From owner-freebsd-stable Mon Nov 27 11: 3:18 2000 Delivered-To: freebsd-stable@freebsd.org Received: from smtp1.dti.net (smtp1.dti.net [206.252.128.177]) by hub.freebsd.org (Postfix) with ESMTP id 26AB137B479 for ; Mon, 27 Nov 2000 11:03:15 -0800 (PST) Received: from dti.net (ynp.noc.dti.net [206.252.134.43]) by smtp1.dti.net (8.11.1/8.11.1/DTI) with ESMTP id eARJ39g97413 for ; Mon, 27 Nov 2000 14:03:09 -0500 (EST) Message-ID: <3A22AFED.80DFE1DA@dti.net> Date: Mon, 27 Nov 2000 14:03:09 -0500 From: Youri Podchosov Organization: Digital Telemedia, Inc. X-Mailer: Mozilla 4.75 [en] (X11; U; SunOS 5.7 sun4u) X-Accept-Language: en, ru, uk MIME-Version: 1.0 To: freebsd-stable@FreeBSD.ORG Subject: Re: Problems with ahc driver? References: <5.0.0.25.2.20001127102321.0469da20@sung.org> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Christian Sung wrote: > > Hi, > > I've been consistently experiencing a problem with one of our FreeBSD boxes > (the only one running a SCSI subsystem with an embedded Adaptec Ultra2 SCSI > controller): > > ahc0: port 0xe400-0xe4ff mem > 0xfebfe000-0xfebfefff irq 16 at device 11.0 on pci0 > ahc1: port 0xe800-0xe8ff mem > 0xfebff000-0xfebfffff irq 16 at device 11.1 on pci0 > > About once a week, the following message is logged in the system's messages > file, and the box stops responding altogether: > > (da0:ahc0:0:0:0): Invalidating pack > > I started experiencing this problem after migrating to 4.1. I then > upgraded to 4.1.1 and later to 4.2-STABLE hoping it was related to minor > problems with the ahc driver that might have been fixed in the later > releases, but unfortunately the problem persists. > > Is there anyone out there experiencing the same problem? Anyone that can > shed some light on what's happening here? As I mentioned, the problem > manifests itself mostly after the box has been up for a few days -- no real > "activity" to speak of, the machine is a router/firewall server. I've been suffering from this problem with all 4.1+ RELEASEs/STABLEs :-( My AHA is plain old 2940UW (BIOS v2.57.2) and it drives two UW Barracudas and Sony SDT9000 tape. The symptoms are just as you put it: after a while, no matter how heavy or light the SCSI bus activity is, all ends up with a dropped disk and a need to reboot (and recover the system, which is much worse). I found a cure for myself, however: I never do soft reboots any more, only through reset button. My understanding is that plain halt/reboot leaves the HA in some weird state that prevents normal operations in the future. Reset (or power cycling) cleans things up. At least this is how it works for me. (Because this is a critical system, and experimenting with failed SCSI operations usually takes a lot of time even to bring the system back to life, I can't afford any comprehensive investigation of this problem and just live with it as long as my workaround allows it to work flawlessly). -- LLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL L Youri N. Podchosov (ynp) /// Email: ynp@ynp.net L L Senior NOC Engineer ())) Web: http://www.ynp.net L L Digital Telemedia, Inc. /// B:212-625-5365 H:718-680-9024 L LLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message