APM OIDC debug logging is somehow verbose but not where it should - RFE
While I was configuring and testing an APM Access policy for an OIDC integration with Entra ID, I was seeing some backchannel errors logged (debug level) some HTTP error 400. The whole logging is quite verbose, but here, not very useful: [...] OAuth Client: failed for server '/Common/entraid_test' using 'authorization_code' grant type (client_id=xxxxxxxxxxxxxx), error: HTTP error 400, Session variable 'session.oauth.client./Common/entra_front_act_oauth_client_ag.errMsg' set to 'HTTP error 400, ' Session variable 'session.oauth.client./Common/entraid_test.errMsg' set to 'HTTP error 400, ' Session variable 'session.oauth.client.last.errMsg' set to 'HTTP error 400, ' So, tcpdump was required to see what exactly the error was in the returned JSON payload from Entra ID. I've got a new RFE from F5 support => (Bug alias 2229965) [RFE] Add extra verbosity for the debug-level logging for OIDC backchannel connection Don't hesitate to open a support case to get it bound to that RFE, the more we are the higher priority will be assigned to implement it (hopefully). đ Alexandre
Using OpenID Connect to authenticate users
Hello all, I want to use OpenID Connect to authenticate my users before gaining access to one of my application. I want to use my bigip as OpenID Provider (ie: the entity that authenticate the users) . My issue is the following: The OpendID provider (my bigip) never provides me with a ID Token. All I have is an âAccess Tokenâ and a âRefresh Tokenâ but no âID Tokenâ. Below is a description of my lab: resource owner: ip address 10.10.255.1 bigip OpenID Provider: virtual server ip 10.10.255.221 (see below for the access policy) *The agents are left with their default values. Client: I user openid debugger (https://oidcdebugger.com/) in order to request the authorization code. Then I request the Tokens using an html code. I do the following for testing : I request the authorization code from the authorization server (ie my virtual server) . For this I use openID Connect debugger to construct the request for me. Here is the request that I send : ? client_id=e1111098ccff5f81859f9fc83eaa000c29267803efc0bb5b & redirect_uri=https://oidcdebugger.com/debug & scope=openid myscope & response_type=code & response_mode=form_post & state=toto & nonce=7r8saarltr After sending this request I enter my credentials (in the logon agent), I click on authorize then the virtual server correctly redirects me to: https://oidcdebugger.com/debug code=cae85f6fd33c6b27f56f536b6fd9af2ca0fc78e69ac3d13799533c143b38b4ac&state=totoâ I now have a correct authorization code that I can exchange for an âaccess tokenâ AND a âID Tokenâ . I then send a POST to get the access and ID Tokens using the following HTML code : *note the presence of the âopenidâ in the scope parameter. However, this is what I get from the authorization server (see in the comment) : -> I have No âID TOKENâ âš Could you please help me configure my access policy so that it supports OpenID Connect and sends me an âID Tokenâ ??Using NGINX Controller API Management Module and NGINX App Protect to secure financial services API transactions
As financial services APIs (such as Open Banking) are concerned primarily with managing access to exposed banking APIs, the security aspect has always been of paramount importance. Securing financial services APIs is a vast topic, as security controls are distributed among different functions, such as user authentication at the Identity Provider level, user authorization and basic API security at the API Gateway level and advanced API security at the WAF level. In this article we will explore how two NGINX products, Controller API Management Module and App Protect, can be deployed to secure the OAuth Authorization Code flow which is a building block of the access controls used to secure many financial services APIs.. Physical setup The setup used to support this article comprises of NGINX Controller API Management Module, providing API Management functions through an instance of NGINX API Gateway and NGINX App Protect deployed on a Kubernetes Ingress Controller providing advanced security for the Kubernetes-deployed demo application, Arcadia Finance. These elements are being deployed and configured in an automated fashion using a Gitlab CI/CD pipeline. The visualization for NGINX App Protect is provided by NAP dashboards deployed in ELK. Note: For the purpose of supporting this lab, APM was configured as an OAuth Authorization Server supporting OpenID Connect. Its configuration, along with the implementation details of the third party banking application (AISP/PISP), acting as an OAuth Client, is beyond the scope of this article. In an OAuth Authorization Code flow, the PSU (End User) is initiating an API request through the Account or Payment Information Services Provider (AISP/PISP Application) which first redirects the end user to the Authorization Server. Strong Customer Authentication is being performed between the end user and Authorization Server which, if successful, will issue an authorization code and redirect the user back to the AISP/PISP Application. The AISP/PISP Application will exchange the authorization code for an ID Token and a JWT Access Token, the latter will be attached as a bearer token to the initial end-user API request which will then be forwarded to the API Gateway. The API Gateway will authenticate the signature of the JWT Access Token by downloading the JSON Web Key (JWK) from the Authorization Server and may apply further security controls by authorising the API call based on JWT claims and/or apply rate limits. Worth noting here is the security function of the API Gateway, which provides positive security by allowing only calls conforming to published APIs, in addition to authentication and authorization functions. The Web Application Firewall function, represented here by the NGINX App Protect deployed on the Kubernetes Ingress Controller (KIC), will add negative security protection, by checking the request against a database of attack signatures, and advanced API security, by validating the API request against the OpenAPI manifest and providing Bot detection capabilities. Configuration To configure the NGINX Controller API Management Module, first create an Application by sending a POST request to 'https://{{ my_controller }}/api/v1/services/environments/env_prod/apps' having the following body: { "metadata": { "name": "app_api", "displayName": "API Application Arcadia", "description": "", "tags": [] }, "desiredState": {} } Then create an Identity Provider, pointed at the Authorization Server's JWK endpoint, by sending a PUT request to 'https://{{ my_controller }}/api/v1/security/identity-providers/bank_idp' having the following body: { "metadata": { "name": "bank_idp", "tags": [] }, "desiredState": { "environmentRefs": [ { "ref": "/services/environments/env_prod" } ], "identityProvider": { "type": "JWT", "jwkFile": { "type": "REMOTE_FILE", "uri": "https://bank.f5lab/f5-oauth2/v1/jwks", "cacheExpire": "12h" } } } } Create an API definition by sending a PUT request to 'https://{{ my_controller }}/api/v1/services/api-definitions/arcadia-api-def/versions/v1' with the following body: { "metadata": { "name": "v1", "displayName": "arcadia-api-def" }, "desiredState": { "specs": { "REST": { "openapi": "3.0.0", "info": { "version": "v1", "title": "arcadia-api-def" }, "paths": {} } } } } Then import the OpenAPI definition by sending a PUT request to 'https://{{ my_controller }}/api/v1/services/api-definitions/arcadia-api-def/versions/v1/import' with the OpenAPI JSON as a request body. Publish the API definition by sending a PUT request to 'https://{{ my_controller }}/api/v1/services/environments/env_prod/apps/app_api/published-apis/prod-api', with the following body: { "metadata": { "name": "prod-api", "displayName": "prod-api", "tags": [] }, "desiredState": { "apiDefinitionVersionRef": { "ref": "/services/api-definitions/arcadia-api-def/versions/v1" }, "gatewayRefs": [ { "ref": "/services/environments/env_prod/gateways/gw_api" } ] } } Declare the necessary back-end components (in this example webapi-kic.nginx-udf.internal Kubernetes workload) by sending a PUT to 'https://{{ my_controller }}/api/v1/services/environments/env_prod/apps/app_api/components/cp_moneytransfer_api' with the following body: { "metadata": { "name": "cp_moneytransfer_api", "displayName": "cp_moneytransfer_api", "tags": [] }, "desiredState": { "ingress": { "uris": { "/api/rest/execute_money_transfer.php": { "php": { "get": { "description": "Send money to a friend", "parameters": [ { "in": "body", "name": "body", "required": true, "schema": { "type": "object" } } ], "responses": { "200": { "description": "200 response" } } }, "matchMethod": "EXACT" } } }, "gatewayRefs": [ { "ref": "/services/environments/env_prod/gateways/gw_api" } ] }, "backend": { "ntlmAuthentication": "DISABLED", "preserveHostHeader": "DISABLED", "workloadGroups": { "wl_mainapp_api": { "loadBalancingMethod": { "type": "ROUND_ROBIN" }, "uris": { "http://webapi-kic.nginx-udf.internal:30276": { "isBackup": false, "isDown": false, "isDrain": false } } } } }, "programmability": { "requestHeaderModifications": [ { "action": "DELETE", "applicableURIs": [], "headerName": "Host" }, { "action": "ADD", "applicableURIs": [], "headerName": "Host", "headerValue": "k8s.arcadia-finance.io" } ] }, "logging": { "errorLog": "DISABLED", "accessLog": { "state": "DISABLED" } }, "security": { "rateLimits": { "policy_1": { "rate": "5000r/m", "burstBeforeReject": 0, "statusCode": 429, "key": "$binary_remote_addr" } }, "conditionalAuthPolicies": { "policy_1": { "action": "ALLOW", "comparisonType": "CONTAINS", "comparisonValues": [ "Payment" ], "sourceType": "JWT_CLAIM", "sourceKey": "scope", "denyStatusCode": 403 } }, "identityProviderRefs": [ { "ref": "/security/identity-providers/bank_idp" } ], "jwtClientAuth": { "keyLocation": "BEARER" } }, "publishedApiRefs": [ { "ref": "/services/environments/env_prod/apps/app_api/published-apis/prod-api" } ] } } Note the 'security' block, specifying the JWT authentication, the Identity Provider from where to download the JWK, the authorization check applied on each request and the rate limit policy. The configuration used to deploy NGINX App Protect on the Kubernetes Ingress Controller can be consulted here. Summary In this article we showed how NGINX Controller API Management Module and NGINX App Protect can be deployed to protect API calls as part of the OAuth Authorization Code flow which is a basic flow used to control the access to many financial services APIs. Links UDF lab environment link.
Authenticate user of native mobile app with OpenId Connect
Does F5 Big IP Access Manager support mobile apps authenticating over OpenId Connect with custom URI redirect_uri? Our native mobile app (iOS and Android) authenticates the user using the Authorization Code Grant flow. How it Works. Our redirect_uri (ie callback uri) is: com.mckesson.wfm.ansos2go://signin We are a software vendor in the Healthcare domain. Our customer who uses F5 Big IP says that this URI is considered invalid by F5 when configuring the OpenId Connect Service Provider. Is that true? If so, how do native mobile app developers perform OIDC authentication with F5? Thanks, Scott UPDATE: I got word from my customer that they set up a rewrite policy, so they could enter the redirect_uri as https:/com.mckesson.wfm.ansos2go://signin. Then, they strip off the https:// in the response to the initial 'authorize' call. This is NUTS! Why does F5 Big IP Access Manager require redirect_uri to be https://...? This totally breaks the OpenId Connect specification which says "The Redirection URI MAY use an alternate scheme, such as one that is intended to identify a callback into a native application." https://openid.net/specs/openid-connect-core-1_0.html#AuthorizationEndpoint
