Re: Re: BUG #5602: Recovering from Hot-Standby file backup leads to the currupted indexes

Hi,

Actually full_page_write being turned off on the master is probably a
problem.

-- Valentine

On Thu, Aug 12, 2010 at 9:43 AM, Fujii Masao <masao(dot)fujii(at)gmail(dot)com> wrote:

> On Thu, Aug 12, 2010 at 4:18 PM, Simon Riggs <simon(at)2ndquadrant(dot)com>
> wrote:
> > The safest approach is to
> >
> > 1. run pg_start_backup() on master, remember LSN
> > 2. copy backup_label from master to standby
> > 3. wait for starting LSN to be applied on standby
>
> ISTM we should wait for the latest checkpoint redo location to reach
> the starting location which pg_start_backup() returns.
>
> > That ensures we don't run without full page writes during backup and
> > that we have an explicit consistency point to work from.
>
> Yeah, the current documented procedure might not work when
> full_page_writes is off on the master. This should be documented.
>
> Regards,
>
> --
> Fujii Masao
> NIPPON TELEGRAPH AND TELEPHONE CORPORATION
> NTT Open Source Software Center
>