POST request body not recorded if the response has a "302 Found" status code
Reported by
pkts...@gmail.com,
Sep 19 2017
|
||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/61.0.3163.79 Safari/537.36 Steps to reproduce the problem: 1. Create account, sign in, and go to http://playground.xwiki.org/xwiki/bin/view/Sandbox/TestPage1 2. Edit the page (add "xyzzy", for example) 3. Start network log in Developer tools 4. Save the page ("Save and view") 5. Look at the POST request (first entry) in the network log What is the expected behavior? I should be able to see the body of the POST in the request. What went wrong? The body of the post is absent from the log, even though the change to the page was saved. Did this work before? N/A Chrome version: 61.0.3163.79 Channel: n/a OS Version: 10.0 Flash Version: File #1: playground.xwiki.org.har shows the problem File #2: www.cs.tut.fi.har shows what a normal POST looks like Note: I'm guessing that it's the 302 response causing this, it could be something else.
,
Sep 20 2017
This seems to be also related with these reports on Stack Overflow: https://stackoverflow.com/questions/46311968/google-chrome-network-developer-tools-form-data-is-missing-from-headers-tabs-o https://stackoverflow.com/questions/46237449/chrome-developer-tools-do-not-show-form-data-on-windows-server-2012
,
Sep 20 2017
Same issue on OSX too with 302 request
,
Sep 20 2017
Tested on Chrome Stable #61.0.3163.91, Canary #63.0.3220.0 and Chrome #50.0.2624.0 on Windows 10, Mac 10.12.6, Ubuntu 14.04 and able to reproduce the issue. Screenshot is attached for reference. This is a non-regression issue and able to reproduce from M-50 #50.0.2624.0. Marking it as untriaged so that issue gets addressed. Thanks.
,
Sep 28 2017
Same here, when the POST request resulted in a status 302, the form data for the POST request is *not* showing. Form data is showing when it is a 200. MacOS High Sierra 10.13, Chrome 61
,
Oct 2 2017
Issue 770130 has been merged into this issue.
,
Oct 3 2017
This is an important feature for developers. We're experiencing the problem on Windows 10 64 bit with Chrome 61.0.3163.100.
,
Oct 3 2017
I do not think this should be marked as a non-regression issue. We have multiple developers working in our team who utilize this feature and it worked until recently. They are currently having to use FireFox to view the POST data.
,
Oct 3 2017
I agree. This is a regression and should be marked as such. It's one of the main features in Chrome Developer Tools that my development team uses. It was working before and now it isn't, must be a regression.
,
Oct 3 2017
,
Oct 5 2017
Hi guys, I got the exactly same issue here: For 302 POST, no form data is displayed in Network Tab. I am using the Version 63.0.3229.0 (Official Build) canary (64-bit)
,
Oct 9 2017
Issue seen in Version 61.0.3163.100 (Official Build) (64-bit) MacOS Sierra v10.12.6
,
Oct 9 2017
Issue present in Chrome 61.0.3163.100 and not present in Chrome 49.0.2623.112
,
Oct 12 2017
Serious bug - Please fix ASAP. This made SAMLResponse POST not show up for debugging.
,
Oct 16 2017
Serious bug. Present in Version 61.0.3163.100 (Official Build) (64-bit) on Mac OSX
,
Oct 16 2017
The problem happens as well with "204 No-Content" response code. Tested Chrome version 61.0.3163.100 on Mac OSX.
,
Oct 16 2017
Our dev team is also finding that 500 response XMLHTTPRequests are not showing responses in dev tools. This means it is not possible to read the exception message if it is returned with a 500 status. These were visible in previous versions. @allada Any updates on the issue from Chromium?
,
Oct 17 2017
Issue present in Chrome 61.0.3163.100 and not present in Chrome 58.0.3029.110 Any updates?
,
Oct 17 2017
This is a killer issue for enterprise app teams working on SAML integrations. Any updates?
,
Oct 17 2017
I'm guessing it will take a while till this bug gets fixed. I am using Firefox' developer console in the meantime
,
Oct 19 2017
who need chrome with working feature https://browser.yandex.com
,
Oct 19 2017
While waiting for the bug fix you can use Fiddler https://www.telerik.com/fiddler
,
Oct 23 2017
It's been two weeks since the bug was assigned. Is there any update/progress on this?
,
Oct 26 2017
Same issue with newest version chrome on Windows 7. I think this priority should be highest.
,
Oct 30 2017
Ironically this works fine in the current version of Opera (48.0.2685.52)
,
Nov 1 2017
This should not be a low bug. this bug is making me use firefox due to not giving me the form data please increase from low to high and fix asap :)
,
Nov 2 2017
The request payload is not shown also when the request is blocked using `Request blocking`. Ironically I can see the request payload fine using a Vivaldi build based on Chrome/61.0.3163.102
,
Nov 3 2017
This is a major bug for my team. We are being forced to use other browsers in order to debug.
,
Nov 6 2017
I hope this isn't done on purpose to avoid unauthorized users to read the passwords saved in the browser (that is indeed a security issue, but this is the wrong way to address it)
,
Nov 6 2017
To all the need-to-comment-with-my-pain-fix-ASAP people. Please just use the freaking star option! :D Note: Its open source if you want to speed things up make a PR Best from A happy developer only interested in when the issue is fixed and not pain comments emails :D The irony is also - this is a pain comment to :D
,
Nov 7 2017
,
Nov 8 2017
Serious bug. I have to switch to firefox to get the request playload
,
Nov 9 2017
This was caused by plznavigate. It appears that when the renderer submits a navigation request to browser the renderer gets notified of the request but somewhere it drops the request's HTTPBody. eostroukhov@ is looking into it. cc: clamy@ (just in case she knows exactly where it does this, otherwise we'll find it).
,
Nov 9 2017
Yes this happens here: https://cs.chromium.org/chromium/src/content/browser/frame_host/navigation_request.cc?q=navigationrequest&dr=CSs&l=617. +creis who reviewed the CL where we introduced this. This was a deliberate choice not to transmit the POST data to the renderer process if the request was redirected to a GET. The main issue here is that DevTools only hears from the navigation after it is ready to commit.
,
Nov 15 2017
,
Nov 15 2017
,
Nov 15 2017
Issue 755997 has been merged into this issue.
,
Nov 15 2017
,
Nov 21 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/649cc21a76d5e7e6dba6077b8013d3b8bf4ccc2d commit 649cc21a76d5e7e6dba6077b8013d3b8bf4ccc2d Author: Nathan Bruer <allada@chromium.org> Date: Tue Nov 21 04:31:34 2017 Allows renderer gets post data if devtools attached w/ plzNavigate Adds a temporary fix to allow devtools to get the post data of navigation request if the renderer has devtools attached. R=caseq,clamy BUG= 766715 Cq-Include-Trybots: master.tryserver.chromium.linux:linux_site_isolation Change-Id: I875ea05e778fc431b6073a6fb6b1c393bcc904be Reviewed-on: https://chromium-review.googlesource.com/770490 Reviewed-by: Charlie Reis <creis@chromium.org> Commit-Queue: Blaise Bruer <allada@google.com> Commit-Queue: Blaise Bruer <allada@chromium.org> Cr-Commit-Position: refs/heads/master@{#518109} [modify] https://crrev.com/649cc21a76d5e7e6dba6077b8013d3b8bf4ccc2d/content/browser/frame_host/navigation_request.cc [add] https://crrev.com/649cc21a76d5e7e6dba6077b8013d3b8bf4ccc2d/third_party/WebKit/LayoutTests/http/tests/inspector-protocol/network/redirect-302-with-post-expected.txt [add] https://crrev.com/649cc21a76d5e7e6dba6077b8013d3b8bf4ccc2d/third_party/WebKit/LayoutTests/http/tests/inspector-protocol/network/redirect-302-with-post.js
,
Nov 21 2017
,
Nov 21 2017
This bug requires manual review: We are only 13 days from stable. Please contact the milestone owner if you have questions. Owners: cmasso@(Android), cmasso@(iOS), gkihumba@(ChromeOS), govind@(Desktop) For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Nov 21 2017
This issue has been exists since M61 and we're very close to M63 stable promotion. Issue listed at #40 is not yet baked/verified in Canary/dev, so I'm afraid it is too late for M63 as at this point as we're only taking absolutely critical and safe merges in.
,
Nov 21 2017
@govind: the fix looks safe to me and 114 stars is on the higher end of our bug spectrum. let's see how it behaves on Canary vs how close M63 promotion actually is. If M63 requires another set of fixes, we could try merging this one in as well.
,
Nov 21 2017
Ok, pls update the bug with canary result. If canary result looks good and merge is approved, merge has to happen latest by 11:30 AM on 11/28 in order to make it to last M63 Beta release. Thank you.
,
Nov 22 2017
When will this fix be added to Canary? I just installed Canary (Version 64.0.3273.3) and the problem still happens. I am willing to help test it. Thanks.
,
Nov 22 2017
There should be a new canary build now,. You can go ahead and test it. Since this issue is not a regression in M63, I wonder if you have considered letting it bake longer and release it with M64 instead.
,
Nov 22 2017
Enthusiastically confirming fixed in 64.0.3275.0. Would be awesome if this could get into M63. Thanks!
,
Nov 23 2017
I can also confirm that it is fixed in 64.0.3275.0. @cmasso: Why do you say that this is not a regression in M63? I can confirm that this issue happens in (63.0.3239.59 (Official Build) beta (64-bit)).
,
Nov 28 2017
allada@/ pfeldman@, how is the change looking in Canary so far? Note: We're cutting M63 last Beta tomorrow. So if change looks good in canary and approved, pls plan to merge your change to M63 branch 3239 before 12:30 PM, Tuesday (11/28/17) so we can take it in for this week last Beta release.
,
Nov 28 2017
,
Nov 28 2017
govind@ Yes confirming it has not been giving issues in canary. I'll merge today pending your approval.
,
Nov 28 2017
Approving merge to M63 branch 3239 based on comments #44, #48 and #52. Please merge ASAP. Thank you.
,
Nov 28 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/34b7aa1d2a17995423a3e4a68215a3ce6ab9008c commit 34b7aa1d2a17995423a3e4a68215a3ce6ab9008c Author: Nathan Bruer <allada@chromium.org> Date: Tue Nov 28 19:26:04 2017 Allows renderer gets post data if devtools attached w/ plzNavigate Adds a temporary fix to allow devtools to get the post data of navigation request if the renderer has devtools attached. TBR=pfeldman,govind R=caseq,clamy BUG= 766715 (cherry picked from commit 649cc21a76d5e7e6dba6077b8013d3b8bf4ccc2d) Cq-Include-Trybots: master.tryserver.chromium.linux:linux_site_isolation Change-Id: I875ea05e778fc431b6073a6fb6b1c393bcc904be Reviewed-on: https://chromium-review.googlesource.com/770490 Reviewed-by: Charlie Reis <creis@chromium.org> Commit-Queue: Blaise Bruer <allada@google.com> Commit-Queue: Blaise Bruer <allada@chromium.org> Cr-Original-Commit-Position: refs/heads/master@{#518109} Reviewed-on: https://chromium-review.googlesource.com/794342 Reviewed-by: Blaise Bruer <allada@chromium.org> Cr-Commit-Position: refs/branch-heads/3239@{#585} Cr-Branched-From: adb61db19020ed8ecee5e91b1a0ea4c924ae2988-refs/heads/master@{#508578} [modify] https://crrev.com/34b7aa1d2a17995423a3e4a68215a3ce6ab9008c/content/browser/frame_host/navigation_request.cc [add] https://crrev.com/34b7aa1d2a17995423a3e4a68215a3ce6ab9008c/third_party/WebKit/LayoutTests/http/tests/inspector-protocol/network/redirect-302-with-post-expected.txt [add] https://crrev.com/34b7aa1d2a17995423a3e4a68215a3ce6ab9008c/third_party/WebKit/LayoutTests/http/tests/inspector-protocol/network/redirect-302-with-post.js
,
Nov 28 2017
Everything should have landed and be present on stable soon. Thanks! Closing bug.
,
Jul 13
Hi, I'm on Mac OSX and using Google Chrome Version 67.0.3396.99 (Official Build) (64-bit) and this bug still seems to exists.
,
Jul 16
@atif089: please file a new bug including reproduction steps and a net-log (http://dev.chromium.org/for-testers/providing-network-details) and referencing this bug. |
||||||||||||||
►
Sign in to add a comment |
||||||||||||||
Comment 1 by ligim...@chromium.org
, Sep 19 2017