Stay Up to Date on TLS with BreakingPoint 9.10
2020-09-16 | 5 min read
By Daniel Munteanu | Since many BreakingPoint releases back, transport layer security (TLS) has been top of mind for Keysight, driven by a diverse set of customer requirements and the continuous evolution in encryption standards. Now, more services and systems are turning on the latest TLS 1.3 encryption standard by default. For example, check the recent move from Microsoft in this blog as well as Google’s announcement here.
Moves like this, where both the client and server communication endpoints are supporting TLS 1.3 by default, are putting increased pressure on middle boxes in many various forms. These depend on the functionality that the middleboxes are performing, be it full man in the middle (MiTM) interception, just cleartext header inspection, or simple passthrough. Such network devices need proper testing and validation with the new production environment realities to ensure that their functions and performance are still within acceptable limits.
In the latest BreakingPoint 9.10 release, we have rounded up the TLS 1.3 functionalities and added a few more generic TLS-related enhancements that empower users to emulate use cases and test scenarios matching the new developments seen in production networks.
Therefore, I want to give you a quick overview of the most important new BreakingPoint TLS 1.3 functionalities.
First, I will start with the ability to downgrade TLS 1.3 sessions to TLS 1.2 on the fly, while keeping other already established TLS 1.3 sessions going. This is an important capability as TLS 1.3 is still relatively new and some network devices are not yet tunned to fully support it. Seamlessly testing with TLS 1.3 and TLS 1.2 (downgraded on the fly from TLS 1.3) in parallel will provide a full, realistic environment to understand the impact on the end-user application traffic as well as performance limits of the network device.
Another important enhancement is the so-called “Middlebox Compatibility mode”. As per RFC8446, “a significant number of middleboxes misbehave when a TLS client/server pair negotiates TLS 1.3. Implementations can increase the chance of making connections through those middleboxes by making the TLS 1.3 handshake look more like a TLS 1.2 handshake”. Moreover, there are middleboxes that even in a full MiTM setup are enforcing this mode as well:
One more important addition is the ability to specify for the emulated clients a list of TLS server name indications (SNIs). While this might sound like a no brainer, it highly simplifies testing for end TLS-enabled infrastructures as users can iterate through hundreds of SNIs, hostnames, and URLs. All these can be done at high scale and high performance, stressing the system under test and exercising it under different test scenarios. This feature can be used for either TLS 1.3 or for TLS 1.2/1.1.
One important improvement in TLS 1.3 compared to TLS 1.2 and previous versions that is also often overlooked is the fact that the client authentication does not expose anymore the client identity on the network providing boosted security when client authentication is required. To properly test such scenarios, starting with release 9.10, BreakingPoint supports client authentication for TLS 1.3 as well.
Other newly introduced TLS 1.3 enhancements include, among others: pre-shared key (PSK), Eduard Curve Certificate support, and the ability to configure the sessions to be reused as a percentage.
There are obviously many other features and enhancements added with the BreakingPoint 9.10 release and I really encourage BreakingPoint users to check out the full list in the release notes, here (requires customer login).