Antwort: [Mod_gzip] "mod_gzip_send_vary=Yes" disables caching on IE? (1.3.26.1a)

[Jordan Russell]

Michael.Schroepl@telekurs.de wrote:
> so you actually do compress *.gif files - what for?
> You would not gain a lot by this, maybe even gain nothing.

I realize this; I wasn't intentionally compressing .gif files.

The caching problem however is *not* limited to embedded images; I've found
that it apparently happens with any type of file, including HTML pages. With
mod_gzip_send_vary=Yes, every time someone visits a page using IE, it's
reloaded from scratch. And -- get this -- using the Back and Forward buttons
in the browser also causes a reload (!). This means that even sites with
100% dynamically-generated pages suffer from bandwidth waste. Clients, of
course, will also observe delays.

> Sure. You should at least serve the correctly working clients
> what they deserve, and that is a HTTP response labelled as
> a negoriation result. That is what "Vary:" is meant for.

Unfortunately, though, less than 5% of clients fall under the "correctly
working" category.

> You can't detect the proxy servers - they aren't required to
> send a "Via:" header. They only do it if they decided being
> nice to you.

"aren't required"? Maybe I'm missing something, but RFC 2616 seems to state
otherwise ("MUST"):


14.45  Via

   The Via general-header field MUST be used by gateways and proxies to
   indicate the intermediate protocols and recipients between the user
   agent and the server on requests, and between the origin server and
   the client on responses. It is analogous to the "Received" field of
   RFC 822 [9] and is intended to be used for tracking message forwards,
   avoiding request loops, and identifying the protocol capabilities of
   all senders along the request/response chain.


Jordan Russell