Re: draft-ietf-tls-http-upgrade reissued

• Messages sorted by: [ date ][ thread ][ subject ][ author ]
• Next message: Ari Gordon-Schlosberg: "Re: Has Netscape 5 become 6?"
• Previous message: Kaming Young: "Has Netscape 5 become 6?"
• Maybe in reply to: Scott Lawrence: "draft-ietf-tls-http-upgrade reissued"

Julien Pierre wrote:

> If the client tries to upgrade to TLS on every request, it will fail 99% of the time,
> because servers don't support it.

Servers that don't support it ignore it, because RFC-2616 doesn't provide an error code to
mean "I don't understand that Upgrade:".  (I just tried it on Apache 1.3.9 and 1.3.12, NES
3.6, and IIS 5.0; they all behaved exactly the same with and without "Upgrade: foo".) This
means that a client that sends the Upgrade: all the time won't break anything; it will cost
a few extra bytes, but not the extra round trips you're talking about.

--
/================================================================\
|John Stracke    | http://www.ecal.com |My opinions are my own.  |
|Chief Scientist |===============================================|
|eCal Corp.      |The cheapest, fastest, most reliable components|
|francis@ecal.com|of a computer system are those that aren't     |
|                |there.--Gordon Bell                            |
\================================================================/

• Next message: Ari Gordon-Schlosberg: "Re: Has Netscape 5 become 6?"
• Previous message: Kaming Young: "Has Netscape 5 become 6?"
• Maybe in reply to: Scott Lawrence: "draft-ietf-tls-http-upgrade reissued"