Permanent"/>
</rule>
</rules>
<outboundRules>
<rule name="Add HSTS Header" enabled="true">
<match serverVariable="RESPONSE_Strict_Transport_Security"
pattern=".*" />
<conditions>
<add input="{HTTPS}" pattern="on" ignoreCase="true" />
</conditions>
<action type="Rewrite" value="max-age=15768000" />
</rule>
</outboundRules>
</rewrite>
</system.webServer>
</configuration>
<httpRuntime enableVersionHeader="false" />
HttpContext.Current.Response.Headers.Remove("Server");
HTTP validation and encoding
Forms authentication
ASP.NET MVC GuidanceASP.NET MVC (Model-View-Controller) is a contemporary web application framework that uses more standardized HTTP communication than the Web Forms postback model. The OWASP Top 10 lists the most prevalent and dangerous threats to web security in the world today and is reviewed every 3 years. This section is based on this. Your approach to securing your web application should be to start at the top threat A1 below and work down, this will ensure that any time spent on security will be spent most effectively spent and cover the top threats first and lesser threats afterwards. After covering the top 10 it is generally advisable to assess for other threats or get a professionally completed Penetration Test.
DO: Using an object relational mapper (ORM) or stored procedures is the most effective way of countering the SQL Injection vulnerability. DO: Use parameterized queries where a direct sql query must be used. e.g. In entity frameworks: var sql = @"Update [User] SET FirstName = @FirstName WHERE Id = @Id";
context.Database.ExecuteSqlCommand(
sql,
new SqlParameter("@FirstName", firstname),
new SqlParameter("@Id", id));
DO NOT: Concatenate strings anywhere in your code and execute them against your database (Known as dynamic sql). NB: You can still accidentally do this with ORMs or Stored procedures so check everywhere. e.g string strQry = "SELECT * FROM Users WHERE UserName='" + txtUser.Text + "' AND Password='" + txtPassword.Text + "'"; EXEC strQry // SQL Injection vulnerability! DO: Practise Least Privilege - Connect to the database using an account with a minimum set of permissions required to do it's job i.e. not the sa account
Ensure cookies are sent via httpOnly: CookieHttpOnly = true, Reduce the time period a session can be stolen in by reducing session timeout and removing sliding expiration: ExpireTimeSpan = TimeSpan.FromMinutes(60),
SlidingExpiration = false
See here for full startup code snippet Ensure cookie is sent over https in the production environment. This should be enforced in the config transforms: <httpCookies requireSSL="true" xdt:Transform="SetAttributes(requireSSL)"/>
<authentication>
<forms requireSSL="true" xdt:Transform="SetAttributes(requireSSL)"/>
</authentication>
Protect LogOn, Registration and password reset methods against brute force attacks by throttling requests (see code below), consider also using ReCaptcha. [HttpPost]
[AllowAnonymous]
[ValidateAntiForgeryToken]
[AllowXRequestsEveryXSecondsAttribute(Name = "LogOn", Message = "You have performed this action more than {x} times in the last {n} seconds.", Requests = 3, Seconds = 60)]
public async Task<ActionResult> LogOn(LogOnViewModel model, string returnUrl)
DO NOT: Roll your own authentication or session management, use the one provided by .Net DO NOT: Tell someone if the account exists on LogOn, Registration or Password reset. Say something like 'Either the username or password was incorrect', or 'If this account exists then a reset token will be sent to the registered email address'. This protects against account enumeration. The feedback to the user should be identical whether or not the account exists, both in terms of content and behaviour: e.g. if the response takes 50% longer when the account is real then membership information can be guessed and tested.
DO NOT: Trust any data the user sends you, prefer white lists (always safe) over black lists You get encoding of all HTML content with MVC3, to properly encode all content whether HTML, javascript, CSS, LDAP etc use the Microsoft AntiXSS library: Install-Package AntiXSS then set in config: <system.web>
<httpRuntime targetFramework="4.5" enableVersionHeader="false" encoderType="Microsoft.Security.Application.AntiXssEncoder, AntiXssLibrary" maxRequestLength="4096" />
DO NOT: Use the [AllowHTML] attribute or helper class @Html.Raw unless you really know that the content you are writing to the browser is safe and has been escaped properly. DO: Enable a content security policy, this will prevent your pages from accessing assets it should not be able to access (e.g. a malicious script): <system.webServer>
<httpProtocol>
<customHeaders>
<add name="Content-Security-Policy" value="default-src 'none'; style-src 'self'; img-src 'self'; font-src 'self'; script-src 'self'" />
...
When you have a resource (object) which can be accessed by a reference (in the sample below this is the id) then you need to ensure that the user is intended to be there // Insecure
public ActionResult Edit(int id)
{
var user = _context.Users.FirstOrDefault(e => e.Id == id);
return View("Details", new UserViewModel(user);
}
// Secure
public ActionResult Edit(int id)
{
var user = _context.Users.FirstOrDefault(e => e.Id == id);
// Establish user has right to edit the details
if (user.Id != _userIdentity.GetUserId())
{
HandleErrorInfo error = new HandleErrorInfo(new Exception("INFO: You do not have permission to edit these details"));
return View("Error", error);
}
return View("Edit", new UserViewModel(user);
}
Ensure debug and trace are off in production. This can be enforced using web.config transforms: <compilation xdt:Transform="RemoveAttributes(debug)" /> <trace enabled="false" xdt:Transform="Replace"/> DO NOT: Use default passwords DO: (When using TLS) Redirect a request made over Http to https: In Global.asax.cs: protected void Application_BeginRequest() { #if !DEBUG
// SECURE: Ensure any request is returned over SSL/TLS in production
if (!Request.IsLocal && !Context.Request.IsSecureConnection) {
var redirect = Context.Request.Url.ToString().ToLower(CultureInfo.CurrentCulture).Replace("http:", "https:");
Response.Redirect(redirect);
}
#endif
}
DO NOT: Store encrypted passwords. DO: Use a strong hash to store password credentials. Use PBKDF2, BCrypt or SCrypt with at least 8000 iterations and a strong key. DO: Enforce passwords with a minimum complexity that will survive a dictionary attack i.e. longer passwords that use the full character set (numbers, symbols and letters) to increase the entropy. DO: Use a strong encryption routine such as AES-512 where personally identifiable data needs to be restored to it's original format. Do not encrypt passwords. Protect encryption keys more than any other asset. Apply the following test: Would you be happy leaving the data on a spreadsheet on a bus for everyone to read. Assume the attacker can get direct access to your database and protect it accordingly. DO: Use TLS 1.2 for your entire site. Get a free certificate from StartSSL.com or LetsEncrypt.org. DO NOT: Allow SSL, this is now obsolete DO: Have a strong TLS policy (see SSL Best Practises), use TLS 1.2 wherever possible. Then check the configuration using SSL Test DO: Ensure headers are not disclosing information about your application. See HttpHeaders.cs , Dionach StripHeaders or disable via web.config: <system.web>
<httpRuntime enableVersionHeader="false"/>
</system.web>
<system.webServer>
<security>
<requestFiltering removeServerHeader="true" />
</security>
<httpProtocol>
<customHeaders>
<add name="X-Content-Type-Options" value="NOSNIFF" />
<add name="X-Frame-Options" value="DENY" />
<add name="X-Permitted-Cross-Domain-Policies" value="master-only"/>
<add name="X-XSS-Protection" value="1; mode=block"/>
<remove name="X-Powered-By"/>
</customHeaders>
</httpProtocol>
DO: Authorize users on all externally facing endpoints. The .Net framework has many ways to authorize a user, use them at method level: [Authorize(Roles = "Admin")]
[HttpGet]
public ActionResult Index(int page = 1)
or better yet, at controller level: [Authorize]
public class UserController
You can also check roles in code using identity features in .net: System.Web.Security.Roles.IsUserInRole(userName, roleName)
DO: Send the anti-forgery token with every Post/Put request: using (Html.BeginForm("LogOff", "Account", FormMethod.Post, new { id = "logoutForm", @class = "pull-right" }))
{
@Html.AntiForgeryToken()
<ul class="nav nav-pills">
<li role="presentation">Logged on as @User.Identity.Name</li>
<li role="presentation"><a href="javascript:document.getElementById('logoutForm').submit()">Log off</a></li>
</ul>
}
Then validate it at the method or preferably the controller level: [HttpPost]
[ValidateAntiForgeryToken]
public ActionResult LogOff()
Make sure the tokens are removed completely for invalidation on logout. /// <summary>
/// SECURE: Remove any remaining cookies including Anti-CSRF cookie
/// </summary>
public void RemoveAntiForgeryCookie(Controller controller)
{
string[] allCookies = controller.Request.Cookies.AllKeys;
foreach (string cookie in allCookies)
{
if (controller.Response.Cookies[cookie] != null && cookie == "__RequestVerificationToken")
{
controller.Response.Cookies[cookie].Expires = DateTime.Now.AddDays(-1);
}
}
}
NB: You will need to attach the anti-forgery token to Ajax requests.
DO: Keep the .Net framework updated with the latest patches DO: Keep your NuGet packages up to date, many will contain their own vulnerabilities. DO: Run the OWASP Dependency checker against your application as part of your build process and act on any high level vulnerabilities. [OWASP Dependency Checker]
A protection against this was introduced in Mvc 3 template. Here is the code: public async Task<ActionResult> LogOn(LogOnViewModel model, string returnUrl)
{
if (ModelState.IsValid)
{
var logonResult = await _userManager.TryLogOnAsync(model.UserName, model.Password);
if (logonResult.Success)
{
await _userManager.LogOnAsync(logonResult.UserName, model.RememberMe);
return RedirectToLocal(returnUrl);
....
private ActionResult RedirectToLocal(string returnUrl)
{
if (Url.IsLocalUrl(returnUrl))
{
return Redirect(returnUrl);
}
else
{
return RedirectToAction("Landing", "Account");
}
}
Other advice:
More information: For more information on all of the above and code samples incorporated into a sample MVC5 application with an enhanced security baseline go to Security Essentials Baseline project XAML Guidance
Windows Forms Guidance
WCF Guidance
Authors and Primary EditorsBill Sempf - bill.sempf(at)owasp.org Other Cheatsheets |
