Re: pg_recvlogical: Prevent flushed data from being re-sent after restarting replication

Hi Fujii,


I see what you mean. For reviewing I started with writing a test that 
just reproduces the bug and documents the current behaviour.

I expected that by applying your patch the test would fail, but then we 
would just update the test accordingly. Surprisingly the test continues 
to pass.

I attached the test for your consideration. I'll have a look again 
tomorrow at both.


Kind regards,

Mircea Cadariu