On Wed, Apr 1, 2020 at 9:02 AM Robert Haas <robertmhaas@gmail.com> wrote:
> I complained
> when you added those error checks to vacuum in back-branches, and
> since that release went out people are regularly tripping those checks
> and taking prolonged outages for a problem that wasn't making them
> unhappy before. I know that in theory those people are better off
> because their database was always corrupted and now they know. But for
> some of them, those prolonged outages are worse than the problem they
> had before. I believe it was irresponsible to decide on behalf of our
> entire user base that they were better off with such a behavior change
> in a supposedly-stable branch, and I believe the same thing here.
I agreed with that decision, FWIW. Though I don't deny that there is
some merit in what you say. This is the kind of high level
philosophical question where large differences of opinion are quite
normal.
I don't think that it's fair to characterize Andres' actions in that
situation as in any way irresponsible. We had an extremely complicated
data corruption bug that he went to great lengths to fix, following
two other incorrect fixes. He was jet lagged from travelling to India
at the time. He went to huge lengths to make sure that the bug was
correctly squashed.
> Actually removing the code is unnecessary, protects
> nobody, and has risk.
Every possible approach has risk. We are deciding among several
unpleasant and risky alternatives here, no?
--
Peter Geoghegan
