From owner-freebsd-hackers  Sat Feb 24 15:47:44 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id PAA27825
          for hackers-outgoing; Sat, 24 Feb 1996 15:47:44 -0800 (PST)
Received: from freebsd.netcom.com (freebsd.netcom.com [198.211.79.3])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id PAA27819
          for <hackers@freebsd.org>; Sat, 24 Feb 1996 15:47:42 -0800 (PST)
Received: by freebsd.netcom.com (8.6.12/SMI-4.1)
	id RAA07014; Sat, 24 Feb 1996 17:51:36 -0600
From: bugs@freebsd.netcom.com (Mark Hittinger)
Message-Id: <199602242351.RAA07014@freebsd.netcom.com>
Subject: re: processes won't die (fwd)
To: hackers@freebsd.org
Date: Sat, 24 Feb 1996 17:51:35 -0600 (CST)
X-Mailer: ELM [version 2.4 PL25]
Content-Type: text
Sender: owner-hackers@freebsd.org
Precedence: bulk

> I wrote:
> > Can you reconfigure your system so that it does not use swap space on the
> > scsi drive?  I noted in your configuration that you had some ide drives
> > with swap space on them.  If you need an additional swap file on the ide
> > drive to handle the load you can do it with vfs.
> > 
> > It would be helpful to me if you could operate for awhile without scsi
> > drive swap space and see if the "hung ps" etc problem goes away.
>
> From: Nate Williams <nate@sri.MT.net>
> 
> Does this problem exist w/current or stable?  I'm running a 2.1R box
> with swap on both under sometimes heavy loads with no problems.
> 

I see the problem under both current and stable.  I'm starting to suspect
that there is a weird swap disk i/o transaction that is screwing up in the
scsi subsystem, and that the same kind of transaction doesn't happen in
the ufs world.

What are your scsi drives and what is your controller?  How much swap space
do you have on your system and how much memory do you have?

Regards,

Mark Hittinger
Netcom/Dallas
bugs@freebsd.netcom.com