On Wed, Mar 04, 2026 at 11:00:00AM +0200, Alexander Lakhin wrote:
> No, I didn't. I agree that eliminating already known failures makes more
> sense. Anyway, we'll have the thread which explain why such failures can
> happen.
Fixed this second instability now with the same trick as 009, down to
v14, as there is no strong requirement for an immediate stop of the
primary to check that a standby is able to to a TLI jump when
reconnecting to the promoted standby.
--
Michael
