Ok, you are right,
I only considered the case without authentication. With authentication
shared caching must not happen.
And its all about shared caches only. So I agree now we should send
Cache-Control: private with GET request too.
Stefan

Unico Hommes wrote:

But that would be a security risk no? I mean if user A has privileges to read a certain resource and user B has not. The client implementation may still decide to show user B the resource that is in the cache on the basis that it is not private to user A. Am I wrong?

--
Unico

Stefan L�tzkendorf wrote:

For GET requests it should not be send by default. They should be
cachable, I think, because it is the standard http GET.

I would prefer a configurable way. May be a Filter to be configured in
the web.xml.

Stefan

Unico Hommes wrote:

James Mason wrote:

http://issues.apache.org/bugzilla/show_bug.cgi?id=23497

It's about a year old. The patch looks innocous, from the rfc:

"private Indicates that all or part of the response message is intended
for a single user and MUST NOT be cached by a shared cache."



Looks like a directive we ought to send not only with GET/HEAD but with other read requests as well (PROPFIND/REPORT/SEARCH/etc). Except perhaps when authentication is turned off or all users operate under the same principal. Although I don't think our implementation has to differentiate between these situations.


--
Unico


--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]




---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


-- Stefan L�tzkendorf -- [EMAIL PROTECTED]


--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to