Re: Initial Schema Sync for Logical Replication

[Peter Smith]

On Thu, Jun 15, 2023 at 4:14 PM Peter Smith <smithpb2...@gmail.com> wrote:
>
> On Thu, Jun 8, 2023 at 1:24 PM Masahiko Sawada <sawada.m...@gmail.com> wrote:
> >
> ...
>
> > We also need to research how to integrate the initial schema
> > synchronization with tablesync workers.  We have a PoC patch[2].
> >
> > Regards,
> >
> > [1] 
> > https://wiki.postgresql.org/wiki/Logical_replication_of_DDLs#Initial_Schema_Sync
> > [2] 
> > https://www.postgresql.org/message-id/CAD21AoCdfg506__qKz%2BHX8vqfdyKgQ5qeehgqq9bi1L-6p5Pwg%40mail.gmail.com
> >
>
> FYI -- the PoC patch fails to apply using HEAD fetched today.
>
> git apply 
> ../patches_misc/0001-Poc-initial-table-structure-synchronization-in-logic.patch
> error: patch failed: src/backend/replication/logical/tablesync.c:1245
> error: src/backend/replication/logical/tablesync.c: patch does not apply
>

After rebasing the PoC patch locally, I found the 'make check' still
did not pass 100%.

# 2 of 215 tests failed.

Here are the differences:

diff -U3 /home/postgres/oss_postgres_misc/src/test/regress/expected/rules.out
/home/postgres/oss_postgres_misc/src/test/regress/results/rules.out
--- /home/postgres/oss_postgres_misc/src/test/regress/expected/rules.out
   2023-06-02 23:12:32.073864475 +1000
+++ /home/postgres/oss_postgres_misc/src/test/regress/results/rules.out
2023-06-15 16:53:29.352622676 +1000
@@ -2118,14 +2118,14 @@
     su.subname,
     st.pid,
     st.leader_pid,
-    st.relid,
+    st.subrelid,
     st.received_lsn,
     st.last_msg_send_time,
     st.last_msg_receipt_time,
     st.latest_end_lsn,
     st.latest_end_time
    FROM (pg_subscription su
-     LEFT JOIN pg_stat_get_subscription(NULL::oid) st(subid, relid,
pid, leader_pid, received_lsn, last_msg_send_time,
last_msg_receipt_time, latest_end_lsn, latest_end_time) ON ((st.subid
= su.oid)));
+     LEFT JOIN pg_stat_get_subscription(NULL::oid) st(subid,
subrelid, pid, leader_pid, received_lsn, last_msg_send_time,
last_msg_receipt_time, latest_end_lsn, latest_end_time) ON ((st.subid
= su.oid)));
 pg_stat_subscription_stats| SELECT ss.subid,
     s.subname,
     ss.apply_error_count,
diff -U3 /home/postgres/oss_postgres_misc/src/test/regress/expected/oidjoins.out
/home/postgres/oss_postgres_misc/src/test/regress/results/oidjoins.out
--- /home/postgres/oss_postgres_misc/src/test/regress/expected/oidjoins.out
2022-10-04 15:11:32.457834981 +1100
+++ /home/postgres/oss_postgres_misc/src/test/regress/results/oidjoins.out
 2023-06-15 16:54:07.159839010 +1000
@@ -265,4 +265,3 @@
 NOTICE:  checking pg_subscription {subdbid} => pg_database {oid}
 NOTICE:  checking pg_subscription {subowner} => pg_authid {oid}
 NOTICE:  checking pg_subscription_rel {srsubid} => pg_subscription {oid}
-NOTICE:  checking pg_subscription_rel {srrelid} => pg_class {oid}

------
Kind Regards,
Peter Smith.
Fujitsu Australia