- From: Willy Tarreau <w@1wt.eu>
- Date: Sat, 5 Aug 2017 18:02:12 +0200
- To: Eitan Adler <lists@eitanadler.com>
- Cc: Mark Nottingham <mnot@mnot.net>, Amos Jeffries <squid3@treenet.co.nz>, "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>
On Sat, Aug 05, 2017 at 08:49:50AM -0700, Eitan Adler wrote: > On 4 August 2017 at 23:08, Willy Tarreau <w@1wt.eu> wrote: > > On Sat, Aug 05, 2017 at 01:55:31PM +1000, Mark Nottingham wrote: > >> Now, we could make an argument to skip over it now and use it when we've > >> exhausted other 4NN code points, but personally my inclination is to do it > >> now; if we don't want it to ossify, the earlier the better. > > > > I have an argument for skipping it, which is that it's part of a > > contiguous range of 3 codes and that we need only one. There are > > isolated holes likes 425 and 427 and I'd rather fill these holes > > when we need a single code, and use larger areas when we need > > contiguous series. > > Is there any existing use case for contiguous error codes within the > hundred ranges? There could be. Look at the number of 30x we have. We could imagine supporting multiple retry codes later with different causes or actions for example. It serves no purpose to puch holes while it can be useful to maintain continuity. > FTR I agree with Amos Jeffries: any formal definition of 418 at least > be consistent with the well-known meaning from the joke. Yes I agree as well. I don't like the status of this code which is neither allocated nor safe to use for certain purposes due to having been heavily abused :-/ Willy
Received on Saturday, 5 August 2017 16:02:49 UTC