Open Bug 1924711 Opened 1 year ago Updated 25 days ago

Intermittent Wd | <test-name> - assert False (is_maximized)

Categories

(Remote Protocol :: Marionette, defect, P5)

Product:
Remote Protocol
Component:
Marionette
Type:
defect

Tracking

(firefox-esr115 unaffected, firefox-esr128 unaffected, firefox131 unaffected, firefox132 unaffected, firefox133 affected, firefox134 affected)

Tracking Flags:
Tracking Status
firefox-esr115 --- unaffected
firefox-esr128 --- unaffected
firefox131 --- unaffected
firefox132 --- unaffected
firefox133 --- affected
firefox134 --- affected

People

(Reporter: intermittent-bug-filer, Unassigned)

References

(Blocks 1 open bug, Regression)

Details

(Keywords: intermittent-failure, regression)

This is a new failure and started around the time when we landed bug 1857571. The failure is:

[task 2024-10-15T07:37:28.711Z] 07:37:28     INFO -         session.window.maximize()
[task 2024-10-15T07:37:28.711Z] 07:37:28     INFO - >       assert is_maximized(session)
[task 2024-10-15T07:37:28.711Z] 07:37:28     INFO - E       assert False
[task 2024-10-15T07:37:28.711Z] 07:37:28     INFO - E        +  where False = is_maximized(<Session ad94b278-b9b1-4bdf-bad2-6ab54af516d7>)

Sasha, could this be a regression?

Flags: needinfo?(aborovova)

I don't really see how it could be related, since nothing has changed around maximizing the window. It's also a different failure compare to what I've observed on Wayland.
In this case, the maximize command seems to set the right size to the window (see the return result of the command), but when we evaluate the script to get the window dimensions, they equal initial values.

[task 2024-10-15T07:37:21.222Z] 07:37:21     INFO - PID 5569 | 1728977841201	webdriver::server	DEBUG	-> POST /session/ad94b278-b9b1-4bdf-bad2-6ab54af516d7/window/maximize {}
[task 2024-10-15T07:37:21.222Z] 07:37:21     INFO - PID 5569 | 1728977841201	Marionette	DEBUG	0 -> [0,720,"WebDriver:MaximizeWindow",{}]
[task 2024-10-15T07:37:22.351Z] 07:37:22     INFO - PID 5569 | 1728977842350	Marionette	DEBUG	0 <- [1,720,null,{"x":0,"y":23,"width":1280,"height":1001}]
[task 2024-10-15T07:37:22.352Z] 07:37:22     INFO - PID 5569 | 1728977842351	webdriver::server	DEBUG	<- 200 OK {"value":{"x":0,"y":23,"width":1280,"height":1001}}
[task 2024-10-15T07:37:22.353Z] 07:37:22     INFO - PID 5569 | 1728977842352	webdriver::server	DEBUG	-> POST /session/ad94b278-b9b1-4bdf-bad2-6ab54af516d7/execute/sync {"script": "\n        return {\n            availWidth: sc ...             windowWidth: window.outerWidth,\n            windowHeight: window.outerHeight,\n        }\n        ", "args": []}
[task 2024-10-15T07:37:22.355Z] 07:37:22     INFO - PID 5569 | 1728977842354	Marionette	DEBUG	0 -> [0,721,"WebDriver:ExecuteScript",{"args":[],"script":"\n        return {\n            availWidth: screen.availWidth,\n            availHeight: screen.availHeight,\n            windowWidth: window.outerWidth,\n            windowHeight: window.outerHeight,\n        }\n        "}]
[task 2024-10-15T07:37:22.356Z] 07:37:22     INFO - PID 5569 | 1728977842355	Marionette	DEBUG	0 <- [1,721,null,{"value":{"availWidth":1280,"availHeight":1001,"windowWidth":800,"windowHeight":600}}]
[task 2024-10-15T07:37:22.357Z] 07:37:22     INFO - PID 5569 | 1728977842355	webdriver::server	DEBUG	<- 200 OK {"value":{"availHeight":1001,"availWidth":1280,"windowHeight":600,"windowWidth":800}}

On Wayland, the return result of the maximize command is already wrong:

[task 2024-10-07T17:07:18.303Z] 17:07:18     INFO - PID 61406 | 1728320838301	Marionette	DEBUG	0 -> [0,720,"WebDriver:MaximizeWindow",{}]
[task 2024-10-07T17:07:18.555Z] 17:07:18     INFO - PID 61406 | 1728320838554	Marionette	DEBUG	0 <- [1,720,null,{"x":0,"y":0,"width":1850,"height":1053}]
[task 2024-10-07T17:07:18.556Z] 17:07:18     INFO - PID 61406 | 1728320838555	webdriver::server	DEBUG	<- 200 OK {"value":{"x":0,"y":0,"width":1850,"height":1053}}
[task 2024-10-07T17:07:18.556Z] 17:07:18     INFO - PID 61406 | 1728320838556	webdriver::server	DEBUG	-> POST /session/b7b085c6-cd07-468a-98d1-01f4fe0a9e4a/execute/sync {"script": "\n        return {\n            availWidth: sc ...             windowWidth: window.outerWidth,\n            windowHeight: window.outerHeight,\n        }\n        ", "args": []}
[task 2024-10-07T17:07:18.557Z] 17:07:18     INFO - PID 61406 | 1728320838556	Marionette	DEBUG	0 -> [0,721,"WebDriver:ExecuteScript",{"args":[],"script":"\n        return {\n            availWidth: screen.availWidth,\n            availHeight: screen.availHeight,\n            windowWidth: window.outerWidth,\n            windowHeight: window.outerHeight,\n        }\n        "}]
[task 2024-10-07T17:07:18.558Z] 17:07:18     INFO - PID 61406 | 1728320838558	Marionette	DEBUG	0 <- [1,721,null,{"value":{"availWidth":1920,"availHeight":1080,"windowWidth":1850,"windowHeight":1053}}]
[task 2024-10-07T17:07:18.559Z] 17:07:18     INFO - PID 61406 | 1728320838558	webdriver::server	DEBUG	<- 200 OK {"value":{"availHeight":1080,"availWidth":1920,"windowHeight":1053,"windowWidth":1850}}
Flags: needinfo?(aborovova)

Maybe it was triggered because of the changes to set.py? Especially moving out the test test_restore_from_minimized to its own file? We never saw this failure before and it started at the same time. So it might be very likely. It happened twice on MacOS so far.

(In reply to Henrik Skupin [:whimboo][⌚️UTC+2] from comment #3)

Maybe it was triggered because of the changes to set.py? Especially moving out the test test_restore_from_minimized to its own file? We never saw this failure before and it started at the same time. So it might be very likely. It happened twice on MacOS so far.

Yes, that could be, since the minimized test was just before the failing one.

It looks like this failure stopped at the end of the day on October 16th. So maybe there was a regression which got fixed. Lets observer for another week and if no further failures happen we can close this bug.

Set release status flags based on info from the regressing bug 1857571

status-firefox134: --- → affected
Summary: Intermittent /webdriver/tests/classic/set_window_rect/set.py | test_restore_from_maximized - assert False (is_maximized) → Intermittent Wd | <test-name> - assert False (is_maximized)
You need to log in before you can comment on or make changes to this bug.