Re: BUG #18545: \dt breaks transaction, calling error when executed in SET SESSION AUTHORIZATION

Tom Lane <tgl@sss.pgh.pa.us>

From: Tom Lane <tgl@sss.pgh.pa.us>

To: Андрей Рачицкий <therealgofman@mail.ru>

Cc: pgsql-bugs@lists.postgresql.org, Robert Haas <robertmhaas@gmail.com>

Date: 2024-07-31T22:59:16Z

Lists: pgsql-bugs, pgsql-hackers

=?UTF-8?B?0JDQvdC00YDQtdC5INCg0LDRh9C40YbQutC40Lk=?= <therealgofman@mail.ru> writes:
> Hi, Tom! Thank you for work on the subject.  After applying patch, problem is no longer reproducible.

Thanks for checking.  I realized that the idea of making
check_session_authorization a no-op was wrong as presented:
if we don't set up an "extra" struct then guc.c would be
unable to restore the setting later, in case say a function
that's run inside the parallel query has a SET
session_authorization clause.  It's probably possible to
revive that idea with more work, but it's not essential to the
bug fix and we're getting close to the August minor releases.
So I pushed the core bug fix, and I'll take another look at
that part later.

			regards, tom lane

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

Fix parallel worker tracking of new catalog relfilenumbers.
- 67bab53d6442 18.0 landed
Restore relmapper state early enough in parallel workers.
- 126ec0bc76d0 18.0 landed
Allow parallel workers to cope with a newly-created session user ID.
- 6e086fa2e771 18.0 landed
- f5f30c22ed69 18.0 landed
- 97380d4803d1 14.13 landed
- 849326e49a5d 16.4 landed
- 68855c03878c 12.20 landed
- 5887dd4894db 17.0 landed
- 48536305370a 15.8 landed
- 216201027d90 13.16 landed
Clean up handling of client_encoding GUC in parallel workers.
- 0ae5b763ea0e 18.0 landed
Fix rd_firstRelfilenodeSubid for nailed relations, in parallel workers.
- fe4d022c8e17 14.0 cited
Fix several mistakes around parallel workers and client_encoding.
- 10c0558ffefc 9.6.0 cited