Re: pg_listener entries deleted under heavy NOTIFY load only on Windows

"Marshall, Steve" <smarshall@wsi.com> writes:
> The solution that seems to work is to call CallNamedPipe repeatedly in a
> loop if it fails.

Seems a bit nervous-making, since there's no principled way to know how
many retries to try.

> I also notice that signals that should fail, do fail.  For example,
> signal 30 seems to be regularly sent to pid 1, and this fails in both
> the orignal code and my modified version.

That's probably this bug:
http://archives.postgresql.org/pgsql-bugs/2008-11/msg00077.php
http://archives.postgresql.org/pgsql-committers/2008-11/msg00132.php

It shouldn't be happening, but it does offer a convenient test case that
your patch still fails when it should fail ;-)

            regards, tom lane