FW: MaxQuery Plan

From: Dirk Elmendorf <delmendo(at)cymitar(dot)net>
To: pgsql-hackers(at)postgresql(dot)org
Subject: FW: MaxQuery Plan
Date: 1998-07-15 03:02:24
Message-ID: XFMail.980714220224.delmendo@cymitar.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

---

Date: Tue, 14 Jul 1998 21:23:42 -0500 (CDT)
From: Dirk Elmendorf <delmendo(at)cymitar(dot)net>
To: pgsql-hacers(at)postgresql(dot)org
Subject: MaxQuery Plan

I posed this question to Bruce Momjian earlier this week. He was unable to give me an answ
er so I was hoping to find enlightenment here.

I recently upgraded to 6.3.2 (after a number of fixes) on Red Hat 4.2 from 6.2/6.1 .
A number of the views I have seem to violate the new reduced QueryPlan Max- This has rende
red one of my databases un-usable.

Out of frustration I replaced the occurance of 8192 with 16384 (8k vs 16k) I wasn't able t
o determine which parts handled just the standard database tuples and which parts handled
the query plans themselves. It compiled and stopped complaining about my views.

My question is:
1. Does anyone know what the implecations of doubling the max tuple are ? (besides increas
ing the amount of memory used by postgres)
2. Is there another more efficient way to achieve the same ends?

Bruce told me that this problem would be fixed in 6.4 , but that's months away-I waited fo
r a long time for 6.3 to stablize (Even though I had some problems still - which Bruce tha
nkfully resolved) I need a number of the bug fixes and features of 6.3 - so staying put w
ith 6.2/6.1 isn't very tenable either. Any insight would greatly appreciated.

_________________________________________________________
Dirk Elmendorf, VP/Development Main: 210-892-4000
Cymitar Technology Group, Inc. Direct: 210-892-4005
Lorene Office Plaza Fax: 210-892-4329
9828 Lorene Lane <http://www.cymitar.com>
San Antonio, TX 78216-4450 <delmendo(at)cymitar(dot)com>
_________________________________________________________

Browse pgsql-hackers by date

  From Date Subject
Next Message Vadim Mikheev 1998-07-15 03:15:09 Re: [HACKERS] "internal error" triggered by EXISTS()
Previous Message Bruce Momjian 1998-07-15 02:45:30 Re: [HACKERS] "internal error" triggered by EXISTS()