Re: Need help debugging slow logical replication - Mailing list pgsql-general

From sunyucong@gmail.com
Subject Re: Need help debugging slow logical replication
Date
Msg-id CAJygYd3Qfo4Pu4DWo1y-Qu6NaTjAwQrhQybR_wNCw-BjR5Jagw@mail.gmail.com
Whole thread Raw
In response to Re: Need help debugging slow logical replication  (Justin <zzzzz.graf@gmail.com>)
Responses Re: Need help debugging slow logical replication
List pgsql-general
Hi Justin, thanks for the response!

> REPLICA IDENTITY DEFAULT   will only use primary keys,  if the publisher  includes those tables, the subscriber when
replayingthe WAL will stop throwing an error not knowing how to replay the UPDATE/DELETE.
 

But I don't see any errors being thrown out in the postgresql logs?
Should I be seeing it complain there? Is postgresql falling back to
replica identity full here?

However I checked that table, it doesn't seem to be making progress at
all: so I suspect you are right that it is the problem.

> Logical Replication is most likely broken at this point.
>
> I suggest stopping logical replication and correcting tables that don't have qualifying indexes for logical
replicationby creating the necessary indexes and avoid using replica identity full.  Then restart logical replication
fromthe beginning.
 
>
>



pgsql-general by date:

Previous
From: Justin
Date:
Subject: Re: Need help debugging slow logical replication
Next
From: Brad White
Date:
Subject: Re: Quoting issue from ODBC