On Thu, Aug 4, 2022 5:49 PM shiy.f...@fujitsu.com <shiy.f...@fujitsu.com> wrote: > > On Thu, Aug 4, 2022 2:28 PM Masahiko Sawada <sawada.m...@gmail.com> > wrote: > > > > On Thu, Aug 4, 2022 at 10:37 AM Amit Kapila <amit.kapil...@gmail.com> > > wrote: > > > > > > On Wed, Aug 3, 2022 at 10:21 AM Amit Kapila > <amit.kapil...@gmail.com> > > wrote: > > > > > > > > Pushed this one and now I'll look at your other patch. > > > > > > > > > > I have pushed the second patch as well after making minor changes in > > > the comments. Alvaro [1] and Tom [2] suggest to back-patch this and > > > they sound reasonable to me. Will you be able to produce back branch > > > patches? > > > > Yes. I've attached patches for backbranches. The updates are > > straightforward on v11 - v15. However, on v10, we don't use > > wait_for_catchup() in some logical replication test cases. The commit > > bbd3363e128dae refactored the tests to use wait_for_catchup but it's > > not backpatched. So in the patch for v10, I didn't change the code > > that was changed by the commit. Also, since wait_for_catchup requires > > to specify $target_lsn, unlike the one in v11 or later, I changed > > wait_for_subscription_sync() accordingly. > > > > Thanks for your patches. > > In the patches for pg11 ~ pg14, it looks we need to add a "=pod" before the > current change in PostgresNode.pm. Right? >
By the way, I notice that in 002_types.pl (on master branch), it seems the "as well" in the following comment should be removed. Is it worth being fixed? $node_subscriber->safe_psql('postgres', "CREATE SUBSCRIPTION tap_sub CONNECTION '$publisher_connstr' PUBLICATION tap_pub WITH (slot_name = tap_sub_slot)" ); # Wait for initial sync to finish as well $node_subscriber->wait_for_subscription_sync($node_publisher, 'tap_sub'); Regards, Shi yu