On Tue, Jun 15, 2021 at 9:51 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> So, it's well over a year later, and so far as I can see exactly
> nothing has been done about snapshot_too_old's problems.
FWIW I think that the concept itself is basically reasonable. The
implementation is very flawed, though, so it hardly enters into it.
> I never liked that feature to begin with, and I would be very
> glad to undertake the task of ripping it out. If someone thinks
> this should not happen, please commit to fixing it ... and not
> "eventually".
ISTM that this is currently everybody's responsibility, and therefore
nobody's responsibility. That's probably why the problems haven't been
resolved yet.
I propose that the revert question be explicitly timeboxed. If the
issues haven't been fixed by some date, then "snapshot too old"
automatically gets reverted without further discussion. This gives
qualified hackers the opportunity to save the feature if they feel
strongly about it, and are actually willing to take responsibility for
its ongoing maintenance.
--
Peter Geoghegan