8000 Forms IT tests breaks with Chrome 136.0.7103.59 · Issue #39540 · keycloak/keycloak · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Forms IT tests breaks with Chrome 136.0.7103.59 #39540

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
2 tasks done
ahus1 opened this issue May 7, 2025 · 6 comments · Fixed by #39565
Closed
2 tasks done

Forms IT tests breaks with Chrome 136.0.7103.59 #39540

ahus1 opened this issue May 7, 2025 · 6 comments · Fixed by #39565
Assignees
@rmartinc
Labels
area/ci Indicates an issue on the CI kind/bug Categorizes a PR related to a bug priority/important Must be worked on very soon release/26.2.5 release/26.3.0 team/continuous-testing team/core-clients

Comments

@ahus1
Copy link
Contributor
ahus1 commented May 7, 2025

Before reporting an issue

  • I have read and understood the above terms for submitting issues, and I understand that my issue may be closed without action if I do not follow them.

Area

No response

Describe the bug

When running with the latest version of chrome 136.0.7103.59, tests like BrowserButtonsTest.appInitiatedRegistrationWithBackButton fail on chrome

Version

main

Regression

  • The issue is a regression

Expected behavior

Tests should pass

Actual behavior

Tests fail, see https://github.com/keycloak/keycloak/actions/runs/14880473986/job/41808705885?pr=39522 and https://github.com/keycloak/keycloak/actions/runs/14885132071/job/41803271126

How to Reproduce?

See CI. Also tested locally.

Anything else?

No response
@ahus1 ahus1 self-assigned this May 7, 2025
@ahus1 ahus1 added kind/bug Categorizes a PR related to a bug status/triage labels May 7, 2025
@ahus1 ahus1 changed the title Forms IT tests break with Chrome 136.0.7103.59 Forms IT tests breaks with Chrome 136.0.7103.59 May 7, 2025
ahus1 added a commit to pruivo/keycloak that referenced this issue May 7, 2025
@ahus1
Downgrade Chrome to unblock CI
b0e6390 
Workaround for keycloak#39540

Signed-off-by: Alexander Schwartz <aschwart@redhat.com>
ahus1 added a commit that referenced this issue May 7, 2025
@ahus1
Downgrade Chrome to unblock CI
a097c5a 
Workaround for #39540

Signed-off-by: Alexander Schwartz <aschwart@redhat.com>
ahus1 added a commit to pruivo/keycloak that referenced this issue May 7, 2025
@ahus1
Downgrade Chrome to unblock CI
0586262 
Workaround for keycloak#39540

Signed-off-by: Alexander Schwartz <aschwart@redhat.com>
@ahus1 ahus1 added team/core-clients area/ci Indicates an issue on the CI labels May 7, 2025
@ahus1
Copy link
Contributor Author
ahus1 commented May 7, 2025

@mposolda / @rmartinc - when testing this locally it was consistently failing for Chrome 136.0.7103.59 locally, so I downgraded to 135.0.7049.114 and then the test started to work again.

The workaround is now part of a097c5a which should now log the version used in the tests which was a bit hard to find out from the previously known good test.

I would appreciate if the core-clients team would have a look. With the workaround, the release branches are unblocked (first main, and as soon as we backport #39537 also 26.2

Thanks!

@ahus1 ahus1 removed their assignment May 7, 2025
ahus1 added a commit that referenced this issue May 7, 2025
@ahus1
Downgrade Chrome to unblock CI
a403ff0 
Workaround for #39540

Signed-off-by: Alexander Schwartz <aschwart@redhat.com>
@ahus1
Copy link
Contributor Author
ahus1 commented May 8, 2025

See comment: a403ff0#r156592758

@rmartinc rmartinc self-assigned this May 8, 2025
@rmartinc
Copy link
Contributor
rmartinc commented May 8, 2025

@ahus1 @tbroyer The comment is OK. If you use default you use the chrome installed in the GHA box, if you specify another thing (a version), that version is installed.

The problem is a real problem. It seems that chrome 136 does not respect the cache-control set to no-store and it caches the pages anyway. see issue https://issues.chromium.org/issues/415773538. They explain that now they check cookies and other things to detect that is not a login (and I think that is correct because I see the back works OK after the login). The problem is that in some places in the middle of the login we are testing that going back shows the expired page. And now that doesn't happen, a explicit refresh is needed. I'm changing the tests to do the refresh in chrome.

rmartinc added a commit to rmartinc/keycloak that referenced this issue May 8, 2025
@rmartinc
Refresh back button for chrome 136
f160ec7 
Closes keycloak#39540

Signed-off-by: rmartinc <rmartinc@redhat.com>
@rmartinc rmartinc mentioned this issue May 8, 2025
Merged
@rmartinc rmartinc added backport/26.2 priority/important Must be worked on very soon and removed status/triage labels May 8, 2025
@ahus1 ahus1 closed this as completed in 048bc88 May 9, 2025
rmartinc added a commit to rmartinc/keycloak that referenced this issue May 9, 2025
@rmartinc
Refresh back button for chrome 136
61e1545 
Closes keycloak#39540

Signed-off-by: rmartinc <rmartinc@redhat.com>
(cherry picked from commit 048bc88)
@rmartinc rmartinc mentioned this issue May 9, 2025
Merged
ahus1 pushed a commit that referenced this issue May 9, 2025
@rmartinc
Refresh back button for chrome 136
0879e5e 
Closes #39540


(cherry picked from commit 048bc88)

Signed-off-by: rmartinc <rmartinc@redhat.com>
@ahus1
Copy link
Contributor Author
ahus1 commented May 9, 2025

@rmartinc - what are your plans for 26.0? Do you plan to backport it to that release branch as well? If yes, please also cherry-pick my workaround first to avoid conflicts.

@rmartinc
Copy link
Contributor
rmartinc commented May 9, 2025

@ahus1 In 26.0 we are using chrome 134 fixed. IIRC there are two issues we should backport to moveto curent chrome in 26.0 (this one and the one about the password detected at leaked by the browser).

@ahus1
Copy link
Contributor Author
ahus1 commented May 9, 2025

Thank you for pointing me to how it works in 26.0. IMHO staying with a fixed version of Chrome in 26.0 might be sufficient. If anything breaks with a newer browser, we'll see it in 26.2 first and then hopefully consciously decide if we want to backport it to 26.0 or not.

InJoDave pushed a commit to InJoDave/keycloak that referenced this issue May 12, 2025
@rmartinc @InJoDave
Refresh back button for chrome 136
fbc7cbf 
Closes keycloak#39540

Signed-off-by: rmartinc <rmartinc@redhat.com>
@douglaspalmer douglaspalmer mentioned this issue May 13, 2025
Closed
2 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@rmartinc rmartinc

Labels
area/ci Indicates an issue on the CI kind/bug Categorizes a PR related to a bug priority/important Must be worked on very soon release/26.2.5 release/26.3.0 team/continuous-testing team/core-clients
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Refresh back button for chrome 136 rmartinc/keycloak
2 participants
@ahus1 @rmartinc
0