Re: Parameterized paths vs index clauses extracted from OR clauses

On 03/05/2013 12:33 AM, Robert Haas wrote:
> On Thu, Feb 28, 2013 at 3:03 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> Whichever way we go, the resulting patch is likely to be too large and
>> invasive for me to feel terribly comfortable about back-patching it into
>> 9.2. AFAICT this issue only arises for indexquals extracted out of
>> larger OR conditions, so maybe it's not worth taking such a risk for.
> EnterpriseDB has received a number of complaints from our customers
> resulting from planner behavior changes which were back-patched; so I
> am not sanguine about back-patching unless the situation is pretty
> darn dire and the fix is pretty darn near certain to be an improvement
> in every case. As we have discussed many times here and on
> pgsql-performance, DBAs seem to prefer a query plan that's the same
> every time, even if it's bad. Updating to get a security fix and
> having plans change under you proves to be an unpleasant surprise.
It also weakens our argument (per versioning policy) that patch releases
are *totally safe* and are something you can apply without extensive
testing and QA per http://www.postgresql.org/support/versioning/ .

I have enough trouble getting people to update as it is.

I'd be interested in being able to turn on fixes/enhancements, but
having behaviour changes that aren't clear wins in all cases turned on
by default seems likely to weaken trust from users who're already way
too leery of updating.

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