New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 725211 link

Starred by 6 users

Issue metadata

Status: Duplicate
Merged: issue 714229
Owner:
Last visit > 30 days ago
Closed: Jun 2017
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux , Windows , Mac
Pri: 1
Type: Bug-Regression



Sign in to add a comment

WebSocket viewer's focus taken from non-default focus - can't see values.

Reported by lmfin...@gmail.com, May 22 2017

Issue description

UserAgent: Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36

Steps to reproduce the problem:
1. Open a website with two or more open WebSocket connections (unfortunately, this is happening on an internal website, so I can't point you to it).
2. Open Dev Tools
3. Click on the Network tab
4. Click on WS to see the WebSocket data
5. Click on a socket other than the first one.

What is the expected behavior?
I expect to see the Frames of the socket I selected, and I expect the view of the frames to be stable, allowing me to see the frames come in over time.

What went wrong?
The Frames view I want opens briefly, but then the view switches to the first socet's frames almost immediately. I haven't found a way to stop the browser from stealing focus away from the view I requested.

Did this work before? Yes 57?

Chrome version: 58.0.3029.110  Channel: stable
OS Version: 6.1 (Windows 7, Windows Server 2008 R2)
Flash Version: 

I don't know what previous version I had, but it used to work.
 

Comment 1 by hdodda@chromium.org, May 23 2017

Cc: hdodda@chromium.org
Labels: Needs-Feedback
@lmfinney-- Could you please provide us the sample test url to reproduce the issue and also help us by providing the screencast for the issue , that would help us in triaging the issue better.

Thanks!

Comment 2 by lmfin...@gmail.com, May 23 2017

As I said, I can't give you the url that I'm working on because it's an internal website, but I can partially duplicate the problem using https://www.websocket.org/echo.html.

1. Navigate to https://www.websocket.org/echo.html
2. Open Dev Tools
3. Click on the Network tab
4. Click on WS to see the WebSocket data (I see one socket here)
5. Click on the "Connect" button on the website (I now see two sockets)
5. Click on the second socket in the list (for me, the name is "?encoding=text")
6. Click on the "Send" button on the website.

At this point, the websocket view jumps to the first socket in the list, away from "?encoding=text", and I have to click again on "?encoding=text" to see what I just sent.

In this app, it's not very annoying, but in our app the socket has a sub-second heartbeat, so the focus is pulled away almost immediately after I click on it, so I don't get to see the socket communication.

I will record a screencast with our actual app and post it later today.
Project Member

Comment 3 by sheriffbot@chromium.org, May 23 2017

Labels: -Needs-Feedback
Thank you for providing more feedback. Adding requester "hdodda@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

Comment 4 by lmfin...@gmail.com, May 23 2017

Unfortunately, I can't install the screencasting software I know (https://screencast-o-matic.com/screen_recorder) on my machine at work, so I can't give you an original screencast. I hope the https://www.websocket.org/echo.html example is sufficient.
Yo, confirming this bug in Chrome 58.0.3029.110 (64-bit) on Win10 and here is the screencast of problem so you can better imagine - https://www.youtube.com/watch?v=BOg8aTRt7rM

I think the problem is that websocket window changes focus to websocket with recent communication. Here I have two websockets each with keep alive messages and the focus is just jumping between them.
Components: Platform>DevTools>Network
Labels: -Pri-2 hasbisect-per-revision ReleaseBlock-Stable M-60 OS-Linux OS-Mac Pri-1
Owner: allada@chromium.org
Status: Assigned (was: Unconfirmed)
Able to reproduce the issue on Windows 10, Mac 10.12.4 and Ubuntu 14.04 using reported version #58.0.3029.110 but the same is not reproducible in the latest canary #60.0.3111.0.

Reverse Bisect Information:
=====================
Good build: 60.0.3085.0  Revision(468227)
Bad Build : 60.0.3084.0  Revision(468182)

Change Log URL: 
https://chromium.googlesource.com/chromium/src/+log/0a54e8804a182388f0aedc8a18524eaa10e76b85..fc4b4d8fbe2657a7d51cef8939e68acc99fae05a

From the above change log possible CL that fixed this issue:
Review-Url: https://codereview.chromium.org/2838673003

allada@ - Could you please check and merge the fix to M60 if it is a valid candidate.

Note: Adding label ReleaseBlock-Stable, as it seems to be a recent regression.

Thanks...!!

Comment 7 by y...@zendesk.com, Jun 1 2017

For the time being, a workaround is to use the filter field above the list of websockets such that the websocket you are interested in is the only websocket listed (or at least the offending websocket with a heartbeat is excluded from the list).

Credits to Tarek El-Sibay in https://groups.google.com/forum/#!topic/google-chrome-developer-tools/sJn5V3yfarU
Your bug is tagged as Release block, please have a fix ASAP.We will take only critical merges when close to Stable.
Cc: krajshree@chromium.org
Labels: Needs-Feedback
Unable to reproduce the issue on Win-10 using latest stable version #59.0.3071.86 and latest canary #61.0.3131.0.

lmfinney@ - Could you please upgrade chrome to latest stable version #59.0.3071.86 or latest canary #61.0.3131.0 and check the issue. 
Please let us know if the issue persist or not.

Thanks...!!
I can confirm that its working now with #59.0.3071.86 on Win10, good job guys! 
As per comment #10, it seems the issue has been fixed in latest stable #59.0.3071.86.
allada@ - Could you please update the status.

Thanks...!!
Status: Fixed (was: Assigned)
Yes this was fixed in a previous patch. I just forgot to merge this bug with the origin bug.
Labels: Merge-TBD
[Auto-generated comment by a script] We noticed that this issue is targeted for M-60; it appears the fix may have landed after branch point, meaning a merge might be required. Please confirm if a merge is required here - if so add Merge-Request-60 label, otherwise remove Merge-TBD label. Thanks.
Has this bug been merged to M60? Branch:3112
Mergedinto: 714229
Status: Duplicate (was: Fixed)
This bug should have been fixed in 59 and 60, yes.
Labels: -Merge-TBD -ReleaseBlock-Stable

Sign in to add a comment