On Wed, 2 Aug 2000, ARATech Cache Team wrote:
> "Pragma: no-cache"
> "Cache-control: no-cache"
>
> When our cache cache encounters above header, it validates the
> object and serve cached copy if valid. So our cache serve cached
> copy when reload request, which yields error report in polymix-3.
I agree with Duane that you are violating HTTP, but let's forget about
HTTP for a second and just use common sense.
A "reload" request is meant to deliver a new copy of the object to the
client. If Polygraph is able to detect that the copy is not really new,
then real-world clients may have a problem with your cache serving stale
info on reload requests. In other words, what may seem like the same
object to you, may be a different object for a particular client-server
operation. Just the fact that object content is not modified does not
mean you can assume you know the semantics behind an HTTP transaction.
A simple real-world example would be sending "reload" requests to the
server to change state information on the _server_ and get new state
info via HTTP response headers. Note that HTTP prohibits sending
extension headers with 304 responses so one cannot achieve the same by
sending IMS requests.
Alex.
This archive was generated by hypermail 2b29 : Tue Jul 10 2001 - 12:00:14 MDT