Re: crash-safe visibility map, take three

On Tue, Nov 30, 2010 at 10:38 AM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com> writes:
>> On 30.11.2010 06:57, Robert Haas wrote:
>>> I can't say I'm totally in love with any of these designs.  Anyone
>>> else have any ideas, or any opinions about which one is best?
>
>> Well, the design I've been pondering goes like this:
>
> Wouldn't it be easier and more robust to just consider VM bit changes to
> be part of the WAL-logged actions?  That would include updating LSNs on
> VM pages and flushing VM pages to disk during checkpoint based on their
> LSN values.  All of these other schemes seem too complicated and not
> provably correct.

What WAL-logged actions?

The problem case is where a page has no tuples or line pointers that
need to be removed, and all we need to do is mark it all-visible. We
don't current WAL-log anything in that case.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company