Re: Re: xReader, double-effort (was: Temporary tables under hot standby)

On Sat, Apr 28, 2012 at 4:43 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov> writes:
>> Tom Lane  wrote:
>>> Simon Riggs  writes:
>>>> Translating WAL is a very hard task.
>>> No kidding. I would think it's impossible on its face.
>
>> Well, the xReader approach (Aakash's GSoC project) is to serve as a
>> proxy for a WAL stream going to a hot standby, to interpret each
>> incoming WAL record by "cheating" and querying the HS before passing
>> the WAL along, and then using sync rep to the HS regardless of
>> whether the incoming stream is sync or async.  What reliability
>> problem do you see with this approach?
>
> Well, first off, it requires a replica that *is* a physical replica
> (else it won't have the same table OIDs, for instance).

It's even harder than that because the physical replica needs to be
able to be shutdown/restarted. We currently restart WAL from last
checkpoint, but if changes have already been made and committed then
the catalog will be incorrect when we replay the WAL again. So, I
think saying very hard is justified.

We basically have a choice of shipping full metadata with every change
or finding some way to avoid doing that. Investigating the latter
seems like a worthwhile task but there are limits, as you say.

--
 Simon Riggs                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services