Judging by the fact that this behaviour started on enwiki on 7 November (this is the first such creation log entry I found), it looks like it was caused by something in the 1.44.0-wmf.2 release.

It looks like https://gerrit.wikimedia.org/r/c/mediawiki/core/+/1069665/29/includes/page/MovePage.php removed a call to setUsePageCreationLog( false ) and didn't replace it with anything else (as far as I can see). I'll ask in T377229 whether or not this change was intentional.

Change #1089972 had a related patch set uploaded (by Ammarpad; author: Ammarpad):

[mediawiki/core@master] page: Restore call to PageUpdater::setUsePageCreationLog()

https://gerrit.wikimedia.org/r/1089972

In T379380#10306402, @A_smart_kitten wrote:

It looks like https://gerrit.wikimedia.org/r/c/mediawiki/core/+/1069665/29/includes/page/MovePage.php removed a call to setUsePageCreationLog( false )

It's indeed the cause.

and didn't replace it with anything else (as far as I can see).

The call to ->setFlags( EDIT_SUPPRESS_RC ) was supposed to magically replace that (and it's already there). However, what would make that possible was reverted in https://gerrit.wikimedia.org/r/c/mediawiki/core/+/1085458

@Ammarpad Thanks for the patch :) Unless I’m missing something, is it just waiting for code review now?

Hi @daniel Do you consider this a regression or was it intentional change? This is result of and https://gerrit.wikimedia.org/r/c/mediawiki/core/+/1069665 and subsequet partial revert.

In T379380#10334772, @Ammarpad wrote:

Hi @daniel Do you consider this a regression or was it intentional change? This is result of and https://gerrit.wikimedia.org/r/c/mediawiki/core/+/1069665 and subsequet partial revert.

That was unintentional. I was hoping we could get rid of setUsePageCreationLog( false ), that didn't quite work out, and I failed to put it back.

Change #1089972 merged by jenkins-bot:

[mediawiki/core@master] page: Restore call to PageUpdater::setUsePageCreationLog()

https://gerrit.wikimedia.org/r/1089972

Is it worth backporting this patch?

In T379380#10335966, @A_smart_kitten wrote:

Is it worth backporting this patch?

IMO no.

I also think the old behavior was correct - it shouldn't be possible to create a page without a page creation log entry. But regardless of whose right this is the sort of edge case that won't cause any major harm if we let it stay broken for an extra week.

Anyone who disagrees is of course welcome to go through the mechanics of backporting.

In T379380#10339005, @Pppery wrote:

IMO no.

I also think the old behavior was correct - it shouldn't be possible to create a page without a page creation log entry. But regardless of whose right this is the sort of edge case that won't cause any minor harm if we let it stay broken for an extra week.

I do see where you’re coming from (even if I don’t know whether I agree with what you’re saying); however, I feel that - if this behaviour is to be modified - it should be done as a result of an intentional proposal, rather than by not reverting an unintentional change. Either way, thanks for the answer re backporting :)