- From: Roy T. Fielding <fielding@kiwi.ICS.UCI.EDU>
- Date: Wed, 19 Feb 1997 07:29:39 -0800
- To: Koen Holtman <koen@win.tue.nl>
- Cc: http-wg@cuckoo.hpl.hp.com
>>I'll second this. Further, I'd say that there is no condition under
>>which a 5xx response is justified.
>
>Say again? If there is no condition, why have a 5xx code at all? The 1.1
>spec says:
I meant in TCN, obviously.
> Response status codes beginning with the digit "5" indicate cases in
> which the server is aware that it has erred or is incapable of
> performing the request.
>
>So I think a server being aware of an internal configuration error qualifies
>for sending a 5xx.
Except that
1) There is no internal configuration error. There is no reason
why a server must be prevented from having multiple levels of
negotiation, aside from the fact it isn't in your conception of TCN.
2) The server has neither erred nor become incapable of servicing
the request, since all it needs to do is supply the client with
whatever message it got from the upstream server.
>> If the response looks bad for TCN,
>>then disable TCN for that response and flush your cache.
>
>Or are you saying that a proxy cache running this algorithm may never return
>a 5xx? The 504 code seems to contradict that.
I am saying that the 506 code has no useful purpose, and shouldn't exist.
....Roy
Received on Wednesday, 19 February 1997 07:51:01 UTC