Come together with the global Drupal community in Rotterdam, 28 Sept – 1 Oct 2026. Sessions, contribution, connection, and Early Bird savings until 8 June.Problem/Motivation
Spin-off from #3516173: Block status code visibility condition should use a status code cache context.
We should have a test that visits one 404 page to warm the dynamic page cache, then visits a different 404 page to make sure that the minimum of page-specific queries run.
Steps to reproduce
Proposed resolution
Remaining tasks
User interface changes
Introduced terminology
API changes
Data model changes
Release notes snippet
Issue fork drupal-3589683
Show commands
Start within a Git clone of the project using the version control instructions.
Or, if you do not have SSH keys set up on git.drupalcode.org:
changes
Comments
Comment #2
catchComment #4
smustgrave commentedDon't mind marking for you but mind explaining the config change?
Comment #5
catch@smustgrave I've reverted the config change, that was a failed attempt to try to trigger the difference that #3516173: Block status code visibility condition should use a status code cache context should show, but we won't be able to see that via Umami without at least one additional change on top of this. Opened yet another issue for that one #3589822: Try to reduce cache context granularity from path conditions. The test coverage added here will be useful for that issue.
Because Umami doesn't show what I want it to show for #3516173: Block status code visibility condition should use a status code cache context I additionally added similar coverage to StandardPerformanceTest which hopefully should.
Comment #6
catchAdditionally needed to add the response status condition on the breadcrumbs block in Olivero to get this to show up. For me it makes sense to change that directly in Olivero - not sure why you'd want breadcrumbs on 404 pages, but we could also change the set-up of the performance test to do it so this doesn't get blocked on that.
Main thing is to unblock the other issue by having a test that actually demonstrates the problem.
Comment #7
smustgrave commentedThis is looking good but appears to have a breadcrumb error now
Comment #8
catchYep that's because it tests with Olivero (which is interesting in itself). Need to see what the verdict is on #3516173-23: Block status code visibility condition should use a status code cache context before doing any more here.