Re: [GENERAL] Re: [ADMIN] Would like to below scenario is possiblefor getting page/block corruption - Mailing list pgsql-admin

From Michael Paquier
Subject Re: [GENERAL] Re: [ADMIN] Would like to below scenario is possiblefor getting page/block corruption
Date
Msg-id CAB7nPqQ64hjZZ4ywNNmbbpcQ9TKaF=R5hzMJGdL19PXNdSv-Lw@mail.gmail.com
Whole thread Raw
In response to Re: [ADMIN] Would like to below scenario is possible for gettingpage/block corruption  (Sreekanth Palluru <sree4pg@gmail.com>)
List pgsql-admin
On Sun, Dec 11, 2016 at 12:00 PM, Sreekanth Palluru <sree4pg@gmail.com> wrote:
> I am looking at possibility of PG introducing corruption if relation extends and before it updates new page with
pageheaderin memory and crash happens? 
>
> Is this possible?

No.

> Does PG updates pageheader when relation get extends?

You need to look at smgrextend() when extension an on-disk relation
file. The page is written in a correct shape.

> If so what details it writes? Or will it be null?
--
Michael


pgsql-admin by date:

Previous
From: Sreekanth Palluru
Date:
Subject: Re: [ADMIN] Would like to below scenario is possible for gettingpage/block corruption
Next
From: Vladimir Borodin
Date:
Subject: Re: [ADMIN] switching replication from standard asynch to slots