From: | Nickolay <nitro(at)zhukcity(dot)ru> |
---|---|
To: | pgsql-performance(at)postgresql(dot)org |
Subject: | Re: Per-database warm standby? |
Date: | 2009-08-15 08:11:14 |
Message-ID: | 4A866DA2.6020408@zhukcity.ru |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
I also have a question about warm standby replication.
What'd be the best solution for the system with 2 db servers (nodes), 1
database and 10 seconds max to switch between them (ready to switch time).
Currently I'm using Slony, but it's kind of slow when doing subscribe
after failover on the failed node (database can be really huge and it
would take a few hours to COPY tables using Slony).
May be WAL replication would be better?
Best regards, Nick.
Tom Lane wrote:
> Craig James <craig_james(at)emolecules(dot)com> writes:
>
>> 8.4 has vastly improved the warm-standby features, but it looks to me like this is still an installation-wide backup, not a per-database backup. That is, if you have (say) a couple hundred databases, and you only want warm-backup on one of them, you can't do it (except using other solutions like Slony). Is that right?
>>
>
> Correct, and that's always going to be true of any WAL-based solution.
>
> regards, tom lane
>
From | Date | Subject | |
---|---|---|---|
Next Message | Jeremy Carroll | 2009-08-15 14:25:45 | Re: Memory reporting on CentOS Linux |
Previous Message | Tom Lane | 2009-08-14 23:21:58 | Re: Scalability in postgres |