Pgbackrest specifying the default DB necessary/correct way ?
List,
I have configured the pgbackrest on a live DB server and a Repo Server. (
EPAS 16, RHEL 9.4 and Pgbackrest 2.52.1 )
On DB Server I have
##############################
[root@db1 ~]# cd
[root@db1 ~]# cat /etc/pgbackrest/pgbackrest.conf
[Demo_Repo]
pg1-path=/data/edb/as16/data
pg1-port=5444
pg1-user=enterprisedb
pg-version-force=16
pg1-database=edb *## Query 1. *
[global]
repo1-host=10.255.0.40
repo1-host-user=postgres
archive-async=y
spool-path=/var/spool/pgbackrest
log-level-console=info
log-level-file=debug
delta=y
[global:archive-get]
process-max=2
[global:archive-push]
process-max=4
##############################################
*## Query 1: *
In this DB server I have other databases than the default "edb"
database. Specifying the above line as pg1-database=edb // I am
not sure this line is necessary or not ?
( I am just learning and exploring PgbackRest) found online some reference
configurations so using like this )
pg1-database=edb // specifying like this, will it block other databases
on this server to get backed up ? IF yes how can I overcome this ?
I want all databases on this server to be backed up to the remote
repository.
It is a production server and I can't perform a trial and error method here
to understand how it works.
Please shed some light on this .
Thanks ,
Krishane
On Wed, Aug 28, 2024 at 1:39 AM KK CHN <kkchn.in@gmail.com> wrote:
In this DB server I have other databases than the default "edb"
database. Specifying the above line as pg1-database=edb // I am
not sure this line is necessary or not ?
The pgbackrest process needs to connect to the database, which means it
needs a user and database. You need this variable if you do not have the
default database, "postgres". If you have a database named postgres, you
can leave this out. Otherwise, yes, it is necessary.
pg1-database=edb // specifying like this, will it block other databases
on this server to get backed up ? IF yes how can I overcome this ?
pgBackRest works on a cluster level, so *all* the databases are backed up.
Indeed, it is not possible to only backup some of the databases. It's the
whole cluster.
( I am just learning and exploring PgbackRest) found online some
reference configurations so using like this )
Probably best to stick to the official docs; this section in particular is
worth a read:
https://pgbackrest.org/user-guide-rhel.html
Cheers,
Greg
Very helpful.
On Wed, Aug 28, 2024 at 5:51 PM Greg Sabino Mullane <htamfids@gmail.com>
wrote:
Show quoted text
On Wed, Aug 28, 2024 at 1:39 AM KK CHN <kkchn.in@gmail.com> wrote:
In this DB server I have other databases than the default "edb"
database. Specifying the above line as pg1-database=edb // I am
not sure this line is necessary or not ?The pgbackrest process needs to connect to the database, which means it
needs a user and database. You need this variable if you do not have the
default database, "postgres". If you have a database named postgres, you
can leave this out. Otherwise, yes, it is necessary.pg1-database=edb // specifying like this, will it block other databases
on this server to get backed up ? IF yes how can I overcome this ?
pgBackRest works on a cluster level, so *all* the databases are backed up.
Indeed, it is not possible to only backup some of the databases. It's the
whole cluster.( I am just learning and exploring PgbackRest) found online some
reference configurations so using like this )
Probably best to stick to the official docs; this section in particular is
worth a read:https://pgbackrest.org/user-guide-rhel.html
Cheers,
Greg