Thanks for filing the issue!

@Reporter: Could you please provide sample Test file/URL and sample credentials that reproduces the issue which help in further triaging the issue in better way.

Thanks!

@viswa.karala@chromium.org: Here is the URL and login credentials you requested...

URL:
https://agentone-qap5.my.salesforce.com

UID||PWD:
admin@agentone-qa.person5.com || Vancouv3rAg3nt! 

Please note that you will be asked for a verification code in order to login but if you provide your IP address I can whitelist it so that you will be able to login without having to provide the verification code.

Otherwise you will need to email me when prompted for the verification code and I can pass it to you.

Steps to repro in the above URL:
1. Go to Clients tab and open the record "Peter Parker"
2. Click the blue "Start new Insurance Case" button
3. Select the 'Begin Application' radio button and in the "What type of product?" question select "Term" then click the Start button
4. When you see the Case Information tab select any state and choose "Term" under Product Type then click Find Available Product
(You should now see the list of products and it is truncated)
5. Sort by Carrier and click the Select button under the iGO e-App column
6. After the screen refreshes to show the form... try scrolling down and back up then click the drop down for "Marital Status"
7. Finally try clicking the "State" drop down (since it is near the truncation the drop down does not work)

Please let me know if you have any questions or if you need the verification code... I am available any time before 5PM Pacific.

Thanks,
Pablo

Thank you for providing more feedback. Adding the requester to the cc list.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

@viswa.karala@chromium.org

Just wondering if there has been any traction on this? I provided the steps and URL to test this but you will need a verification code unless we whitelist your IP address.

Thanks for providing the Test URL and sample credentials!

Tried testing the issue by logging into the URL: https://agentone-qap5.my.salesforce.com with the credentials provided in comment# 2, it is asking for the verification code to enter to login into the site.
#2: We are not sure that we can provide our IP address as per our organisation policy, so it would be better if you provide verification code for triaging the issue.

@Reporter: We people are working in the timezone IST 09:00 to 18:30, so, it will be most helpful for us if you are available in this time to provide the verification which helps us in further triaging the issue in better way. Could you please confirm on which OS you are facing the issue.

Note: I have dropped you the email on regarding the verification code.

Thanks!

@viswa.karala@chromium.org

I have responded on your email regarding the verification code... this is time sensitive so we will need to co-ordinate. We will need to both be online at the same time so I can provide the verification code immediately after you try to login.

Regarding my OS... I am running the latest Chrome beta Version 69.0.3497.42 (Official Build) beta (64-bit) on Windows 10.

Thank you for providing more feedback. Adding the requester to the cc list.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Unable to reproduce the issue on chrome reported version# 69.0.3497.32, latest Beta# 69.0.3497.72 and Stable# 68.0.3440.106 using Mac 10.12.6 and Windows-10 with URL and sample credentials provided in comment# 2. As the issue is Unable to reproduce from TE end, hence adding TE-NeedsTriageHelp label to it and requesting someone from the Dev team to provide further inputs on this issue.
Note: Tentatively adding Blink>Forms>Select component to it, attached screenshots for reference.

Thanks!

Deleted: 873337-1.png 329 KB

	873337-1.png 329 KB View Download

Deleted: 873337-2.png 264 KB

	873337-2.png 264 KB View Download

Reporter, does your machine have non-100% scaling setting (OS-level scaling, not the zoom feature in Chrome)?
Is this reproducing if chrome://flags/#site-isolation-trial-opt-out is "Opt-out"?

tkent@chromium.org

Looks like that setting was the culprit it was set to "Default" and after changing it to "Opt-out" everything is working.

   chrome://flags/#site-isolation-trial-opt-out is "Opt-out"

Two questions:
1. the setting says opt-out is not recommended so what are repercussions of making this change
2. if our customers experience this kind of behavior is it safe and recommended that we tell them to change the setting to "opt-out"

Thank you for providing more feedback. Adding the requester to the cc list.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Comment 10, thank you for the check!
Changing chrome://flags/#site-isolation-trial-opt-out flag is just for investigation of the culprit of this bug. IMO, we may set it "Opt-out" until this bug is fixed.

kenrb@, is this a dupe of  Issue 878703 ?

This issue is marked as a release blocker with no milestone associated. Please add an appropriate milestone.

All release blocking issues should have milestones associated to it, so that the issue can tracked and the fixes can be pushed promptly.

This issue is marked as a release blocker with no OS labels associated. Please add an appropriate OS label.

All release blocking issues should have OS labels associated to it, so that the issue can tracked and promptly verified, once it gets fixed.

Thanks for your time! To disable nags, add the Disable-Nags label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

#12: The mispositioned select menu is very likely to be the same issue but I don't know for certain if the page truncation has the same cause.

We have sent an email to reporter for requesting the verification code in order to reproduce the issue with site isolation flag(chrome://flags/#site-isolation-trial-opt-out is "Opt-out") to check where this has been regressed.

Thanks!

This issue is marked as a release blocker with no milestone associated. Please add an appropriate milestone.

All release blocking issues should have milestones associated to it, so that the issue can tracked and the fixes can be pushed promptly.

This issue is marked as a release blocker with no OS labels associated. Please add an appropriate OS label.

All release blocking issues should have OS labels associated to it, so that the issue can tracked and promptly verified, once it gets fixed.

Thanks for your time! To disable nags, add the Disable-Nags label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Just to update, worked with the reporter for logging into the site to test and confirm the issue in order to find out where it got regressed, but some how unable to reproduce the issue on chrome reported version# 69.0.3497.32 and latest Beta# 69.0.3497.81 using Mac 10.12.6 with URL and sample credentials provided in comment# 2 by making "#site-isolation-trial-opt-out" flag to "Default" and "opt-out".

"#site-isolation-trial-opt-out" flag was introduced from M-66, so tested the issue on chrome version# 66.0.3359.181, but unable to reproduce the issue there also by making "#site-isolation-trial-opt-out" flag to "Default" and "opt-out".
Note: Adding M-70 label to it as per Comment# 13 Target-70.

Thanks!

@viswa.karala@chromium.org

I'm not sure if this is OS related but when I change the setting back to "#site-isolation-trial-opt-out" flag to "Default" the issue re-appears for me. I am also on Chrome beta 69.0.3497.81 but running on Windows 10.

As stated if I set it to "Opt-out" I am no longer getting this issue.

Please let me know if you need to troubleshoot together...

Hi Viswa,

I’ll add the following to a conversation on the  issue 873337  as well but
wanted to make sure this is still being looked at.

We have 40,000 customers about to go live with this release and I wanted to
make sure this issue is not forgotten/ignored.

I noticed you mentioned you are unable to repro this issue but when I reset
the setting “site isolation trial opt out” back to default the issue
reappears.

Can you please let me know what are next steps to assure this issue is
resolved before our customers go live?

Thanks,
Pablo

kenrb@chromium.org
tkent@chromium.org
viswa.karala@chromium.org

Hi,

Just wanted to know the status of this issue, as mentioned when the setting is set back to "default" the issue continues to be reproducible.

We have 40,000 customers soon to go live and most of them are running on Chrome. We cannot expect them to change this setting in order to avoid this issue.

Please let me know if there is anything else I can do to help get some traction on this issue.

Regards,
Pablo

I am actively working on this, unfortunately I don't have an ETA for a fix yet. I will keep this updated.

p.salvia@ if possible can you please provide reduced testcase, Since requesting a verification code might not be ideal case when bisecting a bug.

kenrb@: Are you able to repro this, based on your comment 22?  Sounds like there's two things we need to fix, between the context menu problem ( issue 878703 ?) and the page truncation.  We should focus on getting fixes into trunk for this ASAP, which will help us understand how far it can be merged.  Also, let us know if you spot anything that can be used as a workaround on the page's side, since recommending that users disable Site Isolation is not advised at this point.

There's an M69 respin going out today, which shouldn't be held for this because we don't have a fix available.  Instead, let's see if we can get the fix into at least M70 and possibly a later M69 respin, assuming there is one.

p.salvia@: Can I please get help reproducing this problem? I have been working with repro cases from elsewhere to investigate the wrongly positioned pop-ups, but I'd like to see better what is happening with page truncation.

Alternatively, can you test this on Dev and Canary? The truncation problem looks like other bugs we have seen, and if it is the same then a fix is already on the M69 branch. That didn't make it to the initial M69 launch but I believe a respin will be going out very soon that includes it.

kenrb@chromium.org
pbomm...@chromium.org

Unfortunately this issue is only reproducible from Salesforce so reducing the steps to repro or eliminating the need for the verification code is rather difficult.

If you provide your IP addresses I can whitelist them on the Salesforce org so you will not need to provide a verification code. If this is not an option... I am very happy to test on whatever build you provide to see if the issue is resolved.

 p.salvia@ below are few details about how to bisect Chrome to find the range which helps us a lot in identifying the Culprit Cl. 

Bisect to find culprit CL(Please follow instructions) :
https://www.chromium.org/developers/bisect-builds-py

Other Chrome channels to test if the issue got fixed on Dev or canary which can be downloaded by below URL's :

Dev : https://www.google.com/chrome/dev/
Canary : https://www.google.com/chrome/canary/

p.salvia@: Have you been able to repro this bug on Canary or Dev? As I mentioned I am particularly interested in whether the clipping problem has been fixed on those channels.

@kenrb@chromium.org

Note I have just tried on Canary and was not able to reproduce the truncation issue BUT I was able to reproduce the dropdown boxes position issue.

I will try DEV and add another comment in a few minutes.

@kenrb@chromium.org

I have the same result on DEV the truncation issue is NOT reproducible but the misplaced dropdown boxes continues to be an issue.

@kenrb@chromium.org

I am prepared to test on the bisect builds but have a question...
What should I be putting in the command line as the good and bad revision?

I'm sorry if I missed this in the conversation somewhere.

Thanks,
Pablo

Hi Ken,

Please see added comments 31, 32, and 33 on  issue 873337 .

I tried Dev and Canary and could repro the misplaced dropdowns but not the
truncation issue.

I also had a question regarding bisect builds...

Regards,
Pablo

Thanks, that is a good result. I am working on a fix to the mispositioned select menu popups.

FWIW both of these issues are regressions associated with having nested cross-origin iframes (i.e. an iframe within an iframe). The truncation issue regressed in June and I fixed it in July but only recently merged the fix to M69. I believe we started pushing that out to Stable channel users yesterday, so anybody who has updated to Chrome version 69.0.3497.92 should not see it. Please let us know if that appears not to be the case.

The select element problem regressed earlier this year but unfortunately went unnoticed until last month. Even if I land the fix today it might not be available to users until M70 ships in October. The popup menu being in the wrong position is a temporary problem in a nested iframe after the outer window is scrolled. It persists until something prompts the new position to propagate to the nested iframe -- in my repro cases dismissing the popup and then clicking the element again corrects it. There might be workarounds possible for this bug, such as doing a one pixel resize of the iframe after the main window scrolls. Toggling visibility of the iframe on and off would also work but might be visually unappealing to users.

Thanks Ken.

I have confirmed the truncation issue does not happen in the latest beta
build OR the new stable build.

So only outstanding issue is the misplaced drop downs as you mentioned.

That's great!  Ken has a CL in review that should fix the drop down menu placement at https://chromium-review.googlesource.com/c/chromium/src/+/1225090.

That one is a small one-line fix, and I've verified that it applies cleanly and fixes the bug (using the repro steps in  issue 878703 ) on both the M70 and M69 branches in case we merge it.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/ee24be2aa0c578cf371f64dea1b887635959c833

commit ee24be2aa0c578cf371f64dea1b887635959c833
Author: Ken Buchanan <kenrb@chromium.org>
Date: Thu Sep 13 20:29:27 2018

Propagate window positions down OOPIF chains

A regression from earlier this year affected visual state updates when
they are being propagated to OOPIFs, such that they don't immediately
also propagate to nested OOPIFs underneath a direct child frame. This
was causing temporary mispositioning of things like popup menus.

This CL corrects the flaw.

It also fixes a bug in a test helper class, ShowWidgetMessageFilter,
which incorrectly discards information from messages that were called
before Wait() was called.

Bug:  878703 ,  873337 
Change-Id: I61f42eba4f6b7ab41d96a1db479bceb64ef00954
Reviewed-on: https://chromium-review.googlesource.com/1225090
Reviewed-by: Charlie Reis <creis@chromium.org>
Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
Commit-Queue: Ken Buchanan <kenrb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#591134}
[modify] https://crrev.com/ee24be2aa0c578cf371f64dea1b887635959c833/content/browser/frame_host/cross_process_frame_connector.cc
[modify] https://crrev.com/ee24be2aa0c578cf371f64dea1b887635959c833/content/browser/site_per_process_hit_test_browsertest.cc
[modify] https://crrev.com/ee24be2aa0c578cf371f64dea1b887635959c833/content/test/content_browser_test_utils_internal.cc
[modify] https://crrev.com/ee24be2aa0c578cf371f64dea1b887635959c833/content/test/content_browser_test_utils_internal.h

This has been discussed offline, but it is a low risk change (one line of code, correcting a clear error). Requesting merge.

This bug requires manual review: M70 has already been promoted to the beta branch, so this requires manual review
Please contact the milestone owner if you have questions.
Owners: benmason@(Android), kariahda@(iOS), geohsu@(ChromeOS), abdulsyed@(Desktop)

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/fb11c04111341c75889944ea8e26b8af9c9a97d3

commit fb11c04111341c75889944ea8e26b8af9c9a97d3
Author: Ken Buchanan <kenrb@chromium.org>
Date: Thu Sep 13 21:01:58 2018

Propagate window positions down OOPIF chains

A regression from earlier this year affected visual state updates when
they are being propagated to OOPIFs, such that they don't immediately
also propagate to nested OOPIFs underneath a direct child frame. This
was causing temporary mispositioning of things like popup menus.

This CL corrects the flaw.

It also fixes a bug in a test helper class, ShowWidgetMessageFilter,
which incorrectly discards information from messages that were called
before Wait() was called.

Bug:  878703 ,  873337 
Change-Id: I61f42eba4f6b7ab41d96a1db479bceb64ef00954
Reviewed-on: https://chromium-review.googlesource.com/1225090
Reviewed-by: Charlie Reis <creis@chromium.org>
Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
Commit-Queue: Ken Buchanan <kenrb@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#591134}(cherry picked from commit ee24be2aa0c578cf371f64dea1b887635959c833)
Reviewed-on: https://chromium-review.googlesource.com/1225602
Reviewed-by: Ken Buchanan <kenrb@chromium.org>
Cr-Commit-Position: refs/branch-heads/3551@{#4}
Cr-Branched-From: 08c03be7fdeabd63de8e4efca4c562a2f5a70ba9-refs/heads/master@{#590850}
[modify] https://crrev.com/fb11c04111341c75889944ea8e26b8af9c9a97d3/content/browser/frame_host/cross_process_frame_connector.cc
[modify] https://crrev.com/fb11c04111341c75889944ea8e26b8af9c9a97d3/content/browser/site_per_process_hit_test_browsertest.cc
[modify] https://crrev.com/fb11c04111341c75889944ea8e26b8af9c9a97d3/content/test/content_browser_test_utils_internal.cc
[modify] https://crrev.com/fb11c04111341c75889944ea8e26b8af9c9a97d3/content/test/content_browser_test_utils_internal.h

Note: the merge in comment 41 was to the M71 branch to cover tomorrow's Canary and Dev channel releases.  The M70 and M69 merge requests can be reviewed tomorrow after we confirm the fix is working as expected on M71.

pls update bug with canary result tomorrow morning on version 71.0.3551.3 or higher.

The NextAction date has arrived: 2018-09-14

As per comment# 8 and 18, unable to reproduce the issue on chrome reported version from TE end and as per comment# 37(issue:  878703 , tried reproducing the issue with URL provided in comment# 0), but unable to reproduce the issue. 

Ken Buchanan@ Could you please try to test it on latest chrome#71.0.3552.2 and help us in verifying the fix.

Thanks!

Comment 45: Please use the repro steps in  issue 878703  to verify that fix, which are easier to repro.  Thanks!

p.salvia@: Also, can you verify that the drop down box problem in Salesforce is resolved in today's 71.0.3552.2 Chrome Canary?  Thanks!

The test case on  issue 878703  doesn't appear to be available any longer and I am having trouble putting together a test case that repros on Stable. Confirmation of the fix in Canary would be appreciated.

nasko@ put together another repro case for the select element bug: http://web.evilbit.io/frame_sizing.html

I see the bug reproducing on that page on Windows Stable but everything behaves correctly on Windows Canary.

I've double checked the crash dashboard and I don't see any new crashes introduced related to this in 71.0.3552.2 (not that it was likely, but good to be safe).  Given Ken's verification in comment 48, I think it's worth proceeding with the merges to M69 and M70.

(p.salvia@, still let us know if you're able to confirm the fix in Salesforce in Canary, just to make sure we're on the right track.  We won't block on that, but it will be good to know.)

Approving merge to M69 branch 3497 and M70 branch 3538 based on comment #49 and per offline chat with creis@. Please merge ASAP and mark bug as fixed. Thank you.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/e8544c2740b0d78c5fad7637473fecb39b39dfb1

commit e8544c2740b0d78c5fad7637473fecb39b39dfb1
Author: Ken Buchanan <kenrb@chromium.org>
Date: Fri Sep 14 17:41:15 2018

Propagate window positions down OOPIF chains

A regression from earlier this year affected visual state updates when
they are being propagated to OOPIFs, such that they don't immediately
also propagate to nested OOPIFs underneath a direct child frame. This
was causing temporary mispositioning of things like popup menus.

This CL corrects the flaw.

It also fixes a bug in a test helper class, ShowWidgetMessageFilter,
which incorrectly discards information from messages that were called
before Wait() was called.

Bug:  878703 ,  873337 
Change-Id: I61f42eba4f6b7ab41d96a1db479bceb64ef00954
Reviewed-on: https://chromium-review.googlesource.com/1225090
Reviewed-by: Charlie Reis <creis@chromium.org>
Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
Commit-Queue: Ken Buchanan <kenrb@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#591134}(cherry picked from commit ee24be2aa0c578cf371f64dea1b887635959c833)
Reviewed-on: https://chromium-review.googlesource.com/1226027
Reviewed-by: Ken Buchanan <kenrb@chromium.org>
Cr-Commit-Position: refs/branch-heads/3538@{#406}
Cr-Branched-From: 79f7c91a2b2a2932cd447fa6f865cb6662fa8fa6-refs/heads/master@{#587811}
[modify] https://crrev.com/e8544c2740b0d78c5fad7637473fecb39b39dfb1/content/browser/frame_host/cross_process_frame_connector.cc
[modify] https://crrev.com/e8544c2740b0d78c5fad7637473fecb39b39dfb1/content/browser/site_per_process_hit_test_browsertest.cc
[modify] https://crrev.com/e8544c2740b0d78c5fad7637473fecb39b39dfb1/content/test/content_browser_test_utils_internal.cc
[modify] https://crrev.com/e8544c2740b0d78c5fad7637473fecb39b39dfb1/content/test/content_browser_test_utils_internal.h

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/47b99a368e694ae38346630ce4c621622de4b8b0

commit 47b99a368e694ae38346630ce4c621622de4b8b0
Author: Ken Buchanan <kenrb@chromium.org>
Date: Fri Sep 14 17:40:27 2018

Propagate window positions down OOPIF chains

A regression from earlier this year affected visual state updates when
they are being propagated to OOPIFs, such that they don't immediately
also propagate to nested OOPIFs underneath a direct child frame. This
was causing temporary mispositioning of things like popup menus.

This CL corrects the flaw.

It also fixes a bug in a test helper class, ShowWidgetMessageFilter,
which incorrectly discards information from messages that were called
before Wait() was called.

Bug:  878703 ,  873337 
Change-Id: I61f42eba4f6b7ab41d96a1db479bceb64ef00954
Reviewed-on: https://chromium-review.googlesource.com/1225090
Reviewed-by: Charlie Reis <creis@chromium.org>
Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
Commit-Queue: Ken Buchanan <kenrb@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#591134}(cherry picked from commit ee24be2aa0c578cf371f64dea1b887635959c833)
Reviewed-on: https://chromium-review.googlesource.com/1226026
Reviewed-by: Ken Buchanan <kenrb@chromium.org>
Cr-Commit-Position: refs/branch-heads/3497@{#937}
Cr-Branched-From: 271eaf50594eb818c9295dc78d364aea18c82ea8-refs/heads/master@{#576753}
[modify] https://crrev.com/47b99a368e694ae38346630ce4c621622de4b8b0/content/browser/frame_host/cross_process_frame_connector.cc
[modify] https://crrev.com/47b99a368e694ae38346630ce4c621622de4b8b0/content/browser/site_per_process_hit_test_browsertest.cc
[modify] https://crrev.com/47b99a368e694ae38346630ce4c621622de4b8b0/content/test/content_browser_test_utils_internal.cc
[modify] https://crrev.com/47b99a368e694ae38346630ce4c621622de4b8b0/content/test/content_browser_test_utils_internal.h

kenrb@chromium.org

Apologies about the late reply...

I just tested on Chrome Canary:
- Version 71.0.3561.0 (Official Build) canary (64-bit)

The issue with the misplaced drop down boxes could not be reproduced.