Re: Replication question

From: Jeff Janes <jeff(dot)janes(at)gmail(dot)com>
To: Scot Kreienkamp <Scot(dot)Kreienkamp(at)la-z-boy(dot)com>
Cc: Postgres General <pgsql-general(at)postgresql(dot)org>
Subject: Re: Replication question
Date: 2018-10-22 15:42:50
Message-ID: CAMkU=1xVFTbrXP-=uuRPLoWF3-J-=7uh8O6iER6g1nLqDRG2rA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Mon, Oct 22, 2018, 9:54 AM Scot Kreienkamp <Scot(dot)Kreienkamp(at)la-z-boy(dot)com>
wrote:

> Hi everyone,
>
>
>
> We just moved to PG9.6 from 9.1 (yeah, not my choice to wait this long).
> In 9.1 I had to make the archive location (NFS in my case) available to all
> the mirrors running PG so that they could catch up whenever they fell
> behind. I thought I read somewhere that in 9.6, as long as the WAL log is
> available on disk or in the archive the replication server will provide
> that to the replication client, and my archive NFS mount didn’t have to be
> available to all replication clients. It doesn’t seem to be operating that
> way though. Did I completely remember that wrong or did I misunderstand
> something?
>

The master won't read from the archives for you in order to send to an
replica. But using replication slots, you can keep the needed log files
right in pg_xlog/pg_wal until all replicas get what they need (assuming the
disk is large enough). Then you don't need an archive at all for
replication purposes, still might for pitr purposes.

Perhaps this is what you heard about.

Cheers,

Jeff

>
>

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Scot Kreienkamp 2018-10-22 17:20:40 RE: Replication question
Previous Message Scottix 2018-10-22 15:22:34 Re: Optimizing Postgresql ILIKE while query