F5 BIG-IP and NetApp StorageGRID - Providing Fast and Scalable S3 API for AI apps
F5 BIG-IP, an industry-leading ADC solution, can provide load balancing services for HTTPS servers, with full security applied in-flight and performance levels to meet any enterprise’s capacity targets. Specific to S3 API, the object storage and retrieval protocol that rides upon HTTPS, an aligned partnering solution exists from NetApp, which allows a large-scale set of S3 API targets to ingest and provide objects. Automatic backend synchronization allows any node to be offered up at a target by a server load balancer like BIG-IP. This allows overall storage node utilization to be optimized across the node set, and scaled performance to reach the highest S3 API bandwidth levels, all while offering high availability to S3 API consumers. S3 compatible storage is becoming popular for AI applications due to its superior performance over traditional protocols such as NFS or CIFS, as well as enabling repatriation of data from the cloud to on-prem. These are scenarios where the amount of data faced is large, this drives the requirement for new levels of scalability and performance; S3 compatible object storages such as NetApp StorageGRID are purpose-built to reach such levels. Sample BIG-IP and StorageGRID Configuration This document is based upon tests and measurements using the following lab configuration. All devices in the lab were virtual machine-based offerings. The S3 service to be projected to the outside world, depicted in the above diagram and delivered to the client via the external network, will use a BIG-IP virtual server (VS) which is tied to an origin pool of three large-capacity StorageGRID nodes. The BIG-IP maintains the integrity of the NetApp nodes by frequent HTTP-based health checks. Should an unhealthy node be detected, it will be dropped from the list of active pool members. When content is written via the S3 protocol to any node in the pool, the other members are synchronized to serve up content should they be selected by BIG-IP for future read requests. The key recommendations and observations in building the lab include: Setup a local certificate authority such that all nodes can be trusted by the BIG-IP. Typically the local CA-signed certificate will incorporate every node’s FQDN and IP address within the listed subject alternate names (SAN) to make the backend solution streamlined with one single certificate. Different F5 profiles, such as FastL4 or FastHTTP, can be selected to reach the right tradeoff between the absolute capacity of stateful traffic load-balanced versus rich layer 7 functions like iRules or authentication. Modern techniques such as multi-part uploads or using HTTP Ranges for downloads can take large objects, and concurrently move smaller pieces across the load balancer, lowering total transaction times, and spreading work over more CPU cores. The S3 protocol, at its core, is a set of REST API calls. To facilitate testing, the widely used S3Browser (www.s3browser.com) was used to quickly and intuitively create S3 buckets on the NetApp offering and send/retrieve objects (files) through the BIG-IP load balancer. Setup the BIG-IP and StorageGrid Systems The StorageGrid solution is an array of storage nodes, provisioned with the help of an administrative host, the “Grid Manager”. For interactive users, no thick client is required as on-board web services allow a streamlined experience all through an Internet browser. The following is an example of Grid Manager, taken from a Chrome browser; one sees the three Storage Nodes setup have been successfully added. The load balancer, in our case, the BIG-IP, is set up with a virtual server to support HTTPS traffic and distributed that traffic, which is S3 object storage traffic, to the three StorageGRID nodes. The following screenshot demonstrates that the BIG-IP is setup in a standard HA (active-passive pair) configuration and the three pool members are healthy (green, health checks are fine) and receiving/sending S3 traffic, as the byte counts are seen in the image to be non-zero. On the internal side of the BIG-IP, TCP port 18082 is being used for S3 traffic. To do testing of the solution, including features such as multi-part uploads and downloads, a popular S3 tool, S3Browser, was downloaded and used. The following shows the entirety of the S3Browser setup. Simply create an account (StorageGRID-Account-01 in our example) and point the REST API endpoint at the BIG-IP Virtual Server that is acting as the secure front door for our pool of NetApp nodes. The S3 Access Key ID and Secret values are generated at turn-up time of the NetApp appliances. All S3 traffic will, of course, be SSL/TLS encrypted. BIG-IP will intercept the SSL traffic (high-speed decrypt) and then re-encrypt when proxying the traffic to a selected origin pool member. Other valid load balancer setups exist; one might include an “off load” approach to SSL, whereby the S3 nodes safely co-located in a data center may prefer to receive non-SSL HTTP S3 traffic. This may see an overall performance improvement in terms of peak bandwidth per storage node, but this comes at the tradeoff of security considerations. Experimenting with S3 Protocol and Load Balancing With all the elements in place to start understanding the behavior of S3 and spreading traffic across NetApp nodes, a quick test involved creating a S3 bucket and placing some objects in that new bucket. Buckets are logical collections of objects, conceptually not that different from folders or directories in file systems. In fact, a S3 bucket could even be mounted as a folder in an operating system such as Linux. In their simplest form, most commonly, buckets can simply serve as high-capacity, performant storage and retrieval targets for similarly themed structured or unstructured data. In the first test, we created a new bucket (“audio-clip-bucket”) and uploaded four sample files to the new bucket using S3Browser. We then zeroed the statistics for each pool member on the BIG-IP, to see if even this small upload would spread S3 traffic across more than a single NetApp device. Immediately after the upload, the counters reflect that two StorageGRID nodes were selected to receive S3 transactions. Richly detailed, per-transaction visibility can be obtained by leveraging the F5 SSL Orchestrator (SSLO) feature on the BIG-IP, whereby copies of the bi-directional S3 traffic decrypted within the load balancer can be sent to packet loggers, analytics tools, or even protocol analyzers like Wireshark. The BIG-IP also has an onboard analytics tool, Application Visibility and Reporting (AVR) which can provide some details on the nuances of the S3 traffic being proxied. AVR demonstrates the following characteristics of the above traffic, a simple bucket creation and upload of 4 objects. With AVR, one can see the URL values used by S3, which include the bucket name itself as well as transactions incorporating the object names as URLs. Also, the HTTP methods used included both GETS and PUTS. The use of HTTP PUT is expected when creating a new bucket. S3 is not governed by a typical standards body document, such as an IETF Request for Comment (RFC), but rather has evolved out of AWS and their use of S3 since 2006. For details around S3 API characteristics and nomenclature, this site can be referenced. For example, the expected syntax for creating a bucket is provided, including the fact that it should be an HTTP PUT to the root (/) URL target, with the bucket configuration parameters including name provided within the HTTP transaction body. Achieving High Performance S3 with BIG-IP and StorageGRID A common concern with protocols, such as HTTP, is head-of-line blocking, where one large, lengthy transaction blocks subsequent desired, queued transactions. This is one of the reasons for parallelism in HTTP, where loading 30 or more objects to paint a web page will often utilize two, four, or even more concurrent TCP sessions. Another performance issue when dealing with very large transactions is, without parallelism, even those most performant networks will see an established TCP session reach a maximum congestion window (CWND) where no more segments may be in flight until new TCP ACKs arrive back. Advanced TCP options like TCP exponential windowing or TCP SACK can help, but regardless of this, the achievable bandwidth of any one TCP session is bounded and may also frequently task only one core in multi-core CPUs. With the BIG-IP serving as the intermediary, large S3 transactions may default to “multi-part” uploads and downloads. The larger objects become a series of smaller objects that conveniently can be load-balanced by BIG-IP across the entire cluster of NetApp nodes. As displayed in the following diagram, we are asking for multi-part uploads to kick in for objects larger than 5 megabytes. After uploading a 20-megabyte file (technically, 20,000,000 bytes) the BIG-IP shows the traffic distributed across multiple NetApp nodes to the tune of 160.9 million bits. The incoming bits, incoming from the perspective of the origin pool members, confirm the delivery of the object with a small amount of protocol overhead (bits divided by eight to reach bytes). The value of load balancing manageable chunks of very large objects will pay dividends over time with faster overall transaction completion times due to the spreading of traffic across NetApp nodes, more TCP sessions reaching high congestion window values, and no single-core bottle necks in multicore equipment. Tuning BIG-IP for High Performance S3 Service Delivery The F5 BIG-IP offers a set of different profiles it can run its Local Traffic Manager (LTM) module in accordance with; LTM is the heart of the server load balancing function. The most performant profile in terms of attainable traffic load is the “FastL4” profile. This, and other profiles such as “OneConnect” or “FastHTTP”, can be tied to a virtual server, and details around each profile can be found here within the BIG-IP GUI: The FastL4 profile can increase virtual server performance and throughput for supported platforms by using the embedded Packet Velocity Acceleration (ePVA) chip to accelerate traffic. The ePVA chip is a hardware acceleration field programmable gate array (FPGA) that delivers high-performance L4 throughput by offloading traffic processing to the hardware acceleration chip. The BIG-IP makes flow acceleration decisions in software and then offloads eligible flows to the ePVA chip for that acceleration. For platforms that do not contain the ePVA chip, the system performs acceleration actions in software. Software-only solutions can increase performance in direct relationship to the hardware offered by the underlying host. As examples of BIG-IP virtual edition (VE) software running on mid-grade hardware platforms, results with Dell can be found here and similar experiences with HPE Proliant platforms are here. One thing to note about FastL4 as the profile to underpin a performance mode BIG-IP virtual server is that it is layer 4 oriented. For certain features that involve layer 7 HTTP related fields, such as using iRules to swap HTTP headers or perform HTTP authentication, a different profile might be more suitable. A bonus of FastL4 are some interesting specific performance features catering to it. In the BIG-IP version 17 release train, there is a feature to quickly tear down, with no delay, TCP sessions no longer required. Most TCP stacks implement TCP “2MSL” rules, where upon receiving and sending TCP FIN messages, the socket enters a lengthy TCP “TIME_WAIT” state, often minutes long. This stems back to historically bad packet loss environments of the very early Internet. A concern was high latency and packet loss might see incoming packets arrive at a target very late, and the TCP state machine would be confused if no record of the socket still existed. As such, the lengthy TIME_WAIT period was adopted even though this is consuming on-board resources to maintain the state. With FastL4, the “fast” close with TCP reset option now exists, such that any incoming TCP FIN message observed by BIG-IP will result in TCP RESETS being sent to both endpoints, normally bypassing TIME_WAIT penalties. OneConnect and FastHTTP Profiles As mentioned, other traffic profiles on BIG-IP are directed towards Layer 7 and HTTP features. One interesting profile is F5’s “OneConnect”. The OneConnect feature set works with HTTP Keep-Alives, which allows the BIG-IP system to minimize the number of server-side TCP connections by making existing connections available for reuse by other clients. This reduces, among other things, excessive TCP 3-way handshakes (Syn, Syn-Ack, Ack) and mitigates the small TCP congestion windows that new TCP sessions start with and only increases with successful traffic delivery. Persistent server-side TCP connections ameliorate this. When a new connection is initiated to the virtual server, if an existing server-side flow to the pool member is idle, the BIG-IP system applies the OneConnect source mask to the IP address in the request to determine whether it is eligible to reuse the existing idle connection. If it is eligible, the BIG-IP system marks the connection as non-idle and sends a client request over it. If the request is not eligible for reuse, or an idle server-side flow is not found, the BIG-IP system creates a new server-side TCP connection and sends client requests over it. The last profile considered is the “Fast HTTP” profile. The Fast HTTP profile is designed to speed up certain types of HTTP connections and again strives to reduce the number of connections opened to the back-end HTTP servers. This is accomplished by combining features from the TCP, HTTP, and OneConnect profiles into a single profile that is optimized for network performance. A resulting high performance HTTP virtual server processes connections on a packet-by-packet basis and buffers only enough data to parse packet headers. The performance HTTP virtual server TCP behavior operates as follows: the BIG-IP system establishes server-side flows by opening TCP connections to pool members. When a client makes a connection to the performance HTTP virtual server, if an existing server-side flow to the pool member is idle, the BIG-IP LTM system marks the connection as non-idle and sends a client request over the connection. Summary The NetApp StorageGRID multi-node S3 compatible object storage solution fits well with a high-performance server load balancer, thus making the F5 BIG-IP a good fit. S3 protocol can itself be adjusted to improve transaction response times, such as through the use of multi-part uploads and downloads, amplifying the default load balancing to now spread even more traffic chunks over many NetApp nodes. BIG-IP has numerous approaches to configuring virtual servers, from highest performance L4-focused profiles to similar offerings that retain L7 HTTP awareness. Lab testing was accomplished using the S3Browser utility and results of traffic flows were confirmed with both the standard BIG-IP GUI and the additional AVR analytics module, which provides additional protocol insight.
APM Configuration to Support Duo MFA using iRule
Overview BIG-IP APM has supported Duo as an MFA provider for a long time with RADIUS-based integration. Recently, Duo has added support for Universal Prompt that uses Open ID Connect (OIDC) protocol to provide two-factor authentication. To integrate APM as an OIDC client and resource server, and Duo as an Identity Provider (IdP), Duo requires the user’s logon name and custom parameters to be sent for Authentication and Token request. This guide describes the configuration required on APM to enable Duo MFA integration using an iRule. iRules addresses the custom parameter challenges by generating the needed custom values and saving them in session variables, which the OAuth Client agent then uses to perform MFA with Duo. This integration procedure is supported on BIG-IP versions 13.1, 14.1x, 15.1x, and 16.x. To integrate Duo MFA with APM, complete the following tasks: 1. Choose deployment type: Per-request or Per-session 2. Configure credentials and policies for MFA on the DUO web portal 3. Create OAuth objects on the BIG-IP system 4. Configure the iRule 5. Create the appropriate access policy/policies on the BIG-IP system 6. Apply policy/policies and iRule to the APM virtual server Choose deployment type APM supports two different types of policies for performing authentication functions. Per-session policies: Per-session policies provide authentication and authorization functions that occur only at the beginning of a user’s session. These policies are compatible with most APM use cases such as VPN, Webtop portal, Remote Desktop, federation IdP, etc. Per-request policies: Per-request policies provide dynamic authentication and authorization functionality that may occur at any time during a user’s session, such as step-up authentication or auditing functions only for certain resources. These policies are only compatible with Identity Aware Proxy and Web Access Management use cases and cannot be used with VPN or webtop portals. This guide contains information about setting up both policy types. Prerequisites Ensure the BIG-IP system has DNS and internet connectivity to contact Duo directly for validating the user's OAuth tokens. Configure credentials and policies for MFA on Duo web portal Before you can protect your F5 BIG-IP APM Web application with Duo, you will first need to sign up for a Duo account. 1. Log in to the Duo Admin Panel and navigate to Applications. 2. Click Protect an application. Figure 1: Duo Admin Panel – Protect an Application 3. Locate the entry for F5 BIG-IP APM Web in the applications list and click Protect to get the Client ID, Client secret, and API hostname. You will need this information to configure objects on APM. Figure 2: Duo Admin Panel – F5 BIG-IP APM Web 4. As DUO is used as a secondary authentication factor, the user’s logon name is sent along with the authentication request. Depending on your security policy, you may want to pre-provision users in Duo, or you may allow them to self-provision to set their preferred authentication type when they first log on. To add users to the Duo system, navigate to the Dashboard page and click the Add New... -> Add User button. A Duo username should match the user's primary authentication username. Refer to the https://duo.com/docs/enrolling-users link for the different methods of user enrollment. Refer to Duo Universal Prompt for additional information on Duo’s two-factor authentication. Create OAuth objects on the BIG-IP system Create a JSON web key When APM is configured to act as an OAuth client or resource server, it uses JSON web keys (JWKs) to validate the JSON web tokens it receives from Duo. To create a JSON web key: 1. On the Main tab, select Access > Federation > JSON Web Token > Key Configuration. The Key Configuration screen opens. 2. To add a new key configuration, click Create. 3. In the ID and Shared Secret fields, enter the Client ID and Client Secret values respectively obtained from Duo when protecting the application. 4. In the Type list, select the cryptographic algorithm used to sign the JSON web key. Figure 3: Key Configuration screen 5. Click Save. Create a JSON web token As an OAuth client or resource server, APM validates the JSON web tokens (JWT) it receives from Duo. To create a JSON web token: 1. On the Main tab, select Access > Federation > JSON Web Token > Token Configuration. The Token Configuration screen opens. 2. To add a new token configuration, click Create. 3. In the Issuer field, enter the API hostname value obtained from Duo when protecting the application. 4. In the Signing Algorithms area, select from the Available list and populate the Allowed and Blocked lists. 5. In the Keys (JWK) area, select the previously configured JSON web key in the allowed list of keys. Figure 4: Token Configuration screen 6. Click Save. Configure Duo as an OAuth provider APM uses the OAuth provider settings to get URIs on the external OAuth authorization server for JWT web tokens. To configure an OAuth provider: 1. On the Main tab, select Access > Federation > OAuth Client / Resource Server > Provider. The Provider screen opens. 2. To add a provider, click Create. 3. In the Name field, type a name for the provider. 4. From the Type list, select Custom. 5. For Token Configuration (JWT), select a configuration from the list. 6. In the Authentication URI field, type the URI on the provider where APM should redirect the user for authentication. The hostname is the same as the API hostname in the Duo application. 7. In the Token URI field, type the URI on the provider where APM can get a token. The hostname is the same as the API hostname in the Duo application. Figure 5: OAuth Provider screen 8. Click Finished. Configure Duo server for APM The OAuth Server settings specify the OAuth provider and role that Access Policy Manager (APM) plays with that provider. It also sets the Client ID, Client Secret, and Client’s SSL certificates that APM uses to communicate with the provider. To configure a Duo server: 1. On the Main tab, select Access > Federation > OAuth Client / Resource Server > OAuth Server. The OAuth Server screen opens. 2. To add a server, click Create. 3. In the Name field, type a name for the Duo server. 4. From the Mode list, select how you want the APM to be configured. 5. From the Type list, select Custom. 6. From the OAuth Provider list, select the Duo provider. 7. From the DNS Resolver list, select a DNS resolver (or click the plus (+) icon, create a DNS resolver, and then select it). 8. In the Token Validation Interval field, type a number. In a per-request policy subroutine configured to validate the token, the subroutine repeats at this interval or the expiry time of the access token, whichever is shorter. 9. In the Client Settings area, paste the Client ID and Client secret you obtained from Duo when protecting the application. 10. From the Client's ServerSSL Profile Name, select a server SSL profile. Figure 6: OAuth Server screen 11. Click Finished. Configure an auth-redirect-request and a token-request Requests specify the HTTP method, parameters, and headers to use for the specific type of request. An auth-redirect-request tells Duo where to redirect the end-user, and a token-request accesses the authorization server for obtaining an access token. To configure an auth-redirect-request: 1. On the Main tab, select Access > Federation > OAuth Client / Resource Server > Request. The Request screen opens. 2. To add a request, click Create. 3. In the Name field, type a name for the request. 4. For the HTTP Method, select GET. 5. For the Type, select auth-redirect-request. 6. As shown in Figure 7, specify the list of GET parameters to be sent: request parameter with value depending on the type of policy For per-request policy: %{subsession.custom.jwt_duo} For per-session policy: %{session.custom.jwt_duo} client_id parameter with type client-id response_type parameter with type response-type Figure 7: Request screen with auth-redirect-request (Use “subsession.custom…” for Per-request or “session.custom…” for Per-session) 7. Click Finished. To configure a token-request: 1. On the Main tab, select Access > Federation > OAuth Client / Resource Server > Request. The Request screen opens. 2. To add a request, click Create. 3. In the Name field, type a name for the request. 4. For the HTTP Method, select POST. 5. For the Type, select token-request. 6. As shown in Figure 8, specify the list of POST parameters to be sent: client_assertion parameter with value depending on the type of policy For per-request policy: %{subsession.custom.jwt_duo_token} For per-session policy: %{session.custom.jwt_duo_token} client_assertion_type parameter with value urn:ietf:params:oauth:client-assertion-type:jwt-bearer grant_type parameter with type grant-type redirect_uri parameter with type redirect-uri Figure 8: Request screen with token-request (Use “subsession.custom…” for Per-request or “session.custom…” for Per-session) 7. Click Finished. Configure the iRule iRules gives you the ability to customize and manage your network traffic. Configure an iRule that creates the required sub-session variables and usernames for Duo integration. Note: This iRule has sections for both per-request and per-session policies and can be used for either type of deployment. To configure an iRule: 1. On the Main tab, click Local Traffic > iRules. 2. To create an iRules, click Create. 3. In the Name field, type a name for the iRule. 4. Copy the sample code given below and paste it in the Definition field. Replace the following variables with values specific to the Duo application: <Duo Client ID> in the getClientId function with Duo Application ID. <Duo API Hostname> in the createJwtToken function with API Hostname. For example, https://api-duohostname.com/oauth/v1/token. <JSON Web Key> in the getJwkName function with the configured JSON web key. Note: The iRule ID here is set as JWT_CREATE. You can rename the ID as desired. You specify this ID in the iRule Event agent in Visual Policy Editor. Note: The variables used in the below example are global, which may affect your performance. Refer to the K95240202: Understanding iRule variable scope article for further information on global variables, and determine if you use a local variable for your implementation. when ACCESS_POLICY_AGENT_EVENT { if { [ACCESS::policy agent_id] eq "JWT_CREATE" } { set duo_uname [ACCESS::session data get "session.logon.last.username"] # Inline logic for creating JWT set header "{\"alg\":\"HS512\",\"typ\":\"JWT\"}" set exp [expr {[clock seconds] + 900}] set client_id "<Duo Client ID>" set redirect_uri "https://[ACCESS::session data get session.server.network.name]/oauth/client/redirect" set payload "{\"response_type\": \"code\",\"scope\":\"openid\",\"exp\":${exp},\"client_id\":\"${client_id}\",\"redirect_uri\":\"${redirect_uri}\",\"duo_uname\":\"${duo_uname}\"}" set jwt_duo [ACCESS::oauth sign -header $header -payload $payload -alg HS512 -key "<JSON Web Key>"] ACCESS::session data set session.custom.jwt_duo $jwt_duo # JWT Token creation set aud "<Duo API Hostname>" set jti [string range [clock seconds] 0 31] set token_payload "{\"sub\": \"${client_id}\",\"iss\":\"${client_id}\",\"aud\":\"${aud}\",\"exp\":${exp},\"jti\":\"${jti}\"}" set jwt_duo_token [ACCESS::oauth sign -header $header -payload $token_payload -alg HS512 -key "<JSON Web Key>"] ACCESS::session data set session.custom.jwt_duo_token $jwt_duo_token } } when ACCESS_PER_REQUEST_AGENT_EVENT { if { [ACCESS::perflow get perflow.irule_agent_id] eq "JWT_CREATE" } { set duo_uname [ACCESS::session data get "session.logon.last.username"] set header "{\"alg\":\"HS512\",\"typ\":\"JWT\"}" set exp [expr {[clock seconds] + 900}] set client_id "<Duo Client ID>" set redirect_uri "https://[ACCESS::session data get session.server.network.name]/oauth/client/redirect" set payload "{\"response_type\": \"code\",\"scope\":\"openid\",\"exp\":${exp},\"client_id\":\"${client_id}\",\"redirect_uri\":\"${redirect_uri}\",\"duo_uname\":\"${duo_uname}\"}" set jwt_duo [ACCESS::oauth sign -header $header -payload $payload -alg HS512 -key "<JSON Web Key>"] ACCESS::perflow set perflow.custom $jwt_duo # JWT Token creation set aud "<Duo API Hostname>" set jti [string range [clock seconds] 0 31] set token_payload "{\"sub\": \"${client_id}\",\"iss\":\"${client_id}\",\"aud\":\"${aud}\",\"exp\":${exp},\"jti\":\"${jti}\"}" set jwt_duo_token [ACCESS::oauth sign -header $header -payload $token_payload -alg HS512 -key "<JSON Web Key>"] ACCESS::perflow set perflow.scratchpad $jwt_duo_token } } Note: iRule updated 11/27/2024 to eliminate CMP demotion. Figure 9: iRule screen 5. Click Finished. Create the appropriate access policy/policies on the BIG-IP system Per-request policy Skip this section for a per-session type deployment The per-request policy is used to perform secondary authentication with Duo. Configure the access policies through the access menu, using the Visual Policy Editor. The per-request access policy must have a subroutine with an iRule Event, Variable Assign, and an OAuth Client agent that requests authorization and tokens from an OAuth server. You may use other per-request policy items such as URL branching or Client Type to call Duo only for certain target URIs. Figure 10 shows a subroutine named duosubroutine in the per-request policy that handles Duo MFA authentication. Figure 10: Per-request policy in Visual Policy Editor Configuring the iRule Event agent The iRule Event agent specifies the iRule ID to be executed for Duo integration. In the ID field, type the iRule ID as configured in the iRule. Figure 11: iRule Event agent in Visual Policy Editor Configuring the Variable Assign agent The Variable Assign agent specifies the variables for token and redirect requests and assigns a value for Duo MFA in a subroutine. This is required only for per-request type deployment. Add sub-session variables as custom variables and assign their custom Tcl expressions as shown in Figure 12. subsession.custom.jwt_duo_token = return [mcget {perflow.scratchpad}] subsession.custom.jwt_duo = return [mcget {perflow.custom}] Figure 12: Variable Assign agent in Visual Policy Editor Configuring the OAuth Client agent An OAuth Client agent requests authorization and tokens from the Duo server. Specify OAuth parameters as shown in Figure 13. In the Server list, select the Duo server to which the OAuth client directs requests. In the Authentication Redirect Request list, select the auth-redirect-request configured earlier. In the Token Request list, select the token-request configured earlier. Some deployments may not need the additional information provided by OpenID Connect. You could, in that case, disable it. Figure 13: OAuth Client agent in Visual Policy Editor Per-session policy Configure the Per Session policy as appropriate for your chosen deployment type. Per-request: The per-session policy must contain at least one logon page to set the username variable in the user’s session. Preferably it should also perform some type of primary authentication. This validated username is used later in the per-request policy. Per-session: The per-session policy is used for all authentication. A per-request policy is not used. Figures 14a and 14b show a per-session policy that runs when a client initiates a session. Depending on the actions you include in the access policy, it can authenticate the user and perform actions that populate session variables with data for use throughout the session. Figure 14a: Per-session policy in Visual Policy Editor performs both primary authentication and Duo authentication (for per-session use case) Figure 14b: Per-session policy in Visual Policy Editor performs primary authentication only (for per-request use case) Apply policy/policies and iRule to the APM virtual server Finally, apply the per-request policy, per-session policy, and iRule to the APM virtual server. You assign iRules as a resource to the virtual server that users connect. Configure the virtual server’s default pool to the protected local web resource. Apply policy/policies to the virtual server Per-request policy To attach policies to the virtual server: 1. On the Main tab, click Local Traffic > Virtual Servers. 2. Select the Virtual Server. 3. In the Access Policy section, select the policy you created. 4. Click Finished. Figure 15: Access Policy section in Virtual Server (per-request policy) Per-session policy Figure 16 shows the Access Policy section in Virtual Server when the per-session policy is deployed. Figure 16: Access Policy section in Virtual Server (per-session policy) Apply iRule to the virtual server To attach the iRule to the virtual server: 1. On the Main tab, click Local Traffic > Virtual Servers. 2. Select the Virtual Server. 3. Select the Resources tab. 4. Click Manage in the iRules section. 5. Select an iRule from the Available list and add it to the Enabled list. 6. Click Finished.
Use Fully Qualified Domain Name (FQDN) for GSLB Pool Member with F5 DNS
Normally, we define a specific IP (and port) to be used as GSLB pool member. This article provides a custom configuration to be able to use Fully Qualified Domain Name (FQDN) as GSLB pool member--with all GSLB features like health-check monitoring, load balancing method, persistence, etc. Despite GSLB as a mechanism to distribute traffic across datacenters having reached years of age, it has not become less relevant this recent years. The fact that internet infrastructure still rely heavily on DNS technology means GSLB is continuously used due to is lightweight nature and smooth integration. When using F5 DNS as GSLB solution, usually we are dealing with LTM and its VS as GSLB server and pool member respectively. Sometimes, we will add a non-LTM node as a generic server to provide inter-DC load balancing capability. Either way, we will end up with a pair of IP and port to represent the application, in which we sent a health-check against. Due to the trend of public cloud and CDN, there is a need to use FQDN as GSLB pool member (instead of IP and port pair). Some of us may immediately think of using a CNAME-type GSLB pool to accommodate this. However, there is a limitation in which BIG-IP requires a CNAME-type GSLB pool to use a wideIP-type pool member, in which we will end up with an IP and port pair (again!) We can use "static target", but there is "side-effect" where the pool member will always consider available (which then triggers the question why we need to use GSLB in the first place!). Additionally, F5 BIG-IP TMUI accepts FQDN input when we configure GSLB server and pool member. However, it will immediately translate to IP based on configured DNS. Thus, this is not the solution we are looking for Now this is where F5’s BIG-IP power (a.k.a programmability) comes into play. Enter the realm of customization... We all love customization, but at the same time do not want that to be overly complicated so that life becomes harder on day-2 🙃. Thus, the key is to use some customization, but simple enough to avoid unnecessary complication. Here is one idea to solve our FQDN as GSLB pool problem above The customized configuration object includes 1. External health-check monitor: Dynamically resolve DNS to translate FQDN into IP address Perform health-check monitoring against current IP address Result is used to determine GSLB pool member availability status 2. DNS iRules: Check #1: Checks if GSLB pool attached to wideIP contains only FQDN-type member (e.g. other pool referring to LTM VS is also attached to the wideIP) If false, do nothing (let DNS response refer to LTM VS) Otherwise, perform check #2 Check #2: Checks current health-check status of requested domain name If FQDN is up, modify DNS response to return current IP of FQDN Otherwise, perform fallback action as requirement (e.g. return empty response, return static IP, use fallback pool, etc.) 3. Internal Datagroup: Store current IP of FQDN, updated according to health-check interval Datagroup record value contains current IP if health-check success. Otherwise, the value contains empty data Here are some of the codes, where configured; wideIP is gslb.test.com, while GSLB pool member FQDN is arcadia.f5poc.id 1. External health-check monitor config gtm monitor external gslb_external_monitor { defaults-from external destination *:* interval 10 probe-timeout 5 run /Common/gslb_external_monitor_script timeout 120 #define FQDN here user-defined fqdn arcadia.f5poc.id } External health-check monitor script #!/bin/sh pidfile="/var/run/$MONITOR_NAME.$1..$2.pid" if [ -f $pidfile ] then kill -9 -`cat $pidfile` > /dev/null 2>&1 fi echo "$$" > $pidfile # Obtain current IP for the FQDN resolv=`dig +short ${fqdn}` # The actual monitoring action here curl -fIs -k https://${fqdn}/ --resolve ${fqdn}:443:${resolv} | grep -i HTTP 2>&1 > /dev/null status=$? if [ $status -eq 0 ] then # Actions when health-check success rm -f $pidfile tmsh modify ltm data-group internal fqdn { records replace-all-with { $fqdn { data $resolv } } } echo "sending monitor to ${fqdn} ${resolv} with result OK" | logger -p local0.info echo "up" else # Actions when health-check fails tmsh modify ltm data-group internal fqdn { records replace-all-with { $fqdn { } } } echo "sending monitor to ${fqdn} ${resolv} with result NOK" | logger -p local0.info fi rm -f $pidfile 2. DNS iRules when DNS_REQUEST { set qname [DNS::question name] # Obtain current IP for the FQDN set currentip [class match -value $qname equals fqdn] } when DNS_RESPONSE { set rname [getfield [lindex [split [DNS::answer]] 4] "\}" 1 ] #Check if return is IP address of specially encoded FQDN IP, 10.10.10.10 in this example if {$rname eq "10.10.10.10" }{ #Response is only from pool with external monitor, meaning no other pool is attached to wideIP if {$currentip ne ""}{ #Current FQDN health-check success DNS::answer clear # Use current IP to construct DNS answer section DNS::answer insert "[DNS::question name]. 123 [DNS::question class] [DNS::question type] $currentip" } else { #Current FQDN health-check failed #Define action to be performed here DNS::answer clear } } } 3. Internal Datagroup ltm data-group internal fqdn { records { # Define FQDN as record name arcadia.f5poc.id { # Record data contains IP, where this will be continuously updated by external monitoring script data 158.140.176.219 } } type string } *GSLB virtual server configuration Some testing The resolve will follow whichever current IP address for the FQDN. If a returning CNAME response is required, you can do so by modifying DNS irules above. The logic and code are open to any improvement, so leave your suggestions in the comments if you have any. Thanks!iRule Editor - System Config Editing
In the latest release of the iRule Editor v 0.10.1, I added several new features. This tutorial will walk through System Level Configuration editing allowing you to work with your bigip.conf and bigip_base.conf files without having to open a terminal session to the BIG-IP. Usage:Happy 20th Birthday, BIG-IP TMOS!
I wasn’t in the waiting room with the F5 family, ears and eyes perked for the release announcement of BIG-IP version 9.0. I was a customer back in 2004, working on a government contract at Scott AFB, Illinois. I shared ownership of the F5 infrastructure, pairs of BIG-IPs running version 4.5 on Dell PowerEdge 2250 servers with one other guy. But maybe a month or two before the official first release of TMOS, my F5 account manager dropped off some shiny new hardware. And it was legit purpose-built and snazzy, not some garage-style hacked Frankenstein of COTS parts like the earlier stuff. And you wonder why we chose Dell servers! Anyway, I was a hard-core network engineer at this time, with very little exposure to anything above layer four, and even there, my understanding was limited to ports and ACLs and maybe a little high-level clarity around transport protocols. But application protocols? Nah. No idea. So with this new hardware and an entirely new full-proxy architecture (what’s a proxy, again?) I was overwhelmed. And honestly, I was frustrated with it for the first few days because I didn’t know what I didn’t know and so I struggled to figure out what to do with it, even to replicate my half-proxy configuration in the “new way”. But I’m a curious person. Given enough time and caffeine, I can usually get to the bottom of a problem, at least well enough to arrive at a workable solution. And so I did. My typical approach to anything is to make it work, make it work better, make it work reliably better, then finally make it work reliably and more performantly better. And the beauty here with this new TMOS system is that I was armed with a treasure trove of new toys. The short list I dug into during my beta trial, which lasted for a couple of weeks: The concept of a profile. When you support a few applications, this is no big deal. When you support hundreds, being able to macro configuration snippets within your application and across applications was revolutionary. Not just for the final solution, but also for setting up and executing your test plans. iRules. Yes, technically they existed in 4.x, but they were very limited in scope. With TMOS, F5 introduced the Tcl-based and F5 extended live-traffic scripting environment that unleashed tremendous power and flexibility for network and application teams. I dabbled with this, and thought I understood exactly how useful this was. More on this a little later. A host operating system. I was a router, switch, and firewall guy. Nothing I worked on had this capability. I mean, a linux system built in to my networking device? YES!!! Two things I never knew I always needed during my trial: 1) tcpdump ON BOX. Seriously--mind blown; and 2) perl scripting against config and snmp. Yeah, I know, I laugh about perl now. But 20 years ago, it was the cats pajamas. A fortunate job change Shortly after my trial was over, I interviewed for an accepted a job offer from a major rental car company that was looking to hire an engineer to redesign their application load balancing infrastructure and select the next gear purchase for the effort. We evaluated Cisco, Nortel/Alteon, Radware, and F5 on my recommendation. With our team’s resident architect we drafted the rubric with which we’d evaluate all the products, and whereas there were some layer two performance issues in some packet sizes that were arguably less than real-world, the BIG-IP blew away the competitors across the board. Particularly, though, in configurability and instrumentation. Tcpdump on box was such a game-changer for us. Did we have issues with TMOS version 9? For sure. My first year with TMOS was also TMOS's first year. Bugs are going to happen with any release, but a brand new thing is guaranteed. But F5 support was awesome, and we worked through all the issues in due time. Anyway, I want to share three wins in my first year with TMOS. Win #1 Our first production rollout was in the internet space, on BIG-IP version 9.0.5. That’s right, a .0 release. TMOS was a brand new baby, and we had great confidence throughout our testing. During our maintenance, once we flipped over the BIG-IPs, our rental transaction monitors all turned red and the scripted rental process had increased by 50%! Not good. “What is this F5 stuff? Send it back!!” But it was new, and we knew we had a gem here. We took packet captures on box, of course, then rolled back and took more packet captures, this time through taps because our old stuff didn’t have tcpdump on box. This is where Jason started to really learn about the implications of both a full proxy architecture and the TCP protocol. It turns our our application servers had a highly-tuned TCP stack on them specific to the characteristics of the rental application. We didn’t know this, of course. But since we implemented a proxy that terminates clients at the BIG-IP and starts a new session to the servers, all those customizations for WAN traffic were lost. Once we built a TCP profile specifically for the rental application servers and tested it under WAN emulation, we not only reached parity with the prior performance but beat it by 10%. Huzzah! Go BIG-IP custom protocol stack configuration! Win #2 For the next internal project, I had to rearchitect the terminal server farm. We had over 700 servers in two datacenters supporting over 60,000 thin clients around the world for rental terminals. Any failures meant paper tickets and unhappy staff and customers. One thing that was problematic with the existing solution is that sometimes clients would detach and upon reconnect would connect directly to the server, which skewed the load balancers view of the world and frequently overloaded some servers to the point all sessions on that server would hang until metrics (but usually angry staff) would notify. Remember my iRules comment earlier on differentiators? Well, iRules architect David Hansen happened to be a community hero and was very helpful to me in the DevCentral forums and really opened my eyes to the art of possible with iRules. He was able to take the RDP session token that was being returned by the client, read it, translate it from its Microsoft encoding format, and then forward the session on to the correct server in the backend so that all sessions continued to be accounted for in our load balancing tier. This was formative for me as a technologist and as a member of the DevCentral community. Win #3 2004-2005 was the era before security patching was as visible a responsibility as it is today, but even then we had a process and concerns when there were obstacles. We had an internal application that had a plugin for the web tier that managed all the sessions to the app tier, and this plugin was no longer supported. We were almost a year behind on system and application patches because we had no replacement for this. Enter, again, iRules. I was able to rebuild the logic of the plugin in an iRule that IIRC wasn’t more than 30 lines. So the benefits ended up not only being a solution to that problem, but the ability to remove that web tier altogether, saving on equipment, power, and complexity costs. And that was just the beginning... TMOS was mature upon arrival, but it got better every year. iControl added REST-based API access; clustered multi-processing introduced tremendous performance gains; TMOS got virtualized, and all the home-lab technologists shouted with joy; a plugin architecture allowed for product modules like ASM and APM; solutions that began as iRules like AFM and SSLO became products. It’s crazy how much innovation has taken place on this platform! The introduction of TMOS didn’t just introduce me to applications and programmability. It did that and I’m grateful, but it did so much more. It unlocked in me that fanboy level that fans of sports teams, video game platforms, Taylor Swift, etc, experience. It helped me build an online community at DevCentral, long before I was an employee. Happy 20th Birthday, TMOS! We celebrate and salute you!Load Balancing TCP TLS Encrypted Syslog Messages
Syslog messages sent via TCP are not always evenly distributed among backend syslog servers because multiple syslog messages can be sent in a single TCP connection. This article utilizes the F5 BIG-IP Generic Message Routing Framework (MRF) to evenly distribute syslog messages among backend syslog pool members. This solution also uses TLS to protect the confidentiality of the syslog messages. This article is based off the work done by Mark Lloyd in this DevCentral Technical Article: A Simple One-way Generic MRF Implementation to load balance syslog message. In his article, Mark explains how to setup Generic Message Routing Framework (MRF) to distribute syslog messages sent via TCP to a pool of syslog servers. This article adds the necessary configuration to TLS encrypt, decrypt, and re-encrypt the messages. This was tested on BIG-IP version 17.1.0.1. The first step is to define the message routing protocol. The difference between the default protocol (genericmsg) is the field no-response must be configured to yes if this is a one-way stream. Otherwise, the server side will allocate buffers for return traffic that will cause severe free memory depletion. Note: The message-terminator is set to "%0a", this represents a newline character in hex and is used to separate the syslog messages. This value can be changed if a different delineator is required. ltm message-routing generic protocol simple_syslog_protocol { app-service none defaults-from genericmsg description none disable-parser no max-egress-buffer 32768 max-message-size 32768 message-terminator %0a no-response yes } An iRule must be configured on both the Virtual Server and Generic Transport Config. This iRule must be linked as a profile in both the virtual server and the generic transport configuration. ltm rule mrf_simple { when CLIENT_ACCEPTED { GENERICMESSAGE::peer name "[IP::local_addr]:[TCP::local_port]_[IP::remote_addr]:[TCP::remote_port]" } when SERVER_CONNECTED { GENERICMESSAGE::peer name "[IP::local_addr]:[TCP::local_port]_[IP::remote_addr]:[TCP::remote_port]" } } The next item to configure is the generic transport config. The generic transport config has the generic protocol configured along with the iRule to setup the server-side peers. A server-side SSL profile is also configured here to TLS encrypt the traffic to the backend syslog servers. ltm message-routing generic transport-config simple_syslog_tcp_tc { ip-protocol tcp profiles { serverssl-insecure-compatible { } simple_syslog_protocol { } tcp { } } rules { mrf_simple } } Nodes are defined for the backend Syslog servers. ltm node 10.1.20.201 { address 10.1.20.201 } ltm node 10.1.20.202 { address 10.1.20.202 } A pool is created containing the nodes. ltm pool syslog_pool { members { 10.1.20.201:514 { address 10.1.20.201 } 10.1.20.202:514 { address 10.1.20.202 } } } The next step is to create the generic message routing peer. This peer is used to identify the pool of syslog servers that the syslog messages will be routed to. ltm message-routing generic peer simple_syslog_peer { pool syslog_pool transport-config simple_syslog_tcp_tc } Now that the peer is defined, a generic route can be created to send traffic to the peer. ltm message-routing generic route simple_syslog_route { peers { simple_syslog_peer } } A generic router is configured with the generic route. ltm message-routing generic router simple_syslog_router { app-service none defaults-from messagerouter description none ignore-client-port no max-pending-bytes 23768 max-pending-messages 64 mirror disabled mirrored-message-sweeper-interval 1000 routes { simple_syslog_route } traffic-group traffic-group-1 use-local-connection yes } A client-ssl profile is configured to associate the certificates to the Virtual Server. ltm profile client-ssl syslog-ng_client { app-service none cert-key-chain { syslog-ng-new_f5demos_ca_0 { cert syslog-ng-new chain f5demos_ca key syslog-ng-new } } defaults-from clientssl inherit-ca-certkeychain true inherit-certkeychain false } A virtual server is created to receive incoming TLS-encrypted TCP syslog messages. ltm virtual mrftest_simple { creation-time 2024-10-08:09:37:52 destination 10.1.10.101:514 ip-protocol tcp last-modified-time 2024-10-08:13:55:49 mask 255.255.255.255 profiles { simple_syslog_protocol { } simple_syslog_router { } syslog-ng_client { context clientside } tcp { } } rules { mrf_simple } serverssl-use-sni disabled source 0.0.0.0/0 source-address-translation { type automap } translate-address enabled translate-port enabled vs-index 2 } Conclusion This example is from a use case where a single syslog client was sending to a TCP load balancer. The load balancer was not evenly distributing the load among the backend servers because multiple messages were being sent as part of a single TCP connection. This solution utilizes a generic Message Routing Framework to evenly distribute TCP syslog messages. TLS encryption is also used on the client to load balancer connection as well as the load balancer to backend server connection to protect the confidentiality of the syslog messages.
iRules 101 - #12 - The Session Command
One of the things that makes iRules so incredibly powerful is the fact that it is a true scripting language, or at least based on one. The fact that they give you the tools that TCL brings to the table - regular expressions, string functions, even things as simple as storing, manipulating and recalling variable data - sets iRules apart from the rest of the crowd. It also makes it possible to do some pretty impressive things with connection data and massaging/directing it the way you want it. Other articles in the series: Getting Started with iRules: Intro to Programming with Tcl | DevCentral Getting Started with iRules: Control Structures & Operators | DevCentral Getting Started with iRules: Variables | DevCentral Getting Started with iRules: Directing Traffic | DevCentral Getting Started with iRules: Events & Priorities | DevCentral Intermediate iRules: catch | DevCentral Intermediate iRules: Data-Groups | DevCentral Getting Started with iRules: Logging & Comments | DevCentral Advanced iRules: Regular Expressions | DevCentral Getting Started with iRules: Events & Priorities | DevCentral iRules 101 - #12 - The Session Command | DevCentral Intermediate iRules: Nested Conditionals | DevCentral Intermediate iRules: Handling Strings | DevCentral Intermediate iRules: Handling Lists | DevCentral Advanced iRules: Scan | DevCentral Advanced iRules: Binary Scan | DevCentral Sometimes, though, a simple variable won't do. You've likely heard of global variables in one of the earlier 101 series and read the warning there, and are looking for another option. So here you are, you have some data you need to store, which needs to persist across multiple connections. You need it to be efficient and fast, and you don't want to have to do a whole lot of complex management of a data structure. One of the many ways that you can store and access information in your iRule fits all of these things perfectly, little known as it may be. For this scenario I'd recommend the usage of the session command. There are three main permutations of the session command that you'll be using when storing and referencing data within the session table. These are: session add: Stores user's data under the specified key for the specified persistence mode session lookup: Returns user data previously stored using session add session delete: Removes user data previously stored using session add A simple example of adding some information to the session table would look like: when CLIENTSSL_CLIENTCERT { set ssl_cert [SSL::cert 0] session add ssl $ssl_cert 90 } By using the session add command, you can manually place a specific piece of data into the LTM's session table. You can then look it up later, by unique key, with the session lookup command and use the data in a different section of your iRule, or in another connection all together. This can be helpful in different situations where data needs to be passed between iRules or events that it might not normally be when using a simple variable. Such as mining SSL data from the connection events, as below: when CLIENTSSL_CLIENTCERT { # Set results in the session so they are available to other events session add ssl [SSL::sessionid] [list [X509::issuer] [X509::subject] [X509::version]] 180 } when HTTP_REQUEST { # Retrieve certificate information from the session set sslList [session lookup ssl [SSL::sessionid]] set issuer [lindex sslList 0] set subject [lindex sslList 1] set version [lindex sslList 2] } Because the session table is optimized and designed to handle every connection that comes into the LTM, it's very efficient and can handle quite a large number of items. Also note that, as above, you can pass structured information such as TCL Lists into the session table and they will remain intact. Keep in mind, though, that there is currently no way to count the number of entries in the table with a certain key, so you'll have to build all of your own processing logic for now, where necessary. It's also important to note that there is more than one session table. If you look at the above example, you'll see that before we listed any key or data to be stored, we used the command session add ssl. Note the "ssl" portion of this command. This is a reference to which session table the data will be stored in. For our purposes here there are effectively two session tables: ssl, and uie. Be sure you're accessing the same one in your session lookup section as you are in your session add section, or you'll never find the data you're after. This is pretty easy to keep straight, once you see it. It looks like: session add uie ... session lookup uie Or: session add ssl ... session lookup ssl You can find complete documentation on the session command here, in the iRules, as well as some great examples that depict some more advanced iRules making use of the session command to great success. Check out Codeshare for more examples.Declarative Advanced WAF policy lifecycle in a CI/CD pipeline
The purpose of this article is to show the configuration used to deploy a declarative Advanced WAF policy to a BIG-IP and automatically configure it to protect an API workload by consuming an OpenAPI file describing the application. For this experiment, a Gitlab CI/CD pipeline was used to deploy an API workload to Kubernetes, configure a declarative Adv. WAF policy to a BIG-IP device and tuning it by incorporating learning suggestions exported from the BIG-IP. Lastly, the F5 WAF tester tool was used to determine and improve the defensive posture of the Adv. WAF policy. Deploying the declarative Advanced WAF policy through a CI/CD pipeline To deploy the Adv. WAF policy, the Gitlab CI/CD pipeline is calling an Ansible playbook that will in turn deploy an AS3 application referencing the Adv.WAF policy from a separate JSON file. This allows the application definition and WAF policy to be managed by 2 different groups, for example NetOps and SecOps, supporting separation of duties. The following Ansible playbook was used; --- - hosts: bigip connection: local gather_facts: false vars: my_admin: "xxxx" my_password: "xxxx" bigip: "xxxx" tasks: - name: Deploy AS3 API AWAF policy uri: url: "https://{{ bigip }}/mgmt/shared/appsvcs/declare" method: POST headers: "Content-Type": "application/json" "Authorization": "Basic xxxxxxxxxx body: "{{ lookup('file','as3_waf_openapi.json') }}" body_format: json validate_certs: no status_code: 200 The Advanced WAF policy 'as3_waf_openapi.json' was specified as follows: { "class": "AS3", "action": "deploy", "persist": true, "declaration": { "class": "ADC", "schemaVersion": "3.2.0", "id": "Prod_API_AS3", "API-Prod": { "class": "Tenant", "defaultRouteDomain": 0, "arcadia": { "class": "Application", "template": "generic", "VS_API": { "class": "Service_HTTPS", "remark": "Accepts HTTPS/TLS connections on port 443", "virtualAddresses": ["xxxxx"], "redirect80": false, "pool": "pool_NGINX_API", "policyWAF": { "use": "Arcadia_WAF_API_policy" }, "securityLogProfiles": [{ "bigip": "/Common/Log all requests" }], "profileTCP": { "egress": "wan", "ingress": { "use": "TCP_Profile" } }, "profileHTTP": { "use": "custom_http_profile" }, "serverTLS": { "bigip": "/Common/arcadia_client_ssl" } }, "Arcadia_WAF_API_policy": { "class": "WAF_Policy", "url": "http://xxxx/root/awaf_openapi/-/raw/master/WAF/ansible/bigip/policy-api.json", "ignoreChanges": true }, "pool_NGINX_API": { "class": "Pool", "monitors": ["http"], "members": [{ "servicePort": 8080, "serverAddresses": ["xxxx"] }] }, "custom_http_profile": { "class": "HTTP_Profile", "xForwardedFor": true }, "TCP_Profile": { "class": "TCP_Profile", "idleTimeout": 60 } } } } } The AS3 declaration will provision a separate Administrative Partition ('API-Prod') containing a Virtual Server ('VS_API'), an Adv. WAF policy ('Arcadia_WAF_API_policy') and a pool ('pool_NGINX_API'). The Adv.WAF policy being referenced ('policy-api.json') is stored in the same Gitlab repository but can be downloaded from a separate location. { "policy": { "name": "policy-api-arcadia", "description": "Arcadia API", "template": { "name": "POLICY_TEMPLATE_API_SECURITY" }, "enforcementMode": "transparent", "server-technologies": [ { "serverTechnologyName": "MySQL" }, { "serverTechnologyName": "Unix/Linux" }, { "serverTechnologyName": "MongoDB" } ], "signature-settings": { "signatureStaging": false }, "policy-builder": { "learnOnlyFromNonBotTraffic": false }, "open-api-files": [ { "link": "http://xxxx/root/awaf_openapi/-/raw/master/App/openapi3-arcadia.yaml" } ] }, "modifications": [ ] } The declarative Adv.WAF policy is referencing in turn the OpenAPI file ('openapi3-arcadia.yaml') that describes the application being protected. Executing the Ansible playbook results in the AS3 application being deployed, along with the Adv.WAF policy that is automatically configured according to the OpenAPI file. Handling learning suggestions in a CI/CD pipeline The next step in the CI/CD pipeline used for this experiment was to send legitimate traffic using the API and collect the learning suggestions generated by the Adv.WAF policy, which will allow a simple way to customize the WAF policy further for the specific application being protected. The following Ansible playbook was used to retrieve the learning suggestions: --- - hosts: bigip connection: local gather_facts: true vars: my_admin: "xxxx" my_password: "xxxx" bigip: "xxxxx" tasks: - name: Get all Policy_key/IDs for WAF policies uri: url: 'https://{{ bigip }}/mgmt/tm/asm/policies?$select=name,id' method: GET headers: "Authorization": "Basic xxxxxxxxxxx" validate_certs: no status_code: 200 return_content: yes register: waf_policies - name: Extract Policy_key/ID of Arcadia_WAF_API_policy set_fact: Arcadia_WAF_API_policy_ID="{{ item.id }}" loop: "{{ (waf_policies.content|from_json)['items'] }}" when: item.name == "Arcadia_WAF_API_policy" - name: Export learning suggestions uri: url: "https://{{ bigip }}/mgmt/tm/asm/tasks/export-suggestions" method: POST headers: "Content-Type": "application/json" "Authorization": "Basic xxxxxxxxxxx" body: "{ \"inline\": \"true\", \"policyReference\": { \"link\": \"https://{{ bigip }}/mgmt/tm/asm/policies/{{ Arcadia_WAF_API_policy_ID }}/\" } }" body_format: json validate_certs: no status_code: - 200 - 201 - 202 - name: Get learning suggestions uri: url: "https://{{ bigip }}/mgmt/tm/asm/tasks/export-suggestions" method: GET headers: "Authorization": "Basic xxxxxxxxx" validate_certs: no status_code: 200 register: result - name: Print learning suggestions debug: var=result A sample learning suggestions output is shown below: "json": { "items": [ { "endTime": "xxxxxxxxxxxxx", "id": "ZQDaRVecGeqHwAW1LDzZTQ", "inline": true, "kind": "tm:asm:tasks:export-suggestions:export-suggestions-taskstate", "lastUpdateMicros": 1599953296000000.0, "result": { "suggestions": [ { "action": "add-or-update", "description": "Enable Evasion Technique", "entity": { "description": "Directory traversals" }, "entityChanges": { "enabled": true }, "entityType": "evasion" }, { "action": "add-or-update", "description": "Enable HTTP Check", "entity": { "description": "Check maximum number of parameters" }, "entityChanges": { "enabled": true }, "entityType": "http-protocol" }, { "action": "add-or-update", "description": "Enable HTTP Check", "entity": { "description": "No Host header in HTTP/1.1 request" }, "entityChanges": { "enabled": true }, "entityType": "http-protocol" }, { "action": "add-or-update", "description": "Enable enforcement of policy violation", "entity": { "name": "VIOL_REQUEST_MAX_LENGTH" }, "entityChanges": { "alarm": true, "block": true }, "entityType": "violation" } Incorporating the learning suggestions in the Adv.WAF policy can be done by simple copy&pasting the self-contained learning suggestions blocks into the "modifications" list of the Adv.WAF policy: { "policy": { "name": "policy-api-arcadia", "description": "Arcadia API", "template": { "name": "POLICY_TEMPLATE_API_SECURITY" }, "enforcementMode": "transparent", "server-technologies": [ { "serverTechnologyName": "MySQL" }, { "serverTechnologyName": "Unix/Linux" }, { "serverTechnologyName": "MongoDB" } ], "signature-settings": { "signatureStaging": false }, "policy-builder": { "learnOnlyFromNonBotTraffic": false }, "open-api-files": [ { "link": "http://xxxxxx/root/awaf_openapi/-/raw/master/App/openapi3-arcadia.yaml" } ] }, "modifications": [ { "action": "add-or-update", "description": "Enable Evasion Technique", "entity": { "description": "Directory traversals" }, "entityChanges": { "enabled": true }, "entityType": "evasion" } ] } Enhancing Advanced WAF policy posture by using the F5 WAF tester The F5 WAF tester is a tool that generates known attacks and checks the response of the WAF policy. For example, running the F5 WAF tester against a policy that has a "transparent" enforcement mode will cause the tests to fail as the attacks will not be blocked. The F5 WAF tester can suggest possible enhancement of the policy, in this case the change of the enforcement mode. An abbreviated sample output of the F5 WAF Tester: ................................................................ "100000023": { "CVE": "", "attack_type": "Server Side Request Forgery", "name": "SSRF attempt (AWS Metadata Server)", "results": { "parameter": { "expected_result": { "type": "signature", "value": "200018040" }, "pass": false, "reason": "ASM Policy is not in blocking mode", "support_id": "" } }, "system": "All systems" }, "100000024": { "CVE": "", "attack_type": "Server Side Request Forgery", "name": "SSRF attempt - Local network IP range 10.x.x.x", "results": { "request": { "expected_result": { "type": "signature", "value": "200020201" }, "pass": false, "reason": "ASM Policy is not in blocking mode", "support_id": "" } }, "system": "All systems" } }, "summary": { "fail": 48, "pass": 0 } Changing the enforcement mode from "transparent" to "blocking" can easily be done by editing the same Adv. WAF policy file: { "policy": { "name": "policy-api-arcadia", "description": "Arcadia API", "template": { "name": "POLICY_TEMPLATE_API_SECURITY" }, "enforcementMode": "blocking", "server-technologies": [ { "serverTechnologyName": "MySQL" }, { "serverTechnologyName": "Unix/Linux" }, { "serverTechnologyName": "MongoDB" } ], "signature-settings": { "signatureStaging": false }, "policy-builder": { "learnOnlyFromNonBotTraffic": false }, "open-api-files": [ { "link": "http://xxxxx/root/awaf_openapi/-/raw/master/App/openapi3-arcadia.yaml" } ] }, "modifications": [ { "action": "add-or-update", "description": "Enable Evasion Technique", "entity": { "description": "Directory traversals" }, "entityChanges": { "enabled": true }, "entityType": "evasion" } ] } A successful run will will be achieved when all the attacks will be blocked. ......................................... "100000023": { "CVE": "", "attack_type": "Server Side Request Forgery", "name": "SSRF attempt (AWS Metadata Server)", "results": { "parameter": { "expected_result": { "type": "signature", "value": "200018040" }, "pass": true, "reason": "", "support_id": "17540898289451273964" } }, "system": "All systems" }, "100000024": { "CVE": "", "attack_type": "Server Side Request Forgery", "name": "SSRF attempt - Local network IP range 10.x.x.x", "results": { "request": { "expected_result": { "type": "signature", "value": "200020201" }, "pass": true, "reason": "", "support_id": "17540898289451274344" } }, "system": "All systems" } }, "summary": { "fail": 0, "pass": 48 } Conclusion By adding the Advanced WAF policy into a CI/CD pipeline, the WAF policy can be integrated in the lifecycle of the application it is protecting, allowing for continuous testing and improvement of the security posture before it is deployed to production. The flexible model of AS3 and declarative Advanced WAF allows the separation of roles and responsibilities between NetOps and SecOps, while providing an easy way for tuning the policy to the specifics of the application being protected. Links UDF lab environment link. Short instructional video link.
