On 2020-Oct-28, Alexander Kukushkin wrote:
> Hello,
>
> I know, nobody in their mind should do that, but, if the postmaster
> process is killed with SIGKILL signal, most backend processes
> correctly notice the fact of the postmaster process absence and exit.
> There is one exception though, when there are autovacuum worker
> processes they are continuing to run until eventually finish and exit.
So, if you have a manual vacuum running on the table (with
vacuum_cost_delay=0) and kill -KILL the postmaster, that one also
lingers arbitrarily long afterwards?
(I suppose the problem is not as obvious just because the vacuum
wouldn't run as long, because of no vacuum cost delay; but it'd still be
a problem if you made the table bigger.)