Re: Review: Fix snapshot taking inconsistencies

Robert Haas <robertmhaas@gmail.com>

From: Robert Haas <robertmhaas@gmail.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Marko Tiikkaja <marko.tiikkaja@cs.helsinki.fi>, Steve Singer <ssinger_pg@sympatico.ca>, pgsql-hackers@postgresql.org, Alvaro Herrera <alvherre@commandprompt.com>
Date: 2011-02-28T19:14:59Z
Lists: pgsql-hackers
On Mon, Feb 28, 2011 at 2:01 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Robert Haas <robertmhaas@gmail.com> writes:
>> On Mon, Feb 28, 2011 at 1:45 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>> I'm afraid that the goals of this patch might be similarly obsolete.
>
>> No, I don't think so.  IIUC, the problem is that EXPLAIN ANALYZE runs
>> the rewrite products with different snapshot handling than the regular
>> execution path.
>
> Possibly, but it's not clear to me that this patch fixes that.
> As I said, it's no longer obvious what the patch means to do, and I'd
> like a clear statement of that.

Fair enough.  I assume Marko will provide that shortly.  I believe the
consensus was to make the regular case behave like EXPLAIN ANALYZE
rather than the other way around...

>> So in theory you could turn on auto_explain and have
>> the semantics of your queries change.  That would be Bad.
>
> That's just FUD.  auto_explain doesn't run EXPLAIN ANALYZE.

Oh, woops.  I stand corrected.  But I guess the query might behave
differently with and without EXPLAIN ANALYZE?

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