Re: Query ID Calculation Fix for DISTINCT / ORDER BY and LIMIT / OFFSET

On Mon, 10 Mar 2025 at 12:39, Michael Paquier <michael(at)paquier(dot)xyz> wrote:
>
> On Thu, Mar 06, 2025 at 06:44:27PM -0600, Sami Imseih wrote:
> > Regarding the issue itself, query jumbling behavior is often subjective,
> > making it difficult to classify as a bug. I'm not entirely sure this
> > qualifies as a bug either, but I do believe it should be addressed.
>
> I would call that a bug and something that we should fix, but not
> something that we can really backpatch as this has unfortunately an
> impact on monitoring tools. Stability takes priority in this area in
> already released branches.

I know you reviewed this, Michael, so you're aware; 2d3389c28 did
recently document that we'd only break this in minor versions as a
last resort. So, I agree that it sounds like a master-only fix is in
order.

For the record, the docs [1] read:

"Generally, it can be assumed that queryid values are stable between
minor version releases of PostgreSQL, providing that instances are
running on the same machine architecture and the catalog metadata
details match. Compatibility will only be broken between minor
versions as a last resort."

David

[1] https://www.postgresql.org/docs/current/pgstatstatements.html