Re: [TLS] I-D Action:draft-ietf-tls-rfc4366-bis-09.txt

[Juho Vähä-Herttua <[e-mail]>]

On 13.6.2010, at 12.01, Juho Vähä-Herttua wrote:
> Another benefit comes from DTLS, where TLSRecords themselves need to be fragmented in order to fit inside a single IP packet. If the implementation can negotiate a fragment length with a max_fragment_length extension, it doesn't have to be able to handle the double fragmentation (first fragment handshake, then encrypt, then fragment the encrypted data).

I should get rid of the bad habit of reviewing documents after sending, but I remembered the DTLS specification wrong. The TLSRecords can't be fragmented, but instead the mentioned process is simply "fragment handshake" -> "encrypt". That's my bad, this argument is not valid.

However the draft itself says "memory constraints among some clients, and bandwidth constraints" which both seem to be understandable. With a limited bandwidth small TLSRecord fragments is a good thing, because the data can be processed faster instead of in very large chunks.


Juho