Re: [HACKERS] Improve logical decoding error message (was wal_level> WAL_LEVEL_LOGICAL) - Mailing list pgsql-hackers

From Andres Freund
Subject Re: [HACKERS] Improve logical decoding error message (was wal_level> WAL_LEVEL_LOGICAL)
Date
Msg-id 20170523131210.lpbe42kqaufy4whj@alap3.anarazel.de
Whole thread Raw
In response to Re: [HACKERS] Improve logical decoding error message (was wal_level > WAL_LEVEL_LOGICAL)  (Neha Khatri <nehakhatri5@gmail.com>)
List pgsql-hackers
On 2017-05-23 10:49:54 +0000, Neha Khatri wrote:
> On Tue, 23 May 2017 at 10:55 am, Alvaro Herrera <alvherre@2ndquadrant.com>
> wrote:
> 
> > Neha Khatri wrote:
> > > On Tue, May 23, 2017 at 10:26 AM, Michael Paquier <
> > michael.paquier@gmail.com
> >
> > > > There is no wal_level higher than logical, so the current sense looks
> > > > perfectly fine to me.
> > >
> > > If there is no wal_level higher than logical, should the following error
> > > message indicate to set it >= logical.
> > >
> > >  select * from
> > > pg_create_logical_replication_slot('regression_slot','test_decoding');
> > >  ERROR:  logical decoding requires wal_level >= logical
> >
> > I think it's purposefully ambiguous to cover a possible future
> > extension.

Right, IIRC that's how this notion started.


> Should documentation also have similar statement and indicate future
> possibility.
> 
> What is the benefit of having it just in error message.

I personally wouldn't do anything here, it doesn't seem an issue.


- Andres



pgsql-hackers by date:

Previous
From: tushar
Date:
Subject: Re: [HACKERS] Getting server crash after running sqlsmith
Next
From: Mithun Cy
Date:
Subject: Re: [HACKERS] Proposal : For Auto-Prewarm.