fre 2006-06-02 klockan 13:10 -0400 skrev sknipe@xxxxxxxxxx: > Could you please verify the following. When ETag is used > for .html pages, the caching seems to be at the Browser > level. Does Squid play a role in the if-none-match with > ETags? Caching is still done at the proxy. ETag changes nothing of that. What ETag does is only to allow the use of If-None-Match which is a stronger cache validation than If-Modified-Since, allowing detection of object changes within the same second or when an object is replaced by an older object.. (i.e. restored from backup or similar). > Is there a way to have an HTTP/1.1 or HTTP/1.x response > generated for IE? Squid is HTTP/1.0. It can't respond with HTTP/1.1 as it does not yet support transfer-encoding and a few other critical HTTP/1.1 components wich MUST be supported to announce HTTP/1.1. Responding with HTTP/1.1 without supporting HTTP/1.1 would result in a lot of problems as the transfer encoding changes the HTTP/1.1 message format (it's not just some added header or other trivial change..) Clients are not supposed to make many other differences in their behavior because of this other than to not use transfer encoding. Using If-None-Match within HTTP/1.0 is fine and even required by the HTTP/1.1 specification, but should for completeness be combined with a If-Modified-Since in case the server or proxy does not support If-None-Match. Please note that this is even the case if the server is HTTP/1.1 as supporting If-None-Match is optional. See RFC2616 (HTTP/1.1) section 13.3.4 "Rules for When to Use Entity Tags and Last-Modified Dates" for a detailed description on when to use what and the reasoning behind this.. Regards Henrik
Attachment:
signature.asc
Description: Detta =?ISO-8859-1?Q?=E4r?= en digitalt signerad meddelandedel
