On Tuesday 21 Oct 2014 18:57:28 you wrote:
> If that's true, and this indeed has regressed from 1.5, it's not a
> factor of changing the isolation level, but something else in the
> lifecycle transitions that is making this prone to races. I'm not sure
> what that could be.
I rather suspect it's blind luck or the calling environment that originally
experienced the bug had some subtle change.
On Tuesday 21 Oct 2014 18:57:28 you wrote:
> If that's true, and this indeed has regressed from 1.5, it's not a
> factor of changing the isolation level, but something else in the
> lifecycle transitions that is making this prone to races. I'm not sure
> what that could be.
I rather suspect it's blind luck or the calling environment that originally
experienced the bug had some subtle change.