Re: Avoiding unnecessary reads in recovery

Heikki Linnakangas wrote:
> While working on this, this comment in ReadBuffer caught my eye:
>
>> /*
>> * During WAL recovery, the first access to any data page should
>> * overwrite the whole page from the WAL; so a clobbered page
>> * header is not reason to fail. Hence, when InRecovery we may
>> * always act as though zero_damaged_pages is ON.
>> */
>> if (zero_damaged_pages || InRecovery)
>> {
>
> But that assumption only holds if full_page_writes is enabled, right? I
> changed that in the attached patch as well, but if it isn't accepted
> that part of it should still be applied, I think.

On second thought, my fix still isn't 100% right because one could turn
full_page_writes on before starting replay.

--
Heikki Linnakangas
EnterpriseDB http://www.enterprisedb.com