developer tool's network tab is showing the wrong line number in initiator column for the cached resources
Reported by
siddhart...@gmail.com,
Aug 16 2016
|
|||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/52.0.2743.116 Safari/537.36 Steps to reproduce the problem: 1. Open network tab in developer tool and load a page having multiple resources 2. Check for line numbers of some of the resources in initiator column , It will give you the correct line number for example for a resource the line number in initiator is 900 3. Now visit another page in same application having reference to same resource that has been cached in the previous request 4. Check initiator column for the same resource in network tab it is showing the same line number as 900 with the new page name if it was cached in last request What is the expected behavior? Initiator column should display the correct line number where this resource is references in the new page What went wrong? Check initiator column in second request for the same resource in network tab it is showing the same line number as 900 with the new page name. If the refresh the pace SHIFT+F5 than all the resources are reloaded with the correct line number Did this work before? N/A Chrome version: 52.0.2743.116 Channel: stable OS Version: 6.1 (Windows 7, Windows Server 2008 R2) Flash Version: Shockwave Flash 22.0 r0 Please let me know if further information is required
,
Aug 17 2016
Could you please provide a sample application to reproduce the issue from test team end. A screencast would be helpful
,
Aug 22 2016
I have attached 6 screenshots where gaTracking.js is referenced at line number 2601 in homepage when i clicked on the initiator link it will landed to homepage line number 2601 in resources tab where it is referenced (refer scr2.png). When I visit prescription page the gaTracking.js is still showing line number 2601(refer scr3.png) in prescription page and when I clicked on the prescription page this page is not having line number 2601 (refer scr4.png). When I reloaded the page shift+F5 it is showing the correct line number 1025 (refer scr5.png and scr6.png) The application is not live so i can't provide the URLS if i found same behaviour in a live application than i will share the URLS. Please let me know if further information is required from my side .
,
Aug 29 2016
Thank you for providing more feedback. Adding "Needs-Review" label for tracking. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Sep 12 2016
,
Sep 12 2016
Hmmm interesting, I think this may be caused from a combination of reasons that line up for you. 1) Can you please ensure it still exists on stable? 2) Check to see if it exists on Canary? 3) On stable try disabling cache and see if it exists? Thanks!
,
Sep 13 2016
I further analysed below are the findings. Able to replicate the issue only on the development environments not having valid SSL certificates(non disable cache) not able to replicate on the environment having the valid SSL certificate, Is this something related to SSL certificate? Thanks
,
Sep 13 2016
Reload behavior is/was going through a change in the way it works when on the same domain. I am not sure if SSL follows the same flow path (probably not). Do you have a sample demo for us to investigate this issue? Can you check to see if it exists on Canary?
,
Sep 14 2016
I am able to replicate same issue on Canary also. And I found more details on canary while analysing on both the environments with and without valid SSL. As per my analysis chrom maintains two king of cache memory and disk caches. and issue only exists for the resources stored on memory cache if any resource is stored in disk it is showing the correct line number. not sure why it is storing resources from same domain in memory if not on valid SSL and storing on disk if on valid SSL. Hoe this helps!
,
Sep 14 2016
Correct. I was betting on disk vs memory cache being the issue as well. +toyoshim, do you see why this might be happening on the top of your head?
,
Sep 15 2016
,
Sep 15 2016
It could be possible if initiator information were stored in memory cache, but actually I don't know details how initiator information are reported to the DevTools. allada@, what kind of interface should be called to report such information? If it was called from memory cache or related code, we'd try to fix this. CCed to hiroshige who are working on memory cache.
,
Sep 15 2016
I have another bug dealing with redirects and initiator info. I have identified the bug to be because "willSendRequest" is being sent twice to devtools causing the GUI to act strange. I think we may be dealing with a similar situation. Let me investigate this possibility further before anyone spends any more time on it.
,
Dec 1 2016
,
Dec 1 2017
We need to investigate. Stray guess is preload scanner is caching something.
,
Dec 14 2017
,
Sep 18
|
|||||||||||
►
Sign in to add a comment |
|||||||||||
Comment 1 by l...@chromium.org
, Aug 16 2016