Re: BUG #17212: pg_amcheck fails on checking temporary relations

On Wed, Oct 6, 2021 at 10:19 AM Mark Dilger
<mark(dot)dilger(at)enterprisedb(dot)com> wrote:
> > A return value of 0 cannot be said to indicate that the database is
> > not corrupt;
>
> Nor can a non-zero value be said to indicate that the database is corrupt.

I never said otherwise. I think it's perfectly fine that there are
multiple non-zero return values. It's totally unrelated.

> I'm not sure how the idea that pg_amcheck should never give back a failure code unless there is corruption got inserted into this thread, but I'm not on board with that as an invariant statement.

I agree; I'm also not on board with it as an invariant statement.

> The differences in the upcoming version are
>
> 1) --all no longer means "all relations" but rather "all checkable relations"

Clearly pg_amcheck never checked all relations, because it never
checked indexes that are not B-Tree indexes. I'm pretty sure that I
can poke big holes in almost any positivist statement like that with
little effort.

> 2) checking options should be automatically downgraded under circumstances where they cannot be applied
> 3) unlogged relations during replication are by definition not corrupt
>
> I think #1 and #3 are unsurprising enough that they need no documentation update. #2 is slightly surprising (at least to me) so I updated the docs for it.

To me #2 sounds like a tautology. It could almost be phrased as
"pg_amcheck does not check the things that it cannot check".

--
Peter Geoghegan