From owner-freebsd-current Fri Mar 10 10:57:40 2000 Delivered-To: freebsd-current@freebsd.org Received: from mass.cdrom.com (castles549.castles.com [208.214.165.113]) by hub.freebsd.org (Postfix) with ESMTP id 5B80537BA57 for ; Fri, 10 Mar 2000 10:57:35 -0800 (PST) (envelope-from msmith@mass.cdrom.com) Received: from mass.cdrom.com (localhost [127.0.0.1]) by mass.cdrom.com (8.9.3/8.9.3) with ESMTP id KAA02844; Fri, 10 Mar 2000 10:57:37 -0800 (PST) (envelope-from msmith@mass.cdrom.com) Message-Id: <200003101857.KAA02844@mass.cdrom.com> X-Mailer: exmh version 2.1.1 10/15/1999 To: "O. Hartmann" Cc: freebsd-current@freebsd.org Subject: Re: RAID 5 Failure dumps all terminals In-reply-to: Your message of "Fri, 10 Mar 2000 15:37:09 +0100." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 10 Mar 2000 10:57:37 -0800 From: Mike Smith Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > We use A Mylex DAC960PD RAID 5 controller, attached to 7 > harddrives. One of the drives have had permanent failures, so > we changed it. Well, as I understand RAID 5, rebuilding the defective > drive should not involve the system, but On console (it ttyv0) I get > a message like "mlx0: error reading message log - invalid log entry request". The controller has told us that there's a message waiting for us, but when we go to get the message, it claims that it doesn't exist. The driver doesn't handle being lied to very well; you're probably getting a storm of these messages based on your followups. I'll take a look at this and see if I can reproduce it. > On all other ttyv* a logged in user (root) gets the same message > from kernel. Is this eloquence of the kernel and the mlx-driver a preliminary > feature or is the code still buggy, instable or anything else? I feel a little bit > confused about that, while other systems using RAID 5 are completely quiet > while rebuilding the defective and changed device. I have no idea what "other systems" you're talking about. I consider it prudent to log hardware failures, rather than letting them slip silently by. > On the other hand, how can I redirect console output to another console, on which > nobody never can log in (it should not be ttyv0)? How to prevent users and root > on some ttyv* from getting that much error messages, except on a bunch of > declared ttys? You would want to read the syslogd manpage, which will make this very standard part of your system's configuration somewhat clearer. -- \\ Give a man a fish, and you feed him for a day. \\ Mike Smith \\ Tell him he should learn how to fish himself, \\ msmith@freebsd.org \\ and he'll hate you for a lifetime. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message