Jeff Janes <jeff.janes@gmail.com> writes:
> While stress testing Pavan's 2nd pass vacuum visibility patch, I realized
> that vacuum/visibility was busted. But it wasn't his patch that busted it.
> As far as I can tell, the bad commit was in the
> range 692079e5dcb331..168d3157032879
> Since a run takes 12 to 24 hours, it will take a while to refine that
> interval.
Just eyeballing the commit logs, I'd have to guess that
0ac5ad5134f2769c (the foreign-key-locks patch) is the only change in
that range that seems like it might account for a visibility-ish bug.
If you're lacking patience for a bisection run, try that one first.
regards, tom lane