Re: Error code for "terminating connection due to conflict with recovery"

Simon Riggs <simon@2ndquadrant.com>

From: Simon Riggs <simon@2ndQuadrant.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Magnus Hagander <magnus@hagander.net>, Robert Haas <robertmhaas@gmail.com>, Kevin Grittner <Kevin.Grittner@wicourts.gov>, Bruce Momjian <bruce@momjian.us>, fgp@phlo.org, Tatsuo Ishii <ishii@postgresql.org>, pgsql-hackers@postgresql.org
Date: 2011-02-01T20:25:35Z
Lists: pgsql-hackers
On Tue, 2011-02-01 at 11:43 -0500, Tom Lane wrote:
> Simon Riggs <simon@2ndQuadrant.com> writes:
> > Here's the basic patch, will work on the refactoring if no objections.
> 
> ResolveRecoveryConflictWithDatabase works when you're not in recovery?
> That seems pretty fragile at best.  In any case, this is a 9.2 feature
> at the earliest, please do not expect people to spend time on it now.

Err, no. Sorry if that wasn't clear. 

The patch works but requires refactoring, which I referred to above.
What is now called ResolveRecoveryConflictWithDatabase() would be
changed to a name that makes sense in both normal running and recovery.
That seemed so obvious that I wouldn't be leaving it like that, so I
just referred to that indirectly, which was confusing.

That allows me to refactor out the code that we just spent too long
discussing, so I was hoping to do that in this release. I wasn't
expecting anybody else to spend time on it since it is simple and clear.

I'll add it to 9.2.

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