Advice for geographically dispersed multi master - Mailing list pgsql-general

From Nikolai Lusan
Subject Advice for geographically dispersed multi master
Date
Msg-id cfccf95fb15bd9f8fcc286b174f0b650cc30e1c6.camel@lusan.id.au
Whole thread Raw
Responses Re: Advice for geographically dispersed multi master
Re: Advice for geographically dispersed multi master
List pgsql-general
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

Hi,

/*
   Disclaimer:
   I am primarily a systems administrator with some database and DBA
   history. Although these days I would normally leave this kind of thing
   to a seasoned DBA this is for my personal use, and the use of a non-
   profit organisation that I am doing some volunteer support work for. 
*/


So I am looking to move my current setup of 3 separate, individual,
postgres 11 servers into a multi master cluster. I have done this with
OpenLDAP, which is used for various purposes, but want to extend this to
cover my DB installs.

I have read the various replication and clustering documentation for
postgresql 11, and it looks like what I want is "Synchronous Multimaster
Replication". The organisation I am doing this for does not have the money
to throw at a commercial solution like BRD, but from my reading it looks
like PG v12 may have a method to make such a solution easier to achieve
with stock PG.

The three servers I am using are all in different DC's (2 VPS and one bare
metal install). I would like to have all my servers containing all the PG
databases on each server for various reasons (to supplement backups, and
make recovery faster/easier. To allow applications on different servers
access to various databases locally (rather than transmitting all SQL
requests over network). The list goes on, but I don't want to bore people
more than I already have.). There is also the possibility that a subset of
these databases may be replicated to a fourth, or even fifth location
depending on needs that arise.

So:
   1) Should I just wait for PG v12 to come out of RC status and try to set
      it up then?

   2) Is there are way to reconfigure my current servers to replicate the
      databases to the other servers?

   3) Is there a better way to do this full stop?

Any help/advice is greatly welcomed, and appreciated.

- -- 
Nikolai Lusan <nikolai@lusan.id.au>
-----BEGIN PGP SIGNATURE-----
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=7Rs2
-----END PGP SIGNATURE-----




pgsql-general by date:

Previous
From: Aleš Zelený
Date:
Subject: Re: Wall shiping replica failed to recover database with error:invalid contrecord length 1956 at FED/38FFE208
Next
From: "Enrico Thierbach"
Date:
Subject: Re: partitions vs indexes