[PATCH] Etag responses need Vary header

[Dave Gynn <dgynn@…>]

The Request.check_modified() function which adds the Etag and checks If-None-Match should be adding a "Vary *" header along with the initial response and the 304 response.

​http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.44

This creates a problem of the browser always caching attachments when mod_deflate is used. mod_deflate adds a "Vary Accept-Encoding" header which makes the browser believe that only Accept-Encoding is used to determine cacheability. "Vary *" is needed to let the browser know that other information is used to make caching decisions.

Also note that Request.send_file() function does not send a Cache-Control header. That should be OK if Vary * is sent.

[Christian Boos]

cmlenz, can you validate the proposed approach?

[Christian Boos]

need someone who understands the issue ;-)

[Christian Boos]

Needs a caretaker.

[Christian Boos]

In light of #9936, we should probably fix this one as well.

[Volker Braun <Sage@…>]

I have a suspicion that this bug causes our transparent web proxy to serve stale copies of trac raw-attachments. See ​http://trac.sagemath.org/sage_trac/ticket/11813 for details including HTTP headers. Please let me know if there anything I can do to debug this further.

[Remy Blank]

Preparing for 1.0.

[Ryan J Ollos]

Moved ticket assigned to next-stable-1.0.x since maintenance of 1.0.x is coming to a close. Please move the ticket back if it's critical to fix on 1.0.x.

[Ryan J Ollos]

Milestone renamed