Re: TOAST versus VACUUM, or "missing chunk number 0 for toast value" identified

On Fri, Oct 28, 2011 at 3:11 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Alvaro Herrera <alvherre(at)commandprompt(dot)com> writes:
>> Excerpts from Alvaro Herrera's message of vie oct 28 16:47:13 -0300 2011:
>>> BTW we had previous discussions about dropping pg_database's toast
>>> table.  Maybe this is a good time to do it, even if there's no risk of
>>> this bug (or the hypothetical circularity detoasting problem) showing up
>>> there.
>
> No objection from me.
>
>> Oh, something unrelated I just remembered: we have
>> pg_database.datlastsysoid which seems unused.  Perhaps we should remove
>> that column for cleanliness.
>
> I have a vague recollection that some client-side code uses this to
> figure out what's the dividing line between user and system OIDs.
> pg_dump used to need to know that number, and it can still be useful
> with casts and other things that are hard to tell whether they're built-in.
> While in principle people could use FirstNormalObjectId instead, that
> could come back to bite us if we ever have to increase that constant
> in future releases.  I'm inclined to leave this one alone.

A cursory search of the archives turned up that pgadmin is explicitly
querying it out (or at least did in 2009). Also, there are some
fairly ancient references of discussion about using it via pg_dump to
deal with the 'not dumping user casts wrapping system objects' issue,
but that was deemed unreliable and is probably not relevant.

merlin