Hi Leonid,

On Thu, Feb 5, 2026 at 10:25 PM Leonid Borchuk <[email protected]> wrote:
>
> Hi!
>
> Sounds wise. How long are we going to support 2.X and future 3.X branches
> before migrating all users to rebased to PG16 (4.X?) branch?
>
> If the answer is more than 2-3 years (it seems to me that it is), the
> answer to upgrade is yes, it would be great to upgrade to PG 14.20

Good question. I think we don’t have a finalized maintenance policy
for 2.x and 3.x yet. I think this deserves a dedicated thread so we
can discuss further and reach a community consensus. My guess is that
the support window will largely depend on our community bandwidth and
capacity.

>From a user perspective, 2.x and 3.x are already in use today. If we
can continue to deliver incremental improvements that are mostly
drop-in compatible, that would be very welcome. Many users may have
just adopted 2.x/3.x, and moving to PG16-based versions could be
costly for them.

> The only objection I can think of is that the EOL for PG14 is this year,
> November 12, 2026. See https://www.postgresql.org/support/versioning/. But
> maybe it's not aт objection at all, we could backport the most important
> CVE's. The stability is more important than a fast switch to a new PG
> kernel.

I agree. Also, the PG16-based version will take some time to stabilize
and mature, so in the near term I think 2.x/3.x should still be
encouraged for production users.

Best,
Dianjin Wang

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to