-
-
Notifications
You must be signed in to change notification settings - Fork 14k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
postgresql_17: init at 17.0 #351253
postgresql_17: init at 17.0 #351253
Conversation
Makes building with PostgreSQL 17 succeed. Changelog: https://github.com/tds-fdw/tds_fdw/releases/tag/v2.0.4
9a0f387
to
e01b63a
Compare
e01b63a
to
fd7b7c1
Compare
|
Hm, those build fine for me. Can you see what the error is for you, @kirillrdy?
Those fail on master, too. Haven't looked into that, yet, but shouldn't be critical for this PR. The remaining failures are for PG17. Hopefully those will be resolved over time when more extension updates are available. |
you can ignore plr build error, something with my system
fails in nix-build, but works in firefox, as for other build failures shouldn't be a blocker |
Will try to take a look today. |
|
re jit/llvm + lantern, see above: This already fails for 12-16, and also before this update. So unrelated. re timescaledb_toolkit: pg17 is not supported upstream, yet: timescale/timescaledb-toolkit#820 re pgvecto-rs: pg17 is not supported upstream, yet: tensorchord/pgvecto.rs#607 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I get a different hash for plr:
error: hash mismatch in fixed-output derivation '/nix/store/mxfgna9mmiiykdy1zlbc2qv8fzlznk1z-source.drv':
likely URL: https://github.com/postgres-plr/plr/archive/REL8_4_7.tar.gz
specified: sha256-t5tniqTYNCXb3Q4ii69kr+rx7e9LIkLzlcpqk4X8wiQ=
got: sha256-PdvFEmtKfLT/xfaf6obomPR5hKC9F+wqpfi1heBphRk=
Do you still have the old one so we can investigate what has changed?
Uh, nice catch. Seems like they force pushed the tag, because of a CI failure: https://github.com/postgres-plr/plr/commits/REL8_4_7/ I certainly obtained my hash before October 9, so that last commit is new. For the last two commits I get those hashes now: rev = "ae75fe46aa3d8a2401cc1641ecaf6a31800163b8";
sha256 = "sha256-IlVGqzgJc/B0RrPB3k3UFLVfgXKB9j9FZP3/uJQiTUc=";
rev = "3640a858b3798311ffc90f21c0c3690b046522ac";
sha256 = "sha256-PdvFEmtKfLT/xfaf6obomPR5hKC9F+wqpfi1heBphRk="; One of them matches your hash, but the other one... does not match mine. That's odd... |
My hash matches the commit from August 4: postgres-plr/plr@c9e458b So the 3 commits after that had been force pushed. |
Makes building with PostgreSQL 17 succeed. Changelog: https://github.com/postgres-plr/plr/blob/master/changelog.md#846---2023-08-01
Makes building with PostgreSQL 17 succeed. Changelog: https://github.com/pgroonga/pgroonga/releases
Makes building with PostgreSQL 17 succeed. Changelog: https://github.com/cybertec-postgresql/pg_squeeze/releases The changelog is only available for beta 2, but 1.7.0 was released properly after that without changelog.
Makes building with PostgreSQL 17 succeed. Changelog: eulerto/pg_similarity@pg_similarity_1_0...master
Also fixes the build on linux with clang (JIT).
Makes building with PostgreSQL 17 succeed. Changelog: https://github.com/sraoss/pg_ivm/releases/tag/v1.9
Makes building with PostgreSQL 17 succeed. Changelog: https://github.com/citusdata/pg_cron/releases/tag/v1.6.3 https://github.com/citusdata/pg_cron/releases/tag/v1.6.4
Makes building with PostgreSQL 17 succeed. Changelog: https://github.com/lanterndata/lantern/releases
Makes building with PostgreSQL 17 succeed. Changelog: https://github.com/plv8/plv8/blob/r3.2/Changes
Makes building with PostgreSQL 17 succeed. Changelog: https://github.com/citusdata/postgresql-topn/blob/master/CHANGELOG.md#topn-v270-october-18-2024
Broken on linux (JIT) as well.
Broken on linux (JIT) as well.
fd7b7c1
to
4cc4534
Compare
Maybe we should just mark them as broken to reduce the noise? that would it make way easier for me to merge this PR. |
Agreed. |
Those extensions don't support building with PostgreSQL 17, yet. To make sure they will show up as build failures after the next package upgrade, they all have version checks for their own package version included.
The link_llvm_objects.sh script used a hardcoded /bin/bash shebang, which failed during a JIT-enabled build.
Done - marked all extensions which don't support PG17 as broken with a comment referencing the upstream issue. I also check for the extension's current version to make sure the build will fail instead of being silently marked as broken after the next package update, so that we don't forget checking whether PG17 is then supported. I also fixed
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks!
And again, I'd like to emphasize that your work on the postgresql subsystem is greatly appreciated! <3
Took the new v17 out of #345260 to target master directly, but left the
postgresql: 16 -> 17
over there for later.@Ma27 @SuperSandro2000
Things done
nix.conf
? (See Nix manual)sandbox = relaxed
sandbox = true
nix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD"
. Note: all changes have to be committed, also see nixpkgs-review usage./result/bin/
)Add a 👍 reaction to pull requests you find important.