Re: pg_service.conf and client support - Mailing list pgsql-general

From Adrian Klaver
Subject Re: pg_service.conf and client support
Date
Msg-id bb69ce6b-90b6-12cb-ca0f-8279d48b2b29@aklaver.com
Whole thread Raw
In response to pg_service.conf and client support  (Niels Jespersen <NJN@dst.dk>)
List pgsql-general
On 6/13/20 10:03 PM, Niels Jespersen wrote:
> Hello all
> 
> I just found out about the pg service file. 
> https://www.postgresql.org/docs/current/libpq-pgservice.html
> 
> I don’t know why it took me so long finding this. I have been looking 
> for ways to abstract physical details of data location away for clients 
> (much like Oracle’s tnsnames).
> 
> Want to move a cluster to a new host. Sure, move it, edit 
> pg_service.conf and clients will not know the difference.
> 
> It works great for at least psql and psycopg2.
> 
> But I cannot find anything on pg_service.conf and the Postgres ODBC 
> driver and Npgsql for .Net
> 
> I know pg_service.conf support is implemented through libpq and support 
> for rivers not using libpq is not a given thing.
> 
> But I think the need for the abstraction of connection details is a 
> general one.
> 
> Can anyone shed som light on the ubiquitousness of support for 
> pg_service.conf?
> 
> Are there any other mechanisms with broader support, that can be used 
> instead of pg_service.conf (if support is scarce beyond what builds on 
> libpq)?

Only thing I know of is something like 
pgbouncer(http://www.pgbouncer.org/) where you keep a constant 
connection setting for the pooler and then change the settings for the 
database(s) it connects to.

> 
> Thank you.
> 
> Regards Niels Jespersen
> 


-- 
Adrian Klaver
adrian.klaver@aklaver.com



pgsql-general by date:

Previous
From: Adrian Klaver
Date:
Subject: Re: BUG #11141: Duplicate primary key values corruption
Next
From: Sreerama Manoj
Date:
Subject: Re: A query in Streaming Replication