Ron Johnson wrote:
>>All that we basically need for PITR is to provide management code that
>>lets old WAL segments get archived off to tape (or wherever) rather than
>>deleted, plus some kind of control that lets the roll-forward process be
>>stopped at the desired point-in-time rather than necessarily running to
>>the end of the available WAL data. This isn't a trivial amount of code,
>>but there's no great conceptual difficulty either.
>
>
> Hope everybody realizes that the amount of WALs will get very big
> on active-update systems...
Of course they will be recycled in some point of time or other. And even if
postgresql would provide PITR abilities, that would be nearly useless if WAL is
recycled.. Its a space/time tradeoff issue..
Shridhar
