From owner-freebsd-scsi Thu Jan 18 12:43:40 2001 Delivered-To: freebsd-scsi@freebsd.org Received: from front3m.grolier.fr (front3m.grolier.fr [195.36.216.53]) by hub.freebsd.org (Postfix) with ESMTP id AEBB237B404 for ; Thu, 18 Jan 2001 12:43:21 -0800 (PST) Received: from nas1-107.cgy.club-internet.fr (nas1-107.cgy.club-internet.fr [195.36.197.107]) by front3m.grolier.fr (8.9.3/No_Relay+No_Spam_MGC990224) with ESMTP id VAA07787; Thu, 18 Jan 2001 21:43:18 +0100 (MET) Date: Thu, 18 Jan 2001 20:42:40 +0100 (CET) From: =?ISO-8859-1?Q?G=E9rard_Roudier?= X-Sender: groudier@linux.local To: Andrea Venturoli Cc: freebsd-scsi@FreeBSD.ORG Subject: Re: sym0 errors In-Reply-To: <200101181809.f0II92102460@relay2.flashnet.it> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, 18 Jan 2001, Andrea Venturoli wrote: > Hello.=20 > =20 > What do you make of this error:=20 > =20 > sym0:1 ERROR (0:18) (1-21-0) (1f/9f) (scripta 5f8:11000000).=20 ^ This bit means SCSI GROSS ERROR. It is generally a SCSI ACK/REQ protocol problem detected by the controller SCSI core. A SCSI BUS that hasn't margin enough for the transfer speed used is=20 likely the cause. > sym0: script cmd =3D 11000000=20 > sym0: regdump: da 10 c0 9f 47 1f 01 03 76 01 81 21 80 01 01 09 00 ce 1e 0= 1 08 ff ff ff.=20 > (noperiph:sym0:0:-1:-1): SCSI BUS reset detected.=20 > =20 > then, some=20 > =20 > sym0: unexpected disconnect=20 This one means that the device (target) disconnected the SCSI BUS when this wasn't expected by the initiator. This is the way target signals problems that cannot be reported using the normal SCSI protocol. A SCSI BUS problem that affects handshake (ACK/REQ) can be reported so by the target, for example. =20 > This happens on a FreeBSD 4.2R system with a Tekram DC390U2W with an Ultr= aPlex 40x on the =20 > narrow bus and two Atlas 10K2 on the LVD bus.=20 > Termination is correct.=20 A SCSI BUS including all devices, cables, connectors, etc.., is something complex. The breakage can simply not be visible. I suggest you to decrease the synchronous speed of your devices to half the values currently used. If the problem disappears or gets an order of magnitude less frequent, then a SCSI BUS problem is very likely the cause. (man camcontrol, or set device settings in NVRAM accordingly) Regards, G=E9rard. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message