PostgreSQL 17 release announcement draft

On 9/4/24 5:04 PM, Jonathan S. Katz wrote:

Hi,

Attached is the draft of the PostgreSQL 17 release announcement. This is
a draft of the text that will go into the press kit, with the key
portions to review starting from the top of the document, up until the
"About PostgreSQL" section.

Please provide feedback on content accuracy, notable omissions or items
that should be excluded, or if an explanation is unclear and needs
better phrasing. On the last point, I'm looking to ensure the wording is
clear and is easy to translate into different languages.

Based on feedback, I'll be posting a revision once a day (if there's
feedback) until the review cut-off. We'll have to freeze the
announcement by Mon, Sep 9 @ 12:00 UTC so we can begin the translation
process.

Thank you for your help with the release process!

Please see v2 attached. As per original note, please provide feedback
before Mon, Sep 9 @ 12:00 UTC so we can begin the translation process.

Thanks,

Jonathan

On Fri, 6 Sept 2024 at 19:04, Jonathan S. Katz <jkatz@postgresql.org> wrote:

On 9/4/24 5:04 PM, Jonathan S. Katz wrote:

Hi,

Attached is the draft of the PostgreSQL 17 release announcement. This is
a draft of the text that will go into the press kit, with the key
portions to review starting from the top of the document, up until the
"About PostgreSQL" section.

Please provide feedback on content accuracy, notable omissions or items
that should be excluded, or if an explanation is unclear and needs
better phrasing. On the last point, I'm looking to ensure the wording is
clear and is easy to translate into different languages.

Based on feedback, I'll be posting a revision once a day (if there's
feedback) until the review cut-off. We'll have to freeze the
announcement by Mon, Sep 9 @ 12:00 UTC so we can begin the translation
process.

Thank you for your help with the release process!

Please see v2 attached. As per original note, please provide feedback
before Mon, Sep 9 @ 12:00 UTC so we can begin the translation process.

[`EXPLAIN`](...) now shows the time spent for I/O block reads and writes

I think this needs some adjustment: IIUC the new feature in PG17's
295c36c0 is that we now also track (and show) timings for local
blocks. I/O timings on shared and temp blocks were already tracked
(and displayed with the BUFFERS option) when track_io_timing was
enabled: temp timing was introduced with efb0ef90 in early April 2022,
and the output of IO timings for shared blocks has existed since the
introduction of track_io_timing in 40b9b957 back in late March of
2012.

Kind regards,

Matthias van de Meent

On Fri, 6 Sept 2024 at 19:04, Jonathan S. Katz <jkatz@postgresql.org> wrote:

Please see v2 attached. As per original note, please provide feedback
before Mon, Sep 9 @ 12:00 UTC so we can begin the translation process.

The following sentence was part of the beta1 release announcement, but
is not part of this draft. Did that happen by accident or was it done
on purpose? If on purpose, I'm curious why since I contributed the
feature.

"PostgreSQL 17 also provides better support for asynchronous and more
secure query cancellation routines, which drivers can adopt using the
libpq API."

On 9/6/24 6:40 PM, Jelte Fennema-Nio wrote:

On Fri, 6 Sept 2024 at 19:04, Jonathan S. Katz <jkatz@postgresql.org> wrote:

Please see v2 attached. As per original note, please provide feedback
before Mon, Sep 9 @ 12:00 UTC so we can begin the translation process.

The following sentence was part of the beta1 release announcement, but
is not part of this draft. Did that happen by accident or was it done
on purpose? If on purpose, I'm curious why since I contributed the
feature.

"PostgreSQL 17 also provides better support for asynchronous and more
secure query cancellation routines, which drivers can adopt using the
libpq API."

The beta announcement typically calls out more of the features available
to ensure we can get folks testing them. Features that target API
integrations/internals aren't usually mentioned in the GA announcement.

Jonathan

On 9/6/24 2:01 PM, Matthias van de Meent wrote:

I think this needs some adjustment: IIUC the new feature in PG17's
295c36c0 is that we now also track (and show) timings for local
blocks. I/O timings on shared and temp blocks were already tracked
(and displayed with the BUFFERS option) when track_io_timing was
enabled: temp timing was introduced with efb0ef90 in early April 2022,
and the output of IO timings for shared blocks has existed since the
introduction of track_io_timing in 40b9b957 back in late March of
2012.

Thanks Matthias. I updated the draft to specify this is for local
blocks. I've attached the latest copy.

Thanks,

Jonathan

On Sun, 8 Sept 2024 at 06:44, Jonathan S. Katz <jkatz@postgresql.org> wrote:

I've attached the latest copy.

Is "This release expands on functionality both for managing data in
partitions" still relevant given the MERGE/SPLIT PARTITION was
reverted [1]https://git.postgresql.org/gitweb/?p=postgresql.git;a=commit;h=3890d90c1508125729ed20038d90513694fc3a7b?

David

[1]: https://git.postgresql.org/gitweb/?p=postgresql.git;a=commit;h=3890d90c1508125729ed20038d90513694fc3a7b

On 9/7/24 6:58 PM, David Rowley wrote:

On Sun, 8 Sept 2024 at 06:44, Jonathan S. Katz <jkatz@postgresql.org> wrote:

I've attached the latest copy.

Is "This release expands on functionality both for managing data in
partitions" still relevant given the MERGE/SPLIT PARTITION was
reverted [1]?

AFAICT yes, as identity columns and exclusion constraints can now be
used. It looks like there are a few items with the optimizer (which
looking at the release notes, you'd be aware of :) but unsure if those
should be added.

Timing wise, we're rapidly approaching the Sep 9 12:00 UTC cutoff; if
there's no additional feedback, once I wake up I'll do one more pass
over the text and then freeze it.

Thanks,

Jonathan