Re: Sequence Access Methods, round two

From: Michael Paquier <michael(at)paquier(dot)xyz>
To: Chao Li <li(dot)evan(dot)chao(at)gmail(dot)com>
Cc: Xuneng Zhou <xunengzhou(at)gmail(dot)com>, Andrei Lepikhov <lepihov(at)gmail(dot)com>, Peter Eisentraut <peter(at)eisentraut(dot)org>, Kirill Reshke <reshkekirill(at)gmail(dot)com>, Peter Smith <smithpb2250(at)gmail(dot)com>, Postgres hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Sequence Access Methods, round two
Date: 2025-12-23 08:42:54
Message-ID: aUpWDj1yr3osMbhO@paquier.xyz
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, Dec 23, 2025 at 02:55:45PM +0800, Chao Li wrote:
> I just started to review this patch. While reviewing 0001, I got the
> point where you now add the 3 attributes into pg_attributes, so I
> tried to test this behavior, but a basic “create sequence” command
> failed for me:
> ```
> evantest=# create sequence ts1 start with 1 increment by 1 cache 10 cycle;
> ERROR: access method "seqlocal" does not exist
> ```
>
> Did I do something wrong?

I guess so. 0001 taken independently has no link to seqlocal.
Applying the full set of patches works here:
=# create sequence ts1 start with 1 increment by 1 cache 10 cycle;
CREATE SEQUENCE

I doubt that the regression would pass at all with this class of
failures. You may need to clean your tree and rebuild.
--
Michael

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Chao Li 2025-12-23 08:45:56 Re: Allow GUC settings in CREATE SUBSCRIPTION CONNECTION to take effect
Previous Message Fujii Masao 2025-12-23 08:41:16 Re: Fix wrong reference in pg_overexplain's doc