test(deps): take @nextcloud/vue 9.11 again, to settle whether it caused the E2E failures - #443
Merged
Merged
Conversation
…ed the E2E failures DELIBERATE EXPERIMENT, and revertible in one commit. CI bisected launchpad's 11 E2E failures to the 9.9.0 -> 9.11.0 move: f93f767 9.11.0 E2E failure d158755 9.9.0 E2E success (the pin) But an isolated reproduction contradicts it. Running the same specs against a launchpad-demo instance on :8605, seeded exactly as CI seeds (tests/e2e/seed.sh: e2e-grantee, recipient, firstrunwizard disabled, demo-data skipped), with a local build against 9.11: conditional-visibility-editor.spec.ts 13 passed <- the 7 CI failures do NOT reproduce dashboard-sharing.spec.ts 4 failed <- but at line 83, on the switcher row, NOT on CI's .sharee-option So two runs correlate the version with the failure, and a controlled run says 9.11 passes those tests. Correlation across two CI runs is not causation, and the difference may be load rather than version. This puts 9.11 back so CI can answer it: the 7 recur -> the NcSelect rework really is the cause, and the specs get adapted to the new component the 7 pass -> they were load-flakes, the pin was incidental, and the other ten apps on 9.11 need nothing Either answer is worth more than shipping a pin whose justification does not survive a controlled test.
Contributor
Quality Report — ConductionNL/launchpad @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| composer | ✅ | ✅ 104/104 | |||
| npm | ✅ | ✅ 526/526 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | 🚨 NO VERDICT — enabled but never ran | ||||
| Hydra gates | ✅ |
Quality workflow — 2026-08-30 15:01 UTC
Download the full PDF report from the workflow artifacts.
rubenvdlinde
added a commit
that referenced
this pull request
Aug 30, 2026
#446) The experiment in #443 has its answer, and it is the one that keeps the pin: E2E on 9.11 failed with the SAME ELEVEN tests and the SAME signatures as before. 127 passed, 11 failed 7x conditional-visibility-editor waiting for '.vs__dropdown-option' 4x dashboard-sharing waiting for '.sharee-option' That is deterministic, not load. Two independent CI runs on 9.11 produce an identical failure set, and 9.9.0 produces none, so the NcSelect rework is the cause. My isolated reproduction said otherwise -- those same specs passed on a 9.11 build against a seeded launchpad-demo -- and I have not yet found what differs between that environment and CI. That gap is real and worth closing before the specs are adapted, because adapting them against an environment that does not reproduce the failure fixes nothing. So development goes back to green on the pin while that work happens, rather than sitting red. The adaptation is still the plan; this is not a retreat from it. The one adaptation already understood and shipped separately is stackiq's (#841): 9.10 turned .vs__search from the <input> into a wrapper div around it, measured on a live build. launchpad's two failure modes are different and not yet explained. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Deliberate experiment, revertible in one commit.
CI bisected launchpad's 11 E2E failures to the
9.9.0 → 9.11.0move:An isolated reproduction contradicts it. Running the same specs against a
launchpad-demoinstance on:8605— seeded exactly as CI seeds (tests/e2e/seed.sh:e2e-grantee,recipient, firstrunwizard disabled, demo-data skipped) with a local build against 9.11:So two CI runs correlate the version with the failure, and a controlled run says 9.11 passes those tests. Correlation across two runs is not causation, and the difference may be load rather than version.
What this PR is for
It puts 9.11 back so CI can answer the question:
Either answer is worth more than shipping a pin whose justification does not survive a controlled test. If the failures recur, this is reverted immediately and the pin goes back.