Обсуждение: Hot Standby DB solution

Поиск
Список
Период
Сортировка

Hot Standby DB solution

От
"Nigel Bishop"
Дата:

Hello list,

 

RH 4

PG 8.0.3

 

I’m investigating the use of a second server with a backup copy of the live DB set up and in recovery mode.

 

I want to copy (scp?) the archived WAL files over and apply these to the standby DB.

 

Basically what I am after is any documentation on how to set this up – the backup/recovery documentation section mentions it but does not go into any detail.

 

I had a hunt through the archives and although some people appeared to be using this scenario, I couldn’t find any definitive solutions  

 

One of the things I do not fully understand is how to keep the standby DB in recovery mode.

 

Would 8.0.3 be OK for this or would I be better shifting to 8.1.1?

 

If anyone has any information on the above and is willing to share it, it would be much appreciated.

 

Thanks,

Nigel

 



Communications on or through ioko's computer systems may be monitored or recorded to secure effective system operation and for other lawful purposes.

Unless otherwise agreed expressly in writing, this communication is to be treated as confidential and the information in it may not be used or disclosed except for the purpose for which it has been sent. If you have reason to believe that you are not the intended recipient of this communication, please contact the sender immediately. No employee is authorised to conclude any binding agreement on behalf of ioko with another party by e-mail without prior express written confirmation.

ioko365 Ltd. VAT reg 656 2443 31. Reg no 3048367. All rights reserved.

Re: Hot Standby DB solution

От
"Qingqing Zhou"
Дата:
""Nigel Bishop"" <Nigel.Bishop@ioko.com> wrote

> I'm investigating the use of a second server with a backup copy of the
> live DB set up and in recovery mode.

You may want to check out the Slony I project:

http://gborg.postgresql.org/index.php

BTW, if you want "cold" standby (means if one server crashed, the other will
replay the WAL xlogs(takes seconds or hours, depends how much work needed to
be redo), and startup and take the job), then there is a much easier
solution of connect two machines on a shared-disk array.

Regards,
Qingqing