From owner-freebsd-hackers Wed Jan 24 22:21:44 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id WAA09892 for hackers-outgoing; Wed, 24 Jan 1996 22:21:44 -0800 (PST) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id WAA09884 Wed, 24 Jan 1996 22:21:42 -0800 (PST) Message-Id: <199601250621.WAA09884@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: Jaye Mathisen cc: hackers@freebsd.org Subject: Re: Weird problem, 2940UW -stable, etc. In-reply-to: Your message of "Tue, 23 Jan 1996 20:49:05 PST." Date: Wed, 24 Jan 1996 22:21:42 -0800 From: "Justin T. Gibbs" Sender: owner-hackers@freebsd.org Precedence: bulk >The locks up when INN is running. So I popped a >different disk under /news, and ran fsck on the disk for 30 hours >straight with nary a glitch. Reboot with the RAID disk mounted as /news, >and it failed within an hour. (By fail, I mean I get messages about SCSI >timeout). Note that at no time was any hardware changed or moved, the >only changes here have been fstab changes. Why did you run fsck as your test? Fsck opens the raw device and in doing so may not exercise the same bug if it is a bug. >I have a comm port on the RAID controller, and hooked up a terminal to it >to watch, and I notice that for some reason, it's picking up a SCSI Bus >RESET, right in the middle of things, while the FS is active. Do you get the RESET after the timouts start to happen? Issuing a reset is part of the (very broken at the moment) error recovery code in the driver. If you are getting a reset during normal I/O activity, I'd say your RAID is broken. -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations ===========================================