Re: introduction and ETag question

From: Henrik Nordstrom <hno@dont-contact.us>
Date: Mon, 5 Apr 2004 18:45:35 +0200 (CEST)

On Mon, 5 Apr 2004, Mati wrote:

> Our area of interests is ETag support in squid 3.
>
> During our tests with etag-patched squid 2.5 we discovered some
> differences between squid's behavior and rfc2616...
>
> One of them is major.
>
> I would like to know whether this difference is based on some
> considerations about nature of ETags and therefore should be ported to
> squid 3, or not...

If there is a difference between RFC2616 and the ETag patch to 2.5 then
what is said in RFC2616 is what should apply.

> section 13.3.4 of rfc2616 states:
>
> "proxy, upon receiving a conditional request that includes both a
> Last-Modified date and one or more entity tags as cache validators, MUST
> NOT return a locally cached response to the client unless that cached
> response is consistent with all of the conditional header fields in the
> request"
>
> We understand that if request has both If-Modified-Since and
> If-None-Match, squid should return cached responce if the validation
> holds true for both conditional headers.

Yes.

> But, patched squid 2.5 returns cached response if either of conditional
> headers validates correctly...
>
> Is it correct?

No, This is a bug.

> Are there some other documents or opinions (beside rfc2616) justifying
> this difference?

No.

Regards
Henrik
Received on Mon Apr 05 2004 - 10:45:42 MDT

This archive was generated by hypermail pre-2.1.9 : Thu Apr 29 2004 - 12:00:03 MDT