Re: streaming replication corruption Q's

From: Laurenz Albe <laurenz(dot)albe(at)cybertec(dot)at>
To: ghiureai <isabella(dot)ghiurea(at)nrc-cnrc(dot)gc(dot)ca>, "pgsql-admin(at)postgresql(dot)org" <pgsql-admin(at)postgresql(dot)org>
Subject: Re: streaming replication corruption Q's
Date: 2017-10-05 07:07:27
Message-ID: 1507187247.3119.1.camel@cybertec.at
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-admin

ghiureai wrote:
> I'm reading PG 10 streaming replication manual, I would like to learn
> from your expertise if this type of replication will copy a bad
> segment/bit corrupt from master to slave host , will this possible to
> happened ?

Streaming replication happens on a physical level, so it does not pay
attention to logical consistency.
If the primary is suffering from data corruption, every data
modification containing that data corruption will be replicated.

> What are the most common issues you seeing when running streaming
> replication, any files/bits corruption ( from my understanding
> logical replication will replicate data corruption)where do you have to
> pay attention when setting up a new system like this ,

That question is too general...

Make sure that you are running the same operating system version on
(ideally) the same hardware on all machines.

Either set wal_keep_segments high enough or (more cumbersome, but
more reliable) configure WAL shipping as a fallback when the
standby has a longer down time.

Yours,
Laurenz Albe

In response to

Browse pgsql-admin by date

  From Date Subject
Next Message Günce Kaya 2017-10-05 10:44:59 Postgresql FDW - some difference between remote and target table structure
Previous Message Boris S. 2017-10-05 07:05:27 Re: High CPU usage from a vacuum process but does nothing