Radware Alteon SSL Orchestrator deployment guide
Radware Alteon can act as an SSL Orchestrator by terminating TLS sessions, making selected traffic visible to a security inspection path, and opening a new TLS session toward the origin. With the appropriate licensed features, this is more than SSL offloading: decrypted requests can be filtered by application, sent to an inspection service, or re-encrypted for onward delivery.
Australian hybrid estates make this pattern valuable. Applications may run in on-premises racks or the AWS Sydney and Melbourne regions while staff work across distributed offices and homes; data residency commitments often shape which backend pools are eligible. Because decrypted content can contain personal information, access, logging and retention must account for the Privacy Act 1988 and the Notifiable Data Breaches scheme.
Map traffic before changing the path
Begin with a traffic inventory rather than a certificate change. For each application, record the client entry point, public DNS name, Alteon virtual server, origin pool, backend ports and business owner. Identify mobile, VPN, partner and site-to-site paths, since each may require a different trust model or bypass rule.
Then draw both forward and return paths. Mark where TLS terminates, where plaintext exists, which firewall routes it and how health checks reach each pool. Decide explicitly whether traffic should fail open or fail closed during an Alteon, inspection appliance or certificate failure. Record these decisions in change documentation and test them later under real load.
Prepare the certificate trust chain
Clients must trust the certificate authority that signs the Alteon’s server certificate. A common approach is a private enterprise CA whose root is distributed through mobile-device management to managed Windows, macOS, Android and iOS devices. Test pilot cohorts before broad deployment, and use an expiry alert rather than relying on users to install roots manually.
Configure the orchestrator’s certificate, private key and upstream trust store according to the running Alteon release. Validate origin certificates and hostnames, enable revocation checking where practical, and synchronise Alteon time with reliable NTP sources. Restrict key exports and administrative access while aligning TLS versions and cipher choices with client capability and security policy.
Connect Alteon into the inspection path
A typical inline sequence is client to Alteon on 443, Alteon to the inspection service over a protected segment, inspection service back to Alteon, then Alteon to the origin over TLS. Use a dedicated routed segment or VLAN pair for decrypted traffic and prevent unintended forwarding back to the internet. Confirm that return routes, MTU settings, NAT and firewall rules preserve the source address when the application needs it.
If Alteon and the inspection appliances are virtualised in vSphere, apply VMware network tuning so vMotion traffic does not compete with latency-sensitive inspection flows. Keep orchestration interfaces on appropriately redundant networks and account for storage, CPU and NIC headroom. When running an HA pair, confirm state synchronisation, certificate synchronisation and whether active-standby or active-active operation matches the failure mode.
Configure application delivery policies
Build application delivery policies in a deliberate order. A suitable sequence is listener and certificate selection, TLS parameters, decryption or bypass decision, destination routing, health checks and logging. Exact labels vary by Alteon release and licensed package, but the order remains important. Use the actual hostname or SNI as a policy condition where possible because applications sharing a wildcard certificate may otherwise reach the wrong upstream. Keep explicit rules ahead of broad defaults.
For direct destinations, configure upstream TLS re-encryption with correct server-name validation. For inspected destinations, define how traffic reaches the inspection appliance and how it returns to Alteon. Narrow exceptions for certificate-pinned applications, unsupported protocols and latency-sensitive workloads, and give each exception an owner and review date rather than leaving unexplained gaps.
Limit logs to metadata needed for operations and investigation, such as client address, SNI, policy name, verdict and timing. Avoid recording request bodies, credentials or sensitive form fields. Define retention and access controls that support investigation without creating a second, poorly governed copy of the plaintext data seen by the application.
Validate decryption and resilience
Test the certificate chain from a representative client with openssl s_client -connect app.example.au:443 -servername app.example.au, then confirm that the inspection path receives the expected hostname, headers and request. Compare each hop independently because a valid browser response can still conceal a broken upstream re-encryption or return route. Use synthetic transactions that contain no production secrets.
Measure connection setup time, throughput, CPU, memory and inspection latency under expected peak load. During maintenance, verify graceful draining, HA failover, certificate renewal and restoration of both direct and inspected paths. Confirm ALPN and HTTP/2 behaviour as well as cipher negotiation, since an inspection hop may not support the same features as the public service. Packet captures should be short, access-controlled and performed where secrets can be excluded.
Compare deployment options
The deployment choice depends on where the inspection function belongs and how much traffic must remain visible.
| Deployment option | Decryption path | Advantages | Trade-offs |
|---|---|---|---|
| Alteon reverse proxy with upstream re-encryption | Client to Alteon, then TLS from Alteon to the origin | Centralised TLS policy, load balancing and hostname routing | Alteon sees plaintext, but content inspection requires another capability |
| Alteon service chaining | Client to Alteon, plaintext inspection hop, return to Alteon, TLS to origin | Supports dedicated security appliances and application-specific decisions | Added latency, routing complexity and more components to operate |
| Passive traffic mirroring | Encrypted traffic copied from a switch or virtual network | Low impact and useful for analytics or forensic tooling | Cannot transparently decrypt or control live client sessions |
| Upstream secure web gateway | TLS inspection performed by the firewall or security gateway | Consolidates inspection within network-security tooling | Application routing and policy ownership may be divided between systems |
For most Australian deployments, Alteon service chaining provides a clear balance between application control and inspection coverage, particularly where a dedicated security appliance already exists. The operating model should also demonstrate alignment with the Australian Signals Directorate’s Essential Eight. Carriers and service providers handling voice traffic need to assess the Telecommunications Act 1997 and applicable service-provider codes before introducing decrypted content into their networks.