KASM Workspaces Integration with F5 BIG-IP Zero Trust Access (formerly APM)
Introduction F5 BIG-IP Zero Trust Access, a key component of the F5 Application Delivery and Security Platform (ADSP), helps teams secure apps that are spread across hybrid, multi-cloud and AI environments. In this article, I’ll show you how to secure your Kasm Workspace using F5 BIG-IP Zero Trust Access. Kasm Workspaces Kasm Workspaces is a containerized streaming platform designed for secure, web-based access to desktops, applications, and web browsing. It leverages container technology to deliver virtualized environments directly to users' browsers, enhancing security, scalability, and performance. Commonly used for remote work, cybersecurity, and DevOps workflows, Kasm Workspaces provides a flexible and customizable solution for organizations needing secure and efficient access to virtual resources. As noted in the Kasm Documentation, the Kasm Workspaces Web App Role servers should not be exposed directly to the public. That’s where F5 BIG-IP Zero Trust Access can help. Demo Video Deployment Prerequisites F5 BIG-IP version 17.x Zero Trust Access version 10.x Kasm Workspaces version 1.17 installed and configured properly Configure using Automation Toolchain with AS3 and FAST Templates The F5 BIG-IP Automation Toolchain is a suite of tools designed to automate the deployment, configuration, and management of F5 BIG-IP devices. It enables efficient and consistent management using declarative APIs, templates, and integrations with popular automation frameworks. Application services (FAST) templates are predefined configurations that streamline the deployment and management of applications by providing consistent and repeatable setups. NOTE: The configuration using the Automation Toolchain is well-documented in this DevCentral article, which also includes demo videos: How I did it - “Delivering Kasm Workspaces three ways” Configure Manually Using a Virtual Server This article will focus on the manual configuration of the BIG-IP using a Virtual Server. Configuring it this way will give you a deeper understanding of how all the components work together to create a cohesive solution. Network Environment Linux “External” client IP: 10.1.10.4 BIG-IP “External” Self IP: 10.1.10.10 BIG-IP “Internal” Self IP: 10.1.20.10 Kasm Workspace IP: 10.1.20.23 BIG-IP Configuration Create HTTP Monitor: First, let’s create the HTTP Monitor for the Kasm Workspace server. From Local Traffic > Monitors > click the green plus sign to add a new one. Give it a name, “Kasm-Monitor” in this example Set the Type to HTTP Enter the following for the Send String: GET /api/__healthcheck\r\n Enter the following for the Receive String: OK It should look like this: Set Reverse to Yes and click Finished Create Pool: Next we’ll create the Pool From Local Traffic > Pools > Pool List > click the plus sign to add a new one Give it a name, “Kasm-Pool” in this example Select the Health Monitor you created previously and click the arrows to move it to Active Under Resources specify a Node Name, “Kasm-Server” in this example Specify the IP Address, “10.1.20.23” in this example Set the Service Port to 443, then click Add Click Finished Create Virtual Server: Next we’ll create the Virtual Server From Local Traffic > Virtual Servers > Virtual Server List > click the plus sign to add a new one Give it a Name, “vs_kasm” in this example. Keep the Type as Standard. Set the Destination to the IP Address you want the BIG-IP to listen on for connections to the Kasm server, “10.1.10.100” in this example. Set the Service Port to HTTPS, port 443. Click Finished at the bottom Click on the Virtual Server you just created Click Resources Set the Default Pool to kasm_pool, then click Update The Kasm Virtual Server Status should eventually change to Green when the Health Monitor is successful. NOTE: The Virtual Server configuration in this example has been simplified for demonstration purposes. Additional configuration options will be covered later in this article. Kasm Workspaces Configuration The Kasm Workspace will need a Zone configured with the default settings. Login as Admin and check this from Infrastructure > Zones. You will need at least one Workspace. In this example, I have a Workspace with Chrome, Firefox, Terminal and Ubuntu Jammy Click the WORKSPACES Tab at the top of the screen to see what the Workspace looks like Your view should look like this: Test Kasm Workspaces Login as a User NOTE: The IP Address used to connect to the Kasm Workspaces through the BIG-IP is the Virtual Server listening IP Address 10.1.10.100 When the Workspace loads, click Firefox Choose the option to Launch Session in a new Tab After a moment, Firefox will load Here you can see the F5.com website displayed NOTE: The browser pop-up blocker can prevent the Kasm Workspace applications from successfully launching. You can disable the pop-up blocker or create an exception for the BIG-IP Virtual IP (10.1.10.100). Enable SSL Decryption Enabling SSL Decryption allows you to fully inspect the requests and payloads passing through BIG-IP. From Local Traffic > Virtual Servers > click Virtual Server List Then click the name of your Virtual Server, “vs_kasm” in this example In the Configuration section, set the Protocol Profile (Client) to http Set the SSL Profile (Client) to clientssl Set the SSL Profile (Server) to serverssl NOTE: If you have created your own Client and Server SSL Profiles, you should add them here. The instructions above are for demonstration purposes only. Scroll to the bottom and click Update You’re done! Conclusion F5 BIG-IP Zero Trust Access is a key asset to securing containerized platforms like KASM Workspaces. In this article, you learned how to secure your Kasm Workspace using F5 BIG-IP Zero Trust Access. Related Content How I did it - “Delivering Kasm Workspaces three ways” Download Kasm Workspaces Kasm Documentation
F5 BIG-IP Zero Trust Access
Introduction F5 BIG-IP Zero Trust Access, a key component of the F5 Application Delivery and Security Platform (ADSP), helps teams secure apps that are spread across hybrid, multi-cloud and AI environments. In this article, I’ll highlight some of the key features and use cases addressed by BIG-IP Zero Trust Access. F5 BIG-IP Zero Trust Access improves security and the user experience while managing access to your portfolio of corporate applications. Demo Video What is Zero Trust? Key Zero Trust Concepts Zero Trust is a cybersecurity framework built on the following core concepts: Never Trust Similar to human concepts that trust is not given freely, it is earned Always Verify Authenticate and authorize based on all available data points Continuously Monitor Zero Trust is an ongoing security framework that requires monitoring F5 enables zero-trust architectures that optimize your investments and extend zero-trust security across your entire portfolio. Why is this important? Securing apps is complex because apps are spread across a hybrid, multi-cloud environment. Apps themselves have become hybrid in nature, too. This creates 2 problems: Legacy and custom applications can complicate access security. Apps residing anywhere increases the attack surface. F5 BIG-IP Zero Trust Access secures hybrid application access. Securely managing access to corporate applications is critical to preventing data breaches. Doing it well can also increase efficiencies in business processes and user productivity. A Zero Trust security model can deliver this business value by enabling users to seamlessly and securely access their applications from anywhere regardless of where the application resides. In today’s world of hybrid, multicloud and AI applications, Zero Trust is a must. Application access control is key to any Zero Trust architecture. How does F5 address Zero Trust? F5 Zero Trust Begins with Secure Access to All Apps. The F5 Application Delivery and Security Platform (ADSP) is the foundation for Zero Trust Architectures. F5 ADSP delivers visibility, enforcement, and intelligence where it matters most: the application layer. While there are many important components to Zero Trust, we will be focusing on Zero Trust Application Access: Identity-Aware Proxy - Secure access to apps with a fine-grained approach to user authentication and authorization that enables only per-request context- and identity-aware access. Single Sign-On (SSO) and Access Federation - Integrating with existing SSO and identity federation solutions, users can access all their business apps via a single login, regardless of if the app is SAML enabled or not. OAuth 2.0 and OIDC Support - Enable social login to simplify access authorization from trusted third-party identity providers like Google, LinkedIn, Okta, Azure AD, and others. Identity Aware Proxy (IAP) – A Key Component of Zero Trust Use the Guided Configuration to configure the Identity Aware Proxy. From the BIG-IP UI, go to Access > Guided Configuration > Zero Trust. Select the Identity Aware Proxy You will see a configuration example of Identity Aware Proxy Click Next at the bottom For the Config Properties, give it a name, “IAP_DEMO” in this example Set the below options to On Click Save & Next Enable the F5 Client Posture Check Select your CA Trust Certificate Click Add Give it a Name, “FW_Check” in this example Under Windows, select Firewall and Domain Managed Devices Enter your domain name, “f5lab.local” in this example Click Done Click Save & Next Configure the Virtual Server Properties Switch Advanced Settings to On Set the Destination Address, “10.1.10.100” in this example For the Client SSL Profile, select the Client SSL Certificate, Private Key and Trusted Certificate Authorities For the Server SSL Profile, select your Server SSL Certificate and Private Key Click Save & Next Click Add under Authentication Give it a Name, “AD” in this example Set the Authentication Type to “AAA” Set the Authentication Server Type to Active Directory Choose your Authentication Server, “ad-servers” in this example Check the box for Active Directory Query Properties Under Required Attributes, find “memberOf” and click the arrow to move it to Selected Click MFA Click Add Double click Radius Under Choose Radius Server, select Create New Give it a name, “radius_pool” in this example Enter the Server IP Address, “10.1.20.8” in this example Enter the Secret in the two fields Click Save Click Save & Next Click Add Give it a name, “basic_sso” in this example For the SSO Configuration Object, click Create New The Username Source and Password Source should be set like the following Click Save Click Save & Next Under Applications click Add Give it a name, “iap1.acme.com” in this example Under Application Properties, set the FQDN and Caption to “basic.acme.com” Set the Pool IP Address to 10.1.20.7, Port 443, HTTPS Click Save For the Auth Domain, enter “iap1.acme.com” Click Save & Next Set Primary Authentication to “AD” Click Save & Next Click Add under Contextual Access For the Contextual Access Properties, give it a name, “basic.acme.com” in this example Set the Resource to iap1.acme.com Set Device Posture to FW_CHECK Set Single Sign-On to basic_sso Find the Sales Engineering Group and click Add Select the box for Additional Checks Set the Match Action to Step Up Set Step Up Authentication to Custom Radius based Authentication Click Save & Next The Remediation Page must be changed to a real host where users can download and install the EPI updates In this example, it has been changed to “https://iap1.acme.com/epi/downloads” Click Save & Next Click Save & Next Click Deploy Click Finish when the deployment completes Test the functionality by going to a client computer and accessing https://iap1.acme.com Logon with valid credentials You should see a page like the following Click basic.acme.com Login with valid credentials & click Validate You should see the basic.acme.com web page and be already logged in Note: If you disable the Windows Firewall on the client, you should get a block page similar to the following: Conclusion BIG-IP introduces a powerful access experience. BIG-IP provides a variety of Authentication, Federation, SSO and MFA protocols allowing for modern to legacy protocol translation. BIG-IP integrates with 3 rd parties to enforce identity aware decisions. BIG-IP secures identities for any apps and users anywhere in legacy and modern environments, spanning on-prem, hybrid or cloud locations. The highly scalable and proven Access Security solution that F5 customers know and trust. Related Content Zero Trust Solution Overview Secure Corporate Apps with a Zero Trust Security Model BLOG: F5 BIG-IP Zero Trust Access Zero Trust Application Access for Federal Agencies
What's new in BIG-IP v21.0?
Introduction In November of 2025 F5 released the latest version of BIG-IP software, v21.0. This release is packed with fixes and new features that enhance the F5 Application Delivery and Security Platform (ADSP). These changes complement the Delivery, Security and Deployment aspects of the ADSP. Demo Video: New SSL Orchestrator Features SNI Preservation SNI (Server Name Indication) Preservation is now supported for Inbound Gateway Mode. This preserves the client’s original SNI information as traffic passes through the reverse proxy, allowing backend TLS servers to access and use this information. This enables accurate application routing and supports security workflows like threat detection and compliance enforcement. Previous software versions required custom iRules to enable this functionality. Note: SNI preservation is enabled by default. However, if you have existing Inbound Gateway Topologies, you must redeploy them for the change to take effect. iRule Control for Service Entry and Return Previously, iRules were only available on the entry (ingress) side, limiting customization to traffic entering the Inspection Service. iRule control is now extended to the return-side traffic of Inspection Services. You can now apply iRules on both sides of an Inspection Service (L2, L3, HTTP). This enhancement provides full control over traffic entering and leaving the Inspection Service, enabling more flexible, powerful, and fine-grained traffic handling. The Services page will now include configuration for iRules on service entry and iRules on service return. A typical use-case for this feature is what we call Header Enrichment. In this case, iRules are used to add headers to the payload before sending it to the Inspection Service. The headers could contain the authenticated username/group membership of the person who initiated the connection. This information can be useful for Inspection Services for either logging, policy enforcement, or both. The benefit of this feature is that the authenticated username/group membership header can be removed from the payload on egress, preventing it from being leaked to origin servers. New Access Policy Manager (APM) Features Expanded Exclusion Support for Locked Client Mode Previously, APM-locked client mode allowed a maximum of 10 exclusions, preventing administrators from adding more than 10 destinations. This limitation has now been removed, and the exclusion list can contain more than 10 entries. OAuth Authorization Server Max Claims Data Support The max claim data size is set to 8kb by default, but a large claim size can lead to excessive memory consumption. You must allocate the right amount of memory dynamically as required based on claims configuration. New Features in BIG-IP v21.0.0 Control Plane Performance and Scalability Improvements The BIG-IP 21.0.0 release introduces significant improvements to the BIG-IP control plane, including better scalability and support for large-scale configurations (up to 1 million objects). This includes MCPD efficiency enhancements and eXtremeDB scale improvements. AI Data Delivery Optimize performance and simplify configuration with new S3 data storage integrations. Use cases include secure ingestion for fine-tuning and batch inference, high-throughput retrieval for RAG and embeddings generation, policy-driven model artifact distribution with observability, and controlled egress with consistent security and compliance. F5 BIG-IP optimizes and secures S3 data ingress and egress for AI workloads. Model Context Protocol (MCP) support for AI traffic Accelerate and scale AI workloads with support for MCP that enables seamless communication between AI models, applications, and data sources. This enhances performance, secures connections, and streamlines deployment for AI workloads. F5 BIG-IP optimizes and secures S3 data ingress and egress for AI workloads. Migrating BIG-IP from Entrust to Alternative Certificate Authorities Entrust is soon to be delisted as a certificate authority by many major browsers. Following a variety of compliance failures with industry standards in recent years, browsers like Google Chrome and Mozilla made their distrust for Entrust certificates public last year. As such, Entrust certificates issued on or after November 12, 2024, are deemed insecure by most browsers. Conclusion Upgrade your BIG-IP to version 21.0 today to take advantage of these fixes and new features that enhance the F5 Application Delivery and Security Platform (ADSP). These changes complement the Delivery, Security and Deployment aspects of the ADSP. Related Content SSL Orchestrator Release Notes BIG-IP Release Notes BLOG F5 BIG-IP v21.0: Control plane, AI data delivery and security enhancements Press Release F5 launches BIG-IP v21.0 Introduction to BIG-IP SSL Orchestrator
F5 BIG-IP APM with Omnissa Workspace ONE Access
This article discusses the collaboration between Omnissa and F5 to integrate Workspace ONE Access Cloud with F5 Access Policy Manager (APM). Workspace ONE Access unifies applications and desktops into a single, aggregated workspace, allowing employees to access resources from anywhere while simplifying IT administration through fewer management points and flexible access. The deployment of Workspace ONE Access in the cloud centralizes assets, devices, and applications, enabling secure management of users and data. Organizations benefit from instant upgrades without maintenance outages. The document provides detailed instructions for configuring Workspace ONE Access Cloud as an Identity Provider (IDP) in front of F5 APM as a Service Provider (SP), utilizing APM as a gateway for Omnissa Horizon. This integration aims to offer a comprehensive view of the workspace, ensuring robust DMZ security and scalability with the F5 PCoIP/Blast Proxy in conjunction with Omnissa Horizon.
Improve BIG-IP APM VPN speed with TLS dynamic record size
After successfully setting up BIG-IP APM network access, and running it for sometime, you may be looking for ways to optimize VPN speed for your users. This article discusses one way you can do that. Feature Description Beginning in BIG-IP 12.1.0, the Client SSL profile includes a feature that enables dynamic record size in TLS. When applied to a F5 BIG-IP Access Policy Manager (APM) network access VPN TLS virtual server, this can improve VPN speeds for your users. It has been found that certain protocols, notably HTTP, show better client response times using this method. For more information on the Allow Dynamic Record Sizing setting down to the packet level, refer to the following resources The About dynamic record sizing section of the BIG-IP System: SSL Administration manual. Boosting TLS Performance with Dynamic Record Sizing on BIG-IP on DevCentral. SSL Profiles Part 11: TLS Optimization on DevCentral. Important: Dynamic record size is a TLS enhancement and does not apply to BIG-IP APM network access DTLS virtual servers. Do not enable dynamic record size on DTLS. When you want to optimize network performance, you must allocate time to tune each configuration to match the requirements specific to your environment. Additionally, note that configuration changes that improve performance may increase BIG-IP system resource (CPU, memory) usage. Testing dynamic record size on VPN speeds Having discussed the theory behind the feature, we will now perform tests to see how it affects VPN speeds. Network bandwidth can vary depending on many factors, for instance, peak vs non-peak hours. When more users are connected to a VPN, download speeds can decrease significantly. It is therefore important to establish a baseline network bandwidth and download speed at the beginning: Baseline AWS environment Windows Client (Seattle) -- VPN --> BIG-IP APM (Oregon) --local LAN--> Apache and iperf servers AWS environment: BIG-IP APM 17.1.0 VE on AWS (F5 BIG-IP VE - ALL modules, m5.xlarge, 1 Gbps, AWS) located in us-west-2 Oregon. Note: Ensure you use at least the recommended size (m5.xlarge) and at least 1Gbps on AWS to make sure there are no bandwidth and resource limits. Windows client in located in Seattle Using iperf3 to measure network bandwidth Using curl to download a 377MB apmclient.iso Optional: You can optionally test using the developer tools on your browser. I used firefox; as the results did not differ significantly from curl. They are not included in this article. Baseline test results These are measured with all default settings on BIG-IP APM and dynamic record sizing not enabled: curl download results Average download speed: 4950k C:\Windows\system32>curl -k -o null https://10.0.128.23/apmclient.iso % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 377M 100 377M 0 0 4950k 0 0:01:18 0:01:18 --:--:-- 4988k iperf3 results Network bandwidth: 4873 KB/sec c:\Users\klau\Desktop\iperf-3.1.3-win64>iperf3.exe -c 10.0.128.24 --get-server-output -i 1 -f K -R Connecting to host 10.0.128.24, port 5201 Reverse mode, remote host 10.0.128.24 is sending [ 4] local 10.0.128.31 port 61284 connected to 10.0.128.24 port 5201 [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.00 sec 4.33 MBytes 4434 KBytes/sec [ 4] 1.00-2.00 sec 4.67 MBytes 4785 KBytes/sec [ 4] 2.00-3.00 sec 4.86 MBytes 4977 KBytes/sec [ 4] 3.00-4.00 sec 4.77 MBytes 4878 KBytes/sec [ 4] 4.00-5.00 sec 4.72 MBytes 4834 KBytes/sec [ 4] 5.00-6.00 sec 4.78 MBytes 4898 KBytes/sec [ 4] 6.00-7.00 sec 4.87 MBytes 4989 KBytes/sec [ 4] 7.00-8.00 sec 4.81 MBytes 4925 KBytes/sec [ 4] 8.00-9.00 sec 4.71 MBytes 4823 KBytes/sec [ 4] 9.00-10.00 sec 4.82 MBytes 4934 KBytes/sec - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bandwidth Retr [ 4] 0.00-10.00 sec 48.0 MBytes 4919 KBytes/sec 9 sender [ 4] 0.00-10.00 sec 47.6 MBytes 4873 KBytes/sec receiver Server output: [...] [ 5] 0.00-10.04 sec 48.0 MBytes 4900 KBytes/sec 9 sender Test 1: Enabling dynamic record size from baseline Comparing with baseline results after enabling dynamic record size Baseline: Dynamic record size disabled dynamic record size enabled Percentage improvement curl average download, k 4950 5272 6.51% iperf3 network bandwidth, KBytes/sec 4873 5138 5.44% While this may not appear to be too high on the AWS cloud, there is also received feedback from customers that they see greater improvements in environments, especially in cases where the end-to-end latencies increase. Implementation strategy and recommendations As you plan to introduce this in your environment, take note of the following recommendations: Every environment is unique Many factors can affect network performance. This can range from VLAN settings (For example. MTU), TCP settings, intermediate network device throttling, and so on. You must perform testing in your own environment before enabling the feature. Implement the feature incrementally for a selected group of users. There are different ways to do this. For example, use an iRule to redirect users based on a URL, to a separate virtual server using a different Client SSL profile that has the feature enabled. Refer to SSL::allow_dynamic_record_sizing on Clouddocs. Monitor BIG-IP system logs and resource usage After you enable dynamic record size, make sure that your BIG-IP system continues to function as expected by monitoring the following monitor /var/log/ltm and /var/log/apm log files monitor BIG-IP CPU and memory usage. For example, you can select Dashboard on the Configuration utility, generating a QKview and analyze it in iHealth and so on. For more information, refer to K71764661: Understanding BIG-IP CPU usage and K16419: Overview of BIG-IP memory usage Verify and analyze SSL statistics Use the tmsh command in K41057430: Enhanced SSL profile statistics and check for failures. The SSL Dynamic Record Sizes section should also indicate use of large record sizes. Boosting TLS Performance with Dynamic Record Sizing on BIG-IP Conclusion There are a variety of different ways to improve VPN speeds, and this article describes just one. For other options and considerations, refer to K31143831: VPN for business continuity | Chapter 5: Optimizing Network Access VPN.
Solving Secure Mobile Access with F5 and iOS 7 Per app VPN - Part 1
Overview As an F5 engineer out in the field I’m fortunate in the fact that I get to talk with customers about their projects and security concerns. While it probably would not surprise you to learn that Mobility is a key project for many organizations what does surprise me is how many are still using a layer-3 VPN approach on mobile devices. The major problem with this design is that once the VPN is established any application on the mobile platform can now access the corporate network. As we hear more and more about malware on mobile devices it is critical to start protecting corporate infrastructure by limiting access to corporate applications only. With iOS 7 Apple introduced a great way to accomplish this with their Per app VPN. Per app VPN allows iOS to control which applications have access to the VPN tunnel. This gives organizations the ability to designate which applications are corporate apps and treat everything else as personal. Per app VPN also works in Safari with a per-tab level of granularity. So I can have one tab open watching who the Houston Texan’s take in the first round draft (Johnny Manziel of course) and a second tab that is securely connected to my corporate SharePoint site. To take advantage of the iOS Per app VPN functionality Apple requires an Enterprise Mobile Management (EMM) solution to configure the mobile device and an Enterprise VPN solution like F5’s Access Policy Manager. So, if you’re anything like me you’ve scrolled past this text and straight to the pictures below because you need to deploy this ASAP right? We’ll here we go… Configuration The iOS Per app VPN uses F5’s APM SOCKS Proxy functionality so we'll need TMOS 11.4 or higher installed on the BIG-IP and Edge Client 2.0 or higher installed on the mobile device. 1. Create a new Application Policy Profile and select your default language. 2. Customize the Profile's Visual Policy Builder by adding a Client Cert Inspection object and set the successful branch to Allow 3. Create a new LTM Client SSL Profile: set Client Certificate to request set Trusted Certificate Authority to the CA that signed the certificate installed on the iOS device. 4. Create a new LTM Virtual Server: Add your customer Client SSL profile Select your Access Profile Select the default Connectivity Profile of create a custom connectivity profile with default settings Click the VDI & Java Support box to enable SOCKS proxy capabilities User Experience So What does the end result look like? In the example below I tested the Safari per-tab capabilities by clicking the F5 shortcut icon and seamlessly had access to my test web server. Next Steps In Part 2 we will walk through how I configured AirWatch to perform the user experience demonstration.
