แก้ไข

แชร์ผ่าน


Web sign in with OpenID Connect in Azure Active Directory B2C

OpenID Connect is an authentication protocol, built on top of OAuth 2.0, that can be used to securely sign users in to web applications. By using the Azure Active Directory B2C (Azure AD B2C) implementation of OpenID Connect, you can outsource sign-up, sign in, and other identity management experiences in your web applications to Microsoft Entra ID. This guide shows you how to do so in a language-independent manner. It describes how to send and receive HTTP messages without using any of our open-source libraries.

Note

Most of the open-source authentication libraries acquire and validate the JWT tokens for your application. We recommend exploring those options, rather than implementing your own code. For more information, see Overview of the Microsoft Authentication Library (MSAL), and Microsoft Identity Web authentication library.

OpenID Connect extends the OAuth 2.0 authorization protocol for use as an authentication protocol. This authentication protocol allows you to perform single sign-on. It introduces the concept of an ID token, which allows the client to verify the identity of the user and obtain basic profile information about the user.

OpenID Connect also enables applications to securely acquire access tokens. You can use access tokens to access resources that the authorization server secures. We recommend OpenID Connect if you're building a web application that you host on a server and accessed through a browser. For more information about tokens, see the Overview of tokens in Azure Active Directory B2C

Azure AD B2C extends the standard OpenID Connect protocol to do more than simple authentication and authorization. It introduces the user flow parameter, which enables you to use OpenID Connect to add user experiences to your application, such as sign up, sign in, and profile management.

Send authentication requests

When your web application needs to authenticate the user and run a user flow, it can direct the user to the /authorize endpoint. The user takes action depending on the user flow.

In this request, the client indicates the permissions that it needs to acquire from the user in the scope parameter, and specifies the user flow to run. To get a feel of how the request works, paste the request into your browser and run it. Replace:

  • {tenant} with the name of your tenant.
  • 00001111-aaaa-2222-bbbb-3333cccc4444 with the app ID of an application you registered in your tenant.
  • {application-id-uri}/{scope-name} with the Application ID URI and scope of an application that you registered in your tenant.
  • {policy} with the policy name that you have in your tenant, for example b2c_1_sign_in.
GET /{tenant}.onmicrosoft.com/{policy}/oauth2/v2.0/authorize?
Host: {tenant}.b2clogin.com

client_id=00001111-aaaa-2222-bbbb-3333cccc4444
&response_type=code+id_token
&redirect_uri=https%3A%2F%2Fjwt.ms%2F
&response_mode=fragment
&scope=openid%20offline_access%20{application-id-uri}/{scope-name}
&state=arbitrary_data_you_can_receive_in_the_response
&nonce=12345
Parameter Required Description
{tenant} Yes Name of your Azure AD B2C tenant. If you're using a custom domain, replace tenant.b2clogin.com with your domain, such as fabrikam.com.
{policy} Yes The user flow or policy that the app runs. Specify the name of a user flow that you create in your Azure AD B2C tenant. For example: b2c_1_sign_in, b2c_1_sign_up, or b2c_1_edit_profile.
client_id Yes The application ID that the Azure portal assigned to your application.
nonce Recommended A value included in the request (generated by the application) that is included in the resulting ID token as a claim. The application can then verify this value to mitigate token replay attacks. The value is typically a randomized unique string that can be used to identify the origin of the request.
response_type Yes Must include an ID token for OpenID Connect. If your web application also needs tokens for calling a web API, you can use code+id_token.
scope Yes A space-separated list of scopes. The openid scope indicates a permission to sign in the user and get data about the user in the form of ID tokens. The offline_access scope is optional for web applications. It indicates that your application need a refresh token for extended access to resources. The https://{tenant-name}/{app-id-uri}/{scope} indicates a permission to protected resources, such as a web API. For more information, see Request an access token.
prompt No The type of user interaction that you require. The only valid value at this time is login, which forces the user to enter their credentials on that request.
redirect_uri Yes The redirect_uri parameter of your application, where the server sends authentication responses to your application. It must exactly match one of the redirect_uri parameters that you registered in the Azure portal, except that it must be URL-encoded.
response_mode No The method that is used to send the resulting authorization code back to your application. It can be either query, form_post, or fragment. We recommended you use the form_post response mode for best security.
state No A value that you can include in the request that the authorization server returns in the token response. It can be a string of any content that you want. A randomly generated unique value is typically used for preventing cross-site request forgery attacks. The state is also used to encode information about the user's state in the application before the authentication request occurred, such as the page they were on. If you don't want to register multiple redirect URLs in your Azure portal, you can use the state parameter to differentiate responses in your application from Azure AD B2C service due to different requests.
login_hint No Can be used to prefill the sign-in name field of the sign-in page. For more information, see Prepopulate the sign-in name.
domain_hint No Provides a hint to Azure AD B2C about the social identity provider that should be used for sign-in. If a valid value is included, the user goes directly to the identity provider sign-in page. For more information, see Redirect sign-in to a social provider.
Custom parameters No Custom parameters that can be used with custom policies. For example, dynamic custom page content URI, or key-value claim resolvers.

At this point, the user is asked to complete the workflow. The user might have to enter their username and password, sign in with a social identity, or sign up for the directory. There could be any other number of steps depending on how the user flow is defined.

After the user completes the user flow, a response is returned to your application at the indicated redirect_uri parameter, by using the method that you specify in the response_mode parameter. The response is the same for each of the preceding cases, independent of the user flow.

A successful response using response_mode=fragment would look like:

GET https://jwt.ms/#
id_token=eyJ0eXAiOiJKV1QiLCJhbGciOiJSUzI1NiIsIng1dCI6Ik5HVEZ2ZEstZnl0aEV1Q...
&code=AwABAAAAvPM1KaPlrEqdFSBzjqfTGBCmLdgfSTLEMPGYuNHSUYBrq...
&state=arbitrary_data_you_can_receive_in_the_response
Parameter Description
id_token The ID token that the application requested. You can use the ID token to verify the user's identity and begin a session with the user.
code The authorization code that the application requested, if you used response_type=code+id_token. The application can use the authorization code to request an access token for a target resource. Authorization codes typically expire after about 10 minutes.
state If a state parameter is included in the request, the same value should appear in the response. The application should verify that the state values in the request and response are identical.

Error responses can also be sent to the redirect_uri parameter so that the application can handle them appropriately:

GET https://jwt.ms/#
error=access_denied
&error_description=AADB2C90091%3a+The+user+has+cancelled+entering+self-asserted+information.%0d%0aCorrelation+ID%3a+xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx%0d%0aTimestamp%3a+xxxx-xx-xx+xx%3a23%3a27Z%0d%0a
&state=arbitrary_data_you_can_receive_in_the_response
Parameter Description
error A code that can be used to classify the types of errors that occur.
error_description A specific error message that can help identify the root cause of an authentication error.
state If a state parameter is included in the request, the same value should appear in the response. The application should verify that the state values in the request and response are identical.

Validate the ID token

Just receiving an ID token isn't enough to authenticate the user. Validate the ID token's signature and verify the claims in the token per your application's requirements. Azure AD B2C uses JSON Web Tokens (JWTs) and public key cryptography to sign tokens and verify that they're valid.

Note

Most of the open-source authentication libraries validate the JWT tokens for your application. We recommend exploring those options, rather than implementing your own validation logic. For more information, see Overview of the Microsoft Authentication Library (MSAL), and Microsoft Identity Web authentication library.

Azure AD B2C has an OpenID Connect metadata endpoint, which allows an application to get information about Azure AD B2C at runtime. This information includes endpoints, token contents, and token signing keys. There's a JSON metadata document for each user flow in your B2C tenant. For example, the metadata document for the b2c_1_sign_in user flow in fabrikamb2c.onmicrosoft.com is located at:

https://fabrikamb2c.b2clogin.com/fabrikamb2c.onmicrosoft.com/b2c_1_sign_in/v2.0/.well-known/openid-configuration

One of the properties of this configuration document is jwks_uri, whose value for the same user flow would be:

https://fabrikamb2c.b2clogin.com/fabrikamb2c.onmicrosoft.com/b2c_1_sign_in/discovery/v2.0/keys

To determine which user flow was used to sign an ID token, you have two options. First, the user flow name is included in the acr claim in the ID token, see claim representing user flow. Your other option is to encode the user flow in the value of the state parameter when you issue the request, and then decode it to determine which user flow was used. Either method is valid.

After you acquire the metadata document from the OpenID Connect metadata endpoint, you can use the RSA 256 public keys to validate the signature of the ID token. There might be multiple keys listed at this endpoint, each identified by a kid claim. The header of the ID token also contains a kid claim, which indicates which of these keys was used to sign the ID token.

To verify the tokens from Azure AD B2C, you need to generate the public key using the exponent(e) and modulus(n). To do so, you need to learn how to generate the public key in a programming language of your choice. The official documentation on Public Key generation with the RSA protocol can be found here: https://tools.ietf.org/html/rfc3447#section-3.1

After you validate the signature of the ID token, there are various claims that you need to verify. For instance:

  • Validate the nonce claim to prevent token replay attacks. Its value should be what you specified in the sign-in request.
  • Validate the aud claim to ensure that the ID token was issued for your application. Its value should be the application ID of your application.
  • Validate the iat and exp claims to make sure that the ID token hasn't expired.

There are also several more validations that you should perform. The validations are described in detail in the OpenID Connect Core Spec. You might also want to validate more claims, depending on your scenario. Some common validations include:

  • Ensure that the user/organization signed up for the application.
  • Ensure that the user has proper authorization/privileges.
  • Ensure that a certain strength of authentication has occurred, such as Microsoft Entra multifactor authentication.

After the ID token is validated, you can begin a session with the user. You can use the claims in the ID token to obtain information about the user in your application. Uses for this information include display, records, and authorization.

Get a token

If you need your web application to only run user flows, you can skip the next few sections. These sections are applicable only to web applications that need to make authenticated calls to a web API, which is protected by Azure AD B2C itself.

You can redeem the authorization code that you acquired (by using response_type=code+id_token) for a token to the desired resource by sending a POST request to the /token endpoint. In Azure AD B2C, you can request access tokens for other APIs as usual by specifying their scope(s) in the request.

You can also request an access token for your app's own back-end Web API. In this case, you use the app's client ID as the requested scope, which results in an access token with that client ID as the "audience":

POST https://{tenant}.b2clogin.com/{tenant}.onmicrosoft.com/{policy}/oauth2/v2.0/token HTTP/1.1
Host: {tenant}.b2clogin.com
Content-Type: application/x-www-form-urlencoded

grant_type=authorization_code
&client_id=00001111-aaaa-2222-bbbb-3333cccc4444
&scope=00001111-aaaa-2222-bbbb-3333cccc4444 offline_access
&code=AwABAAAAvPM1KaPlrEqdFSBzjqfTGBCmLdgfSTLEMPGYuNHSUYBrq...
&redirect_uri=urn:ietf:wg:oauth:2.0:oob
Parameter Required Description
{tenant} Yes Name of your Azure AD B2C tenant
{policy} Yes The user flow that was used to acquire the authorization code. You can't use a different user flow in this request. Add this parameter to the query string, not to the POST body.
client_id Yes The application ID that the Azure portal assigned to your application.
client_secret Yes, in Web Apps The application secret that was generated in the Azure portal. Client secrets are used in this flow for Web App scenarios, where the client can securely store a client secret. For Native App (public client) scenarios, client secrets can't be securely stored, therefore not used on this flow. If using a client secret, change it on a periodic basis.
code Yes The authorization code that you acquired in the beginning of the user flow.
grant_type Yes The type of grant, which must be authorization_code for the authorization code flow.
redirect_uri No The redirect_uri parameter of the application where you received the authorization code.
scope No A space-separated list of scopes. The openid scope indicates a permission to sign in the user and get data about the user in the form of id_token parameters. It can be used to get tokens to your application's own back-end web API, which is represented by the same application ID as the client. The offline_access scope indicates that your application needs a refresh token for extended access to resources.

A successful token response looks like:

{
    "not_before": "1442340812",
    "token_type": "Bearer",
    "access_token": "eyJ0eXAiOiJKV1QiLCJhbGciOiJSUzI1NiIsIng1dCI6Ik5HVEZ2ZEstZnl0aEV1Q...",
    "scope": "00001111-aaaa-2222-bbbb-3333cccc4444 offline_access",
    "expires_in": "3600",
    "expires_on": "1644254945",
    "refresh_token": "AAQfQmvuDy8WtUv-sd0TBwWVQs1rC-Lfxa_NDkLqpg50Cxp5Dxj0VPF1mx2Z...",
}
Parameter Description
not_before The epoch time at which the token becomes valid.
token_type The token type value. Bearer is the only type that is supported.
access_token The signed JWT token that you requested.
scope The valid scopes for the token.
expires_in The length of time that the access token is valid (in seconds).
expires_on The epoch time when the access token becomes invalid.
refresh_token An OAuth 2.0 refresh token. The application can use this token to acquire more tokens after the current token expires. Refresh tokens can be used to retain access to resources for extended periods of time. The scope offline_access must have been used in both the authorization and token requests in order to receive a refresh token.

Error responses look like:

{
    "error": "invalid_grant",
    "error_description": "AADB2C90080: The provided grant has expired. Please re-authenticate and try again. Current time: xxxxxxxxxx, Grant issued time: xxxxxxxxxx, Grant expiration time: xxxxxxxxxx\r\nCorrelation ID: xxxxxxxx-xxxx-xxxX-xxxx-xxxxxxxxxxxx\r\nTimestamp: xxxx-xx-16 xx:10:52Z\r\n"
}
Parameter Description
error A code that can be used to classify types of errors that occur.
error_description A message that can help identify the root cause of an authentication error.

Use the token

After you successfully acquire an access token, you can use the token in requests to your back-end web APIs by including it in the Authorization header:

GET /tasks
Host: mytaskwebapi.com
Authorization: Bearer eyJ0eXAiOiJKV1QiLCJhbGciOiJSUzI1NiIsIng1dCI6Ik5HVEZ2ZEstZnl0aEV1Q...

Refresh the token

Access tokens and ID tokens are short-lived. After they expire, you must refresh them to continue to access resources. When you refresh the access token, Azure AD B2C returns a new token. The refreshed access token will have updated nbf (not before), iat (issued at), and exp (expiration) claim values. All other claim values are similar to those in the previous access token.

Refresh a token by submitting another POST request to the /token endpoint. This time, provide the refresh_token parameter instead of the code parameter:

POST https://{tenant}.b2clogin.com/{tenant}.onmicrosoft.com/{policy}/oauth2/v2.0/token HTTP/1.1
Host: {tenant}.b2clogin.com
Content-Type: application/x-www-form-urlencoded

grant_type=refresh_token
&client_id=00001111-aaaa-2222-bbbb-3333cccc4444
&scope=openid offline_access
&refresh_token=AwABAAAAvPM1KaPlrEqdFSBzjqfTGBCmLdgfSTLEMPGYuNHSUYBrq...
&redirect_uri=urn:ietf:wg:oauth:2.0:oob
Parameter Required Description
{tenant} Yes Name of your Azure AD B2C tenant
{policy} Yes The user flow that was used to acquire the original refresh token. You can't use a different user flow in this request. Add this parameter to the query string, not to the POST body.
client_id Yes The application ID that the Azure portal assigned to your application.
client_secret Yes, in Web Apps The application secret that was generated in the Azure portal. Client secrets are used in this flow for Web App scenarios, where the client can securely store a client secret. For Native App (public client) scenarios, client secrets can't be securely stored, therefore not used on this call. If using a client secret, please change it on a periodic basis.
grant_type Yes The type of grant, which must be refresh_token for this part of the authorization code flow.
refresh_token Yes The original refresh token that was acquired in the second part of the flow. The offline_access scope must be used in both the authorization and token requests in order to receive a refresh token.
redirect_uri No The redirect_uri parameter of the application where you received the authorization code.
scope No A space-separated list of scopes. The openid scope indicates a permission to sign in the user and get data about the user in the form of ID tokens. It can be used to send tokens to your application's own back-end web API, which is represented by the same application ID as the client. The offline_access scope indicates that your application needs a refresh token for extended access to resources.

A successful token response looks like:

{
    "not_before": "1442340812",
    "token_type": "Bearer",
    "access_token": "eyJ0eXAiOiJKV1QiLCJhbGciOiJSUzI1NiIsIng1dCI6Ik5HVEZ2ZEstZnl0aEV1Q...",
    "scope": "00001111-aaaa-2222-bbbb-3333cccc4444 offline_access",
    "expires_in": "3600",
    "refresh_token": "AAQfQmvuDy8WtUv-sd0TBwWVQs1rC-Lfxa_NDkLqpg50Cxp5Dxj0VPF1mx2Z...",
    "refresh_token_expires_in": "1209600"
}
Parameter Description
not_before The epoch time at which the token becomes valid.
token_type The token type value. Bearer is the only type that is supported.
access_token The signed JWT token that was requested.
scope The valid scopes for the token.
expires_in The length of time that the access token is valid (in seconds).
refresh_token An OAuth 2.0 refresh token. The application can use this token to acquire additional tokens after the current token expires. Refresh tokens can be used to retain access to resources for extended periods of time.
refresh_token_expires_in The length of time that the refresh token is valid (in seconds).

Error responses look like:

{
    "error": "invalid_grant",
    "error_description": "AADB2C90129: The provided grant has been revoked. Please reauthenticate and try again.\r\nCorrelation ID: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx\r\nTimestamp: xxxx-xx-xx xx:xx:xxZ\r\n",
}
Parameter Description
error A code that can be used to classify types of errors that occur.
error_description A message that can help identify the root cause of an authentication error.

Send a sign-out request

When you want to sign the user out of the application, it isn't enough to clear the application's cookies or otherwise end the session with the user. Redirect the user to Azure AD B2C to sign out. If you fail to do so, the user might be able to reauthenticate to your application without entering their credentials again. For more information, see Azure AD B2C session behavior.

To sign out the user, redirect the user to the end_session_endpoint that is listed in the OpenID Connect metadata document described earlier:

GET https://{tenant}.b2clogin.com/{tenant}.onmicrosoft.com/{policy}/oauth2/v2.0/logout?post_logout_redirect_uri=https%3A%2F%2Fjwt.ms%2F
Parameter Required Description
{tenant} Yes Name of your Azure AD B2C tenant. If you're using a custom domain, replace tenant.b2clogin.com with your domain, such as fabrikam.com.
{policy} Yes The user flow that you specify in the authorization request. For example, if the user signed-in with the b2c_1_sign_in user flow, specify b2c_1_sign_in in the sign-out request.
id_token_hint No A previously issued ID token to pass to the logout endpoint as a hint about the end user's current authenticated session with the client. The id_token_hint ensures that the post_logout_redirect_uri is a registered reply URL in your Azure AD B2C application settings. For more information, see Secure your logout redirect.
client_id No* The application ID that the Azure portal assigned to your application.

*This is required when using Application isolation SSO configuration and Require ID Token in logout request is set to No.
post_logout_redirect_uri No The URL that the user should be redirected to after successful sign out. If it isn't included, Azure AD B2C shows the user a generic message. Unless you provide an id_token_hint, you shouldn't register this URL as a reply URL in your Azure AD B2C application settings.
state No If you include a state parameter in the authorization request, the authorization server returns the same value in the response to the post_logout_redirect_uri. The application should verify that the state value in the request and response are identical.

Upon a sign-out request, Azure AD B2C invalidates the Azure AD B2C cookie-based session, and attempts to sign out from federated identity providers. For more information, see Single sign-out.

Secure your logout redirect

After logout, the user is redirected to the URI you specify in the post_logout_redirect_uri parameter, regardless of the reply URLs that you specify for the application. However, if a valid id_token_hint is passed, and the Require ID Token in logout requests is turned on, Azure AD B2C verifies that the value of post_logout_redirect_uri matches one of the application's configured redirect URIs before performing the redirect. If no matching reply URL was configured for the application, an error message is displayed and the user isn't redirected.

To set the required ID Token in logout requests, see Configure session behavior in Azure Active Directory B2C.

Next steps