Horton

Re: Support logical replication of DDLs

vignesh C <vignesh21@gmail.com>

View thread
From: vignesh C <vignesh21@gmail.com>
To: Zheng Li <zhengli10@gmail.com>
Cc: li jie <ggysxcq@gmail.com>, Ajin Cherian <itsajin@gmail.com>, Peter Smith <smithpb2250@gmail.com>, Dilip Kumar <dilipbalaut@gmail.com>, Alvaro Herrera <alvherre@alvh.no-ip.org>, "houzj.fnst@fujitsu.com" <houzj.fnst@fujitsu.com>, Amit Kapila <amit.kapila16@gmail.com>, Masahiko Sawada <sawada.mshk@gmail.com>, Japin Li <japinli@hotmail.com>, rajesh singarapu <rajesh.rs0541@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-12-14T10:34:44Z
Lists: pgsql-hackers

Commits

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

  1. Add a run_as_owner option to subscriptions.

  2. Refactor pgoutput_change().

  3. Print the correct aliases for DML target tables in ruleutils.

  4. Fix object identity string for transforms

  5. Add grantable MAINTAIN privilege and pg_maintain role.

  6. Get rid of recursion-marker values in enum AlterTableType

  7. Release cache tuple when no longer needed

  8. Empty search_path in logical replication apply worker and walsender.

  9. Refactor format_type APIs to be more modular

  10. Use wrappers of PG_DETOAST_DATUM_PACKED() more.

Attachments

On Wed, 14 Dec 2022 at 03:56, Zheng Li <zhengli10@gmail.com> wrote:
>
> Attaching V50 patch set which adds support for SECURITY LABEL commands.
> Also fixed an issue in deparse_AlterRelation from the last conflict resolution.
>
> On Mon, Dec 12, 2022 at 9:37 PM li jie <ggysxcq@gmail.com> wrote:
> >
> > I noticed that the issue of ownership seems to have not been considered.
> > For example, if a user 'a' from the publishing side creates a table t1,
> > the owner of t1 is not user 'a' after it is replicated to the subscribing side.
> > This is a situation that has not been encountered in previous DML replication.
> >
> > I think the ownership relationship should not be lost,
> > and we should perhaps add it back,
> > like pg_dump "ALTER TABLE public.t1 OWNER TO a;",
> > even though we do not currently support the replication of USER.
>
> I agree that this is a useful scenario. On the other hand, I think it
> also makes sense to keep the ownership of objects to the subscription
> owner in other use cases. Maybe we can make it optional whether to
> inherit the original ownership of the publisher on the subscriber
> using a new subscription_parameter. For example, original_ownership:
> CREATE SUBSCRIPTION mysub CONNECTION 'conninfo' PUBLICATION mypub WITH
> (original_ownership=true);
>
> Then we can change the apply worker to set the ownership of new
> objects accordingly. We also need to make the owner information
> available in the logicalddlmsg, currently we don't log this
> information.

The patch was not compiling because of recent commit
60684dd834a222fefedd49b19d1f0a6189c1632e which removes ACL_VACUUM and
ACL_ANALYZE, updated the patch to handle the changes in ddl
replication of grant command accordingly. The attached patch has the
changes for the same.

Regards,
Vignesh