Support for Drupal 7 is ending on 5 January 2025—it’s time to migrate to Drupal 10! Learn about the many benefits of Drupal 10 and find migration tools in our resource center.
Many tests are now failing with UpdatePathTestBase, could be a core regression.
The first test we see this issue with is https://dispatcher.drupalci.org/job/default/37588/consoleFull
The most recent is https://dispatcher.drupalci.org/job/default/37779/consoleFull
Total of 22 tests have failed like this today..
Its possible that something in #2382675: hook_entity_create() affects the data of new translations of existing entities in unexpected and undocumented ways is causing some kind of racey condition to happen.
Comment | File | Size | Author |
---|---|---|---|
#6 | 2616808.revert-2382675.patch | 20.57 KB | alexpott |
| |||
#6 | 2616808.revert-2382675.patch | 20.57 KB | alexpott |
| |||
#6 | 2616808.revert-2382675.patch | 20.57 KB | alexpott |
| |||
#6 | 2616808.revert-2382675.patch | 20.57 KB | alexpott |
| |||
#6 | 2616808.revert-2382675.patch | 20.57 KB | alexpott |
|
Comments
Comment #2
MixologicComment #3
xjmI've seen very similar failures before. I have a link to https://www.drupal.org/pift-ci-job/66336 in my email from Oct. 30:
Comment #4
alexpottLet's test the hypothesis that this is caused by #2382675: hook_entity_create() affects the data of new translations of existing entities in unexpected and undocumented ways
Comment #5
MixologicSo the only other time this shows up in any of the log files prior to today is back in october 29th-30th:
Fyi you can go from the full console back to drupal.org from the very top of the output
Patch tests:
https://dispatcher.drupalci.org/job/default/29248 (https://www.drupal.org/pift-ci-job/65190)
https://dispatcher.drupalci.org/job/default/29276 (https://www.drupal.org/pift-ci-job/65248)
https://dispatcher.drupalci.org/job/default/29560 (https://www.drupal.org/pift-ci-job/65532)
https://dispatcher.drupalci.org/job/default/29645 (https://www.drupal.org/pift-ci-job/65748)
https://dispatcher.drupalci.org/job/default/29709 (https://www.drupal.org/pift-ci-job/65874)
Branch tests
https://dispatcher.drupalci.org/job/php5.6_mysql5.5/168/ (https://www.drupal.org/pift-ci-job/65632)
https://dispatcher.drupalci.org/job/php5.6_mysql5.5/169/ (https://www.drupal.org/pift-ci-job/66336 the one xjm linked)
DrupalCI definitely had a lot of other problems on the 29th/30th, so I wouldnt be surprised if those jobs are red herrings. nevertheless they failed in the same way so can provide some clues, hopefully, to what is causing this.
Comment #6
alexpottA different approach...
Comment #10
alexpottI've reverted #2382675: hook_entity_create() affects the data of new translations of existing entities in unexpected and undocumented ways - tentatively marking as rtbc. Let's see how things work out.
Comment #24
alexpottYep this revert has fixed the random error.
Comment #25
plach@Mixologic:
@berdir wrote in #2382675-48: hook_entity_create() affects the data of new translations of existing entities in unexpected and undocumented ways:
Comment #26
plachReopening this and repurposing it to figure out what's actually wrong with the
is_subclass_of()
check, see #2382675-50: hook_entity_create() affects the data of new translations of existing entities in unexpected and undocumented ways.Comment #27
plachIt's probably a PHP 5.5 bug, see #2496337-84: [plach] Testing issue. What patch version are the 5.5 tests running on?
Comment #28
plach#2382675: hook_entity_create() affects the data of new translations of existing entities in unexpected and undocumented ways was committed again, no way to reproduce it locally, not even via a DrupalCI testbot, so setting to maintainer needs more info.
Comment #37
larowlanReviewed this as part of the Bug Smash Initiative, will reach out to @plach on slack to see if this can now be closed.
Comment #38
plachIt seems this was never reproduced since then, closing.
Comment #39
plachCrediting people