no cache control but still read from disk cache when download img
Reported by
denghong...@gmail.com,
Oct 12 2017
|
|||||||
Issue descriptionUserAgent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/61.0.3163.100 Safari/537.36 Example URL: https://en.wikipedia.org/wiki/IMG_(company) Steps to reproduce the problem: 1. open https://en.wikipedia.org/wiki/IMG_(company) in chrome 2. check inspector network tab, filter out img belongs to https://upload.wikimedia.org 3. refresh tab 4. img will directly read from disk/memory instead of validate What is the expected behavior? Safari, Firefox will validate img by sending a request with etag to server What went wrong? response without cache-control, but with etag, must be validated. Did this work before? N/A Chrome version: 61.0.3163.100 Channel: stable OS Version: OS X 10.12.3 Flash Version: Shockwave Flash 27.0 r0
,
Oct 12 2017
how heuristic work? any document?
,
Oct 16 2017
,
Oct 16 2017
Unable to reproduce the issue in Win-10 and MacBook Pro (Retina, 15-inch, Mid 2014), 10.12.6 using chrome reported version #61.0.3163.100 and latest canary #64.0.3241.0. Following are the steps followed to reproduce the issue. ------------ 1. opened https://en.wikipedia.org/wiki/IMG_(company) in chrome 2. checked inspector network tab, filtered out img belonging to https://upload.wikimedia.org 3. refreshed tab 4. Observed that in the header's section there was response with cache-control and validated img by sending a request with etag to server as expected. Attaching screen cast for reference. denghongcai@ - Could you please verify the screen cast and please let us know if any thing missed from our side. Also please check the issue on latest canary #64.0.3241.0 by creating a new profile without any apps and extensions and please let us know if the issue still persist or not. Thanks...!!
,
Oct 17 2017
,
Oct 18 2017
the image whose name starts with 45px from your captured screen video, it reads from cache without cache-control
,
Oct 18 2017
Thank you for providing more feedback. Adding requester "krajshree@chromium.org" to the cc list and removing "Needs-Feedback" label. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Oct 19 2017
This resource in the screenshot in #0 has a large age. In the absence of any explicit lifetime indicator, it looks like the UA is well within the requirements for assuming the resource is fresh based on the freshness_lifetime heuristic. The presence of the Etag itself doesn't mandate revalidation. I'll add the cache folks who might have a more definitive answer.
,
Oct 27 2017
Could someone from Internals>Network>Cache team please take a look into this issue. Thanks..!
,
Oct 27 2017
To be more explicit about the heuristic, it uses 10% of time between now and last-modified if that is available, and the response code is appropriate for that (which includes 200) which is the heuristic suggested in section 4.2.2 of RFC 7234: https://tools.ietf.org/html/rfc7234#section-4.2.2 " If the response has a Last-Modified header field (Section 2.2 of [RFC7232]), caches are encouraged to use a heuristic expiration value that is no more than some fraction of the interval since that time. A typical setting of this fraction might be 10%. " If you always want the resource to be revalidated, please consider something like must-revalidate or max-age=0.
,
Oct 27 2017
Closing issue since the current behavior is compliant. |
|||||||
►
Sign in to add a comment |
|||||||
Comment 1 Deleted