Horton

Re: DROP OWNED BY fails to clean out pg_init_privs grants

Daniel Gustafsson <daniel@yesql.se>

View thread View on postgr.es
From: Daniel Gustafsson <daniel@yesql.se>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Hannu Krosing <hannuk@google.com>, "David G. Johnston" <david.g.johnston@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>, Stephen Frost <sfrost@snowman.net>, Andres Freund <andres@anarazel.de>, Noah Misch <nmisch@google.com>
Date: 2024-05-24T11:27:18Z
Lists: pgsql-hackers 
> On 24 May 2024, at 01:01, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> 
> Hannu Krosing <hannuk@google.com> writes:
>> While the 'DROP OWNED BY fails to clean out pg_init_privs grants'
>> issue is now fixed,we have a similar issue with REASSIGN OWNED BY that
>> is still there:
> 
> Ugh, how embarrassing.  I'll take a look tomorrow, if no one
> beats me to it.

I had a look, but I didn't beat you to a fix since it's not immediately clear
to me how this should work for REASSING OWNED (DROP OWNED seems a simpler
case).  Should REASSIGN OWNED alter the rows in pg_shdepend matching init privs
from SHARED_DEPENDENCY_OWNER to SHARED_DEPENDENCY_INITACL, so that these can be
mopped up with a later DROP OWNED?  Trying this in a POC patch it fails with
RemoveRoleFromInitPriv not removing the rows, shortcircuiting that for a test
seems to make it work but is it the right approach?

--
Daniel Gustafsson

Commits

  1. Allow meson builds to run test_pg_dump test in installcheck mode.

  2. Remove recordExtensionInitPriv[Worker]'s ownerId argument.

  3. Improve tracking of role dependencies of pg_init_privs entries.

  4. Fix failure to track role dependencies of pg_init_privs entries.

  5. Drop global objects after completed test