[08:53:50] <snitch>	 [[Tech]]; L10nM4st3r; /* Autoarchive */; https://meta.wikimedia.org/w/index.php?diff=23780878&oldid=23776419&rcid=24942503
[12:17:22] <Nemo_bis>	 Platonides: that's what I remember as well
[12:55:45] <Dreamy_Jazz>	 Is there a common set of reasons why tests would fail inconsistently and only when run with other extensions?
[12:56:01] <Dreamy_Jazz>	 Trying to deal with https://phabricator.wikimedia.org/T317411 but I have no clue where to start.
[12:57:41] <Dreamy_Jazz>	 I don't think it's because the database table wasn't reset before the test runs because the test fails in a way that would indicate that the DB is missing rows
[12:57:57] <Dreamy_Jazz>	 So if there were extra entries it wouldn't make this test case fail
[14:14:14] <Dreamy_Jazz>	 https://gerrit.wikimedia.org/r/c/mediawiki/core/+/803620 broke Special:CheckUser's paging links
[14:14:34] <Dreamy_Jazz>	 Should I put in a full revert or how else best?
[14:14:45] <Dreamy_Jazz>	 I'm not really wanting to write a patch that fixes the issue
[14:14:54] <Dreamy_Jazz>	 *issue in core
[14:15:44] <Dreamy_Jazz>	 It probably affects more than just CheckUser as it makes a @stable to override method unused
[14:41:09] <RhinosF1>	 Dreamy_Jazz: make it UBN and block the train
[14:41:20] <RhinosF1>	 That's likely the best way to get advice on a weekend
[14:41:38] <RhinosF1>	 And it definitely would be an issue if it landed in prod
[15:16:19] <Platonides>	 Dimona already set it UBN
[15:16:57] <Platonides>	 I'm pretty sure this will be a bug in the tests
[15:17:41] <Platonides>	 maybe it's sme other test cleaning "too much"
[15:18:46] <RhinosF1>	 Platonides: no that's not the same bug
[15:18:56] <RhinosF1>	 There's another one linked above about pagination
[15:19:28] <Platonides>	 oh
[15:19:40] <Platonides>	 I initially thought the phab and gerrit links were about the same issue
[16:01:30] <Dreamy_Jazz>	 Apologies. The gerrit link is already in prod
[16:01:43] <Dreamy_Jazz>	 The phab link is a test issue not yet in prod
[16:02:33] <Dreamy_Jazz>	 So I'm not sure blocking the train would work here
[16:03:01] <Dreamy_Jazz>	 for the gerrit link
[16:03:11] <Dreamy_Jazz>	 As blocking the next train won't stop the code from going in
[16:21:40] <Dreamy_Jazz>	 Actually I'm going to need to revert this change or at least have a number of things fixed
[16:22:05] <Dreamy_Jazz>	 I can't even make use of the new system as the makeLink method is private
[16:26:44] <Dreamy_Jazz>	 I've created a revert at https://gerrit.wikimedia.org/r/c/mediawiki/core/+/830985