On Sat, 2023-03-25 at 19:58 -0400, Dave Cramer wrote:
> Well that means that connection poolers have to all be fixed. There
> are more than just pgbouncer.
> Seems rather harsh that a new feature breaks a connection pooler or
> makes the pooler unusable.
Would it actually break connection poolers as they are now? Or would,
for example, pgbouncer just not set the binary_format parameter on the
outbound connection, and therefore just return everything as text until
they add support to configure it?
I'll admit that GUCs wouldn't have this problem at all, but it would be
nice to know how much of a problem it is before we decide between a
protocol extension and a GUC.
Regards,
Jeff Davis
