9

So we now know who wrote the bad code (Robin Seggelmann). And we have an idea of why it is needed: Why does TLS need an explicit heartbeat protocol?

We can also understand why the client supplies the length: Heartbleed: Why does the client supply the length of the message at all?

But who proposed the form of the ssl heartbeat? Could there be some nefarious input from the NSA in the hopes that someone would code it the wrong way and leave it vulnerable for easier snooping?

I've heard rumors that they added flaws to several encryption schemes (if I could quote which ones and where I would), could this be yet another way they introduced insecurity in their desire to have a copy of everything?

I know this sounds like I might be wearing a tinfoil hat, but given the lengths they have gone to since 9/11, this does not seem out of the realm of possible.

1 Answer 1

10

The RFC6520 describing heartbeats was had R. Seggelmann of Muenster Univ. of Appl. Sciences as its first author. Seggelmann denies intentionally introducing the flaw and claims to be unaffiliated with any intelligence agency. Similarly, the NSA and the Director of Homeland Security has publicly denied knowing about Heartbleed vulnerability, though Bloomberg reported from two people familiar with the matter that the NSA did know about the vulnerability and regularly used it.

Granted, looking at egress logs it appears attackers were using heartbleed style attacks in the wild last year from IP addresses that appear to have been used to systematically log IRC chats (making it suspicious to be intelligence agencies). (So even if they didn't introduce heartbeats, it appears they were aware of it and didn't publicly disclose it).

Further, weirdness things is that two independent groups apparently discovered heartbleed (google and codenomicon) on the same day over two years after OpenSSL released the vulnerable HB protocol in 1.0.1. The google engineer tweeted claimed "Heap allocation patterns make private key exposure unlikely #heartbleed #dontpanic", while the codenomicon tweeted "We're very certain" encryption keys can be stolen.

1
  • 1
    So the same guy that did the insecure implementation, is the guy who authored the RFC in the first place. Along with M. Tuexen who has 22 RFCs to his credit.
    – boatcoder
    Apr 14, 2014 at 12:29

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .