Re: Does no-store in request imply no-cache?

in reference to your statement

"Actually, unsafe request methods MUST invalidate the cache; since the
response may not follow the same path as the request, there's no way
for a cache to know whether the request was successful or not -- better
safe than sorry."

The "path" I'm talking about is a chain of TCP-based HTTP agents not 
about IP hosts (irrelevant in this discussion).

talking about IP and tracerts is a complete red herring.  These agents 
are the parties in TCP connections.  Sure, the IP packets may go via 
different routers between the endpoints, but the endpoints are the 

On 18/10/2010 3:44 p.m., Eric J. Bowman wrote:
> Adrien de Croy wrote:
>> I don't see how on a TCP connection a response message can go via a
>> different path than the request message.  That would break everything.
> IP is connectionless.  Asymmetrical routing is quite common on the
> Internet, don't make me post a traceroute...  TCP is what keeps
> everything from breaking.
> -Eric

Received on Monday, 18 October 2010 03:24:46 UTC