Re: pg_ugprade and missing timeline

From: Alexey Klyukin <alexk(at)hintbits(dot)com>
To: Kelley Reynolds <kelley(dot)reynolds(at)rubyscale(dot)com>
Cc: "pgsql-admin(at)postgresql(dot)org" <pgsql-admin(at)postgresql(dot)org>
Subject: Re: pg_ugprade and missing timeline
Date: 2015-03-24 00:17:07
Message-ID: CAAS3tyJ5b-CiwdJ-Lx=8rSgV_bER4AAJeZQ463GqgvcMDKNmBA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin

Hi,

On Thursday, March 19, 2015, Kelley Reynolds <kelley(dot)reynolds(at)rubyscale(dot)com>
wrote:

> I have just been bitten by the same pg_upgrade/timeline bug that you
> experienced in this thread:
>
>
> http://www.postgresql.org/message-id/CAAS3tyKOvC-5dRqQLj+T=tuHmrsVetBznFOwRgvLFYuzjtE90w@mail.gmail.com
>
> I read your description about how you manually placed a timeline file on
> the new master but I can't quite figure out exactly what should go in
> there, I have no existing timeline files and I have no idea what one looks
> like. Can you give me some more details on how to create that file to get
> streaming replication working again?

I think the fastest way to get a timeline file is to initialize a new
PostgreSQL cluster and promote it afterwards. Make sure you have archiving
turned on and look at the archive destination after promotion. You will get
a new .history file. Just copy this file to your master, renaming it to
the timeline number the replica requires.

This is rather a hack than a proper solution though...

Regards,
Alexey

--
Regards,
Alexey Klyukin

Browse pgsql-admin by date

  From Date Subject
Next Message Ruth Melendo 2015-03-24 09:33:36 Re: How to desconfigure BDR?
Previous Message Vladimir Borodin 2015-03-23 14:45:12 Re: [GENERAL] [pgadmin-support] Issue with a hanging apply process on the replica db after vacuum works on primary