From owner-freebsd-bugs Sun Jul 6 09:20:05 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA26970 for bugs-outgoing; Sun, 6 Jul 1997 09:20:05 -0700 (PDT) Received: (from gnats@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA26956; Sun, 6 Jul 1997 09:20:01 -0700 (PDT) Date: Sun, 6 Jul 1997 09:20:01 -0700 (PDT) Message-Id: <199707061620.JAA26956@hub.freebsd.org> To: freebsd-bugs Cc: From: Justin Gibbs Subject: Re: kern/4039: 2940UW, IBM DCAS 32160 -- hungs if ultra speed (40 MB/sec) selected Reply-To: Justin Gibbs Sender: owner-bugs@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk The following reply was made to PR kern/4039; it has been noted by GNATS. From: Justin Gibbs To: news@FreeBSD.ORG Cc: FreeBSD-gnats-submit@FreeBSD.ORG Subject: Re: kern/4039: 2940UW, IBM DCAS 32160 -- hungs if ultra speed (40 MB/sec) selected Date: Sun, 06 Jul 1997 10:12:51 -0600 > doing a newfs on the second drive while doing I/O will > cause a system hung (no response, no reboot), if > 40 MB/sec was selected as IO-speed for each SCSI drive > in the adaptec BIOS. Will hung not always, but very often. This sounds very much like the buffer deadlock problem that was fixed recently in current by Bruce. My guess is that you are newfsing the block device instead of the raw device (which you shouldn't be doing anyway) and that the faster I/O speed increases the likelyhood of hitting the deadlock. If you can still reproduce this bug while performing the newfs on the raw partition, let me know. -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations ===========================================