Re: PG 14 release notes, first draft
Bruce Momjian <bruce@momjian.us>
On Tue, May 18, 2021 at 08:54:56AM -0700, Peter Geoghegan wrote: > On Tue, May 18, 2021 at 8:48 AM Bruce Momjian <bruce@momjian.us> wrote: > > I am thiking the vacuum_cleanup_index_scale_factor existance in a PG 14 > > postgresql.conf will throw an error, unlike 13.x, so I do think we need > > to mention this so people will know to remove it from their > > postgresql.conf before upgrades, right? I don't think the PG 13.3 > > release note mention really makes it clear it has to be removed. In a > > dump/restore, so we retain the reloption > > vacuum_cleanup_index_scale_factor and just ignore it, or drop it on > > restore? I am hoping it is the later. > > There is no dump/restore hazard on upgrade to 14, since the > vacuum_cleanup_index_scale_factor reloption remains in place (it's > just not in psql tab completion anymore, nor is it documented, etc). So it is the former behavior --- "so we retain the reloption vacuum_cleanup_index_scale_factor and just ignore it"? > It is possible (though I would certainly say unlikely) that the > vacuum_cleanup_index_scale_factor GUC will be in somebody's > postgresql.conf from an earlier version, and won't be recognized on > upgrade to v14. So maybe you need to say something about that > particular issue -- which could be framed as finishing off the process > started by the 13.3 commit. But that's it. Yes. -- Bruce Momjian <bruce@momjian.us> https://momjian.us EDB https://enterprisedb.com If only the physical world exists, free will is an illusion.
Commits
-
doc: add mention of +4GB windows file handling in PG14 relnotes
- 90855908b751 14.0 landed
-
doc: PG 14 relnotes fixes
- 86b222b09071 14.0 landed
-
doc: PG 14 relnote updates
- a2559d409372 14.0 landed
-
doc: PG 14 relnotes, adjust pg_{read|write}_all_data entry
- 7ce7d07e1c5f 14.0 landed
-
Update PG 14 relnotes for vacuum_cost_page_miss
- 8dcae7f0a3d6 14.0 landed
-
doc: update PG 14 release notes with recent feedback
- 5eb1b27d2067 14.0 landed
-
Update replication statistics after every stream/spill.
- 592f00f8dec6 14.0 cited
-
Add information of total data processed to replication slot stats.
- f5fc2f5b23d1 14.0 cited
-
Use NameData datatype for slotname in stats.
- cca57c1d9bf7 14.0 cited
-
Add back vacuum_cleanup_index_scale_factor parameter.
- effdd3f3b633 14.0 cited
-
Don't consider newly inserted tuples in nbtree VACUUM.
- 9f3665fbfc34 14.0 cited
-
pgbench: Improve time logic.
- 547f04e7348b 14.0 cited
-
Track total amounts of times spent writing and syncing WAL data to disk.
- ff99918c625a 14.0 cited
-
Allow decoding at prepare time in ReorderBuffer.
- a271a1b50e9b 14.0 cited
-
Extend the output plugin API to allow decoding of prepared xacts.
- 0aa8a01d04c8 14.0 cited
-
Fix nbtree cleanup-only VACUUM stats inaccuracies.
- 48e1291342dd 14.0 cited
-
Execute invalidation messages for each XLOG_XACT_INVALIDATIONS message
- d7eb52d7181d 14.0 cited
-
Fix our Windows stat() emulation to handle file sizes > 4GB.
- bed90759fcbc 14.0 cited
-
Trigger autovacuum based on number of INSERTs
- b07642dbcd8d 13.0 cited
-
Speed up planning when partitions can be pruned at plan time.
- 428b260f87e8 12.0 cited
-
Remove support for password_encryption='off' / 'plain'.
- eb61136dc75a 10.0 cited
-
Increase work_mem and maintenance_work_mem defaults by 4x
- 848ae330a497 9.4.0 cited