Re: BUG #18964: `ALTER DATABASE ... RESET ...` fails to reset extension parameters that no longer exist
Nathan Bossart <nathandbossart@gmail.com>
From: Nathan Bossart <nathandbossart@gmail.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Laurenz Albe <laurenz.albe@cybertec.at>, mert@futo.org, pgsql-bugs@lists.postgresql.org
Date: 2025-07-30T19:13:16Z
Lists: pgsql-bugs
Commits
Same data as JSON:
GET /api/v1/messages/:b64id/commits
the thread's linked commits as JSON, with link sources.
API reference →
-
Allow resetting unknown custom GUCs with reserved prefixes.
- f79ca73d75cb 15.14 landed
- b998ce3272ab 16.10 landed
- 9eb6068fb64c 19 (unreleased) landed
- 7b9674a8b180 18.0 landed
- 39ff056365e4 17.6 landed
-
Allow ALTER SYSTEM to set unrecognized custom GUCs.
- 2d870b4aeff1 17.0 cited
-
Disallow setting bogus GUCs within an extension's reserved namespace.
- 88103567cb8f 15.0 cited
Attachments
On Mon, Jul 28, 2025 at 05:27:50PM -0400, Tom Lane wrote: > Nathan Bossart <nathandbossart@gmail.com> writes: >> I looked into how easily this back-patched to v15 (where commit 88103567c >> was added), and I noticed that the ALTER SYSTEM code looks a bit different >> on v15 and v16. Furthermore, I didn't see a simple way to fix it on those >> versions without first back-patching commit 2d870b4. From the commit >> message, it looks like Tom didn't back-patch it at the time due to a lack >> of complaints. I'm currently thinking we should first back-patch that one >> to at least v15, if not all supported versions, before applying my proposed >> patch. Thoughts? > > No objection here. Now that that's a couple years old, it should have > had enough time to bake. It applies relatively cleanly down to v15. The only thing I had to change was to replace guc_free() with free() on v15. I was a little worried about changing the signature of check_GUC_name_for_parameter_acl(), but codesearch.debian.net doesn't show any outside uses. -- nathan