Re: HalfSipHash Acceptable Usage

On Tue, 2016-12-20 at 16:36 -0500, Theodore Ts'o wrote:

On Mon, Dec 19, 2016 at 06:32:44PM +0100, Jason A. Donenfeld wrote:

quoted

1) Anything that requires actual long-term security will use
SipHash2-4, with the 64-bit output and the 128-bit key. This includes
things like TCP sequence numbers. This seems pretty uncontroversial to
me. Seem okay to you?

Um, why do TCP sequence numbers need long-term security?  So long as
you rekey every 5 minutes or so, TCP sequence numbers don't need any
more security than that, since even if you break the key used to
generate initial sequence numbers seven a minute or two later, any
pending TCP connections will have timed out long before.

See the security analysis done in RFC 6528[1], where among other
things, it points out why MD5 is acceptable with periodic rekeying,
although there is the concern that this could break certain hueristics
used when establishing new connections during the TIME-WAIT state.

[1] https://tools.ietf.org/html/rfc6528

We do not use rekeying for TCP ISN, not anymore after commit
6e5714eaf77d79ae1 (where we switched from MD4 to MD5 )

It might hurt some common cases and I do not believe it is mandated by a
current (ie not obsolete) RFC.

Our clock has a 64 ns resolution and 274 second period (commit
9b42c336d0641) (compared to 4 usec one in RFC 6528)

I do not see why SipHash, if faster than MD5 and more secure, would be a
problem.

Same for syncookies.

BTW, we probably should add a ratelimit on SYNACK retransmits,
because it seems that attackers understood linux kernels resist to
synfloods, and they (the bad guys) use reflection attacks.