Return-Path: postman 
Delivery-Date: Sun, 19 Sep 93 19:28:38 PDT
Return-Path: postman
Received: by postgres.Berkeley.EDU (5.61/1.29)
	id AA14578; Sun, 19 Sep 93 19:24:24 -0700
Resent-From: postman (POSTGRES mailing list)
Resent-Message-Id: <9309200224.AA14578@postgres.Berkeley.EDU>
Sender: owner-postman@postgres.Berkeley.EDU
X-Return-Path: jean@gso.SAIC.COM
Received: from [139.121.25.1] by postgres.Berkeley.EDU (5.61/1.29)
	id AA14570; Sun, 19 Sep 93 19:24:17 -0700
Received: from norns.css.gov ([192.33.134.10]) by gso.SAIC.COM (4.1/SMI-4.1)
	id AA03157; Sun, 19 Sep 93 19:27:22 PDT
Received: by norns.css.gov (4.1/SMI-4.1)
	id AA01765; Sun, 19 Sep 93 19:25:00 PDT
Date: Sun, 19 Sep 93 19:25:00 PDT
From: jean@gso.SAIC.COM (Jean Anderson)
Message-Id: <9309200225.AA01765@norns.css.gov>
To: postgres@postgres.berkeley.edu
Subject: Re: more survey summary
Resent-To: postgres-dist
Resent-Date: Sun, 19 Sep 93 19:24:22 PDT

aoki@postgres.Berkeley.EDU (Paul M. Aoki) writes:
> hmm.  i haven't been able to actually crash [core] the postmaster 
> since it was rewritten for 4.1 (not that it's hard to crash the 
> backend).

I wasn't able to get the postmaster to crash under 4.1 until I was moved 
from my old trusty Sun 4/110 onto a Sparc IPX. Concurrency tests that
had "passed" on the 4/110 failed spectacularly on the IPX.  I couldn't 
even get two processes running concurrently reliably. After some nosing 
about, I realized that the Postgres shared memory kernel mods had not 
made it onto the new machine.

It never hurts to check the hardware....

 -jean
  jean@gso.saic.com