From owner-freebsd-scsi Mon Feb 11 12:16:22 2002 Delivered-To: freebsd-scsi@freebsd.org Received: from mail.libertysurf.net (mail.libertysurf.net [213.36.80.91]) by hub.freebsd.org (Postfix) with ESMTP id 47B2C37B402 for ; Mon, 11 Feb 2002 12:16:17 -0800 (PST) Received: from [192.168.1.129] (212.129.9.146) by mail.libertysurf.net (5.1.053) id 3C633FD4000B9298; Mon, 11 Feb 2002 21:12:29 +0100 Date: Sun, 10 Feb 2002 22:12:31 +0100 (CET) From: =?ISO-8859-1?Q?G=E9rard_Roudier?= X-X-Sender: To: Frank Altpeter Cc: Subject: Re: SCSI parity error detected In-Reply-To: <20020211170640.C245@ision.net> Message-ID: <20020210215543.P1982-100000@gerard> 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 List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Mon, 11 Feb 2002, Frank Altpeter wrote: > hello! > > After many hours of testing with a good friend we didn't work out any > solution for my problem - he recommended asking you, because you are the > cracks that wrote the driver :-)) > > Well... something about my hardware first... i'm running an x86 system, > powered with a gigabyte (GA-7VTXE+) board and an Athlon XP 1500+ CPU. > There are four SCSI devices connected via a Dawicontrol DC-2945U > hostadapter (symbios 53C875 chipset). > > The device list is as follows: > > murphy# camcontrol devlist -v > scbus0 on sym0 bus 0: > at scbus0 target 0 lun 0 (pass0,da0) > at scbus0 target 1 lun 0 (pass1) > at scbus0 target 2 lun 0 (pass2) > at scbus0 target 4 lun 0 (pass3,cd0) > < > at scbus0 target -1 lun -1 () > scbus-1 on xpt0 bus 0: > < > at scbus-1 target -1 lun -1 (xpt0) > > > Target 0 is an SCSI-2 harddisk, and the targets 1 and 2 are SCA disks, > plugged in with some SCA-SCSI-2 adapter. > > The main disk (target 0) is working: > > da0 at sym0 bus 0 target 0 lun 0 > da0: Fixed Direct Access SCSI-2 device > da0: 20.000MB/s transfers (20.000MHz, offset 15), Tagged Queueing Enabled > da0: 4357MB (8925000 512 byte sectors: 255H 63S/T 555C) > > But the two other disks show up these errors: > sym0: SCSI parity error detected: SCR1=3D1 DBC=3D11000000 SBCL=3D29 > sym0: SCSI parity error detected: SCR1=3D3 DBC=3D11000000 SBCL=3D29 > sym0: SCSI parity error detected: SCR1=3D1 DBC=3D11000000 SBCL=3D29 > sym0: SCSI parity error detected: SCR1=3D3 DBC=3D11000000 SBCL=3D29 > sym0: SCSI parity error detected: SCR1=3D1 DBC=3D11000000 SBCL=3D29 > sym0: SCSI parity error detected: SCR1=3D3 DBC=3D11000000 SBCL=3D29 > sym0: SCSI parity error detected: SCR1=3D1 DBC=3D11000000 SBCL=3D29 The PCI SCSI chip is reporting SCSI PARITY ERROR in DATA IN phase. > (da1:sym0:0:1:0): got CAM status 0x50 > (da1:sym0:0:1:0): fatal error, failed to attach to device > (da1:sym0:0:1:0): lost device > (da1:sym0:0:1:0): removing device entry > (da2:sym0:0:2:0): got CAM status 0x50 > (da2:sym0:0:2:0): fatal error, failed to attach to device > (da2:sym0:0:2:0): lost device > (da2:sym0:0:2:0): removing device entry > > Since i used to have an adaptec 1542cf before, i'm quite sure that the > disks and the SCA adapters are ok, because the system was running fine > with the old card (yes, i know it's a shame to run fast SCA disks with > this setting, but i got them really cheap :). The hardware combination was probably ok for the 1542, but it doesn't seem so for the 53c875. As you should|do know, there are major differences between the 2 controllers: - 1542 is 8 bit SCSI, 53c875 is 16 bit SCSI - 1542 is FAST-10, 53c875 is FAST-20. OTOH, the offending devices exhibit some *SUN* proprietary firmware. Given that SUN guys are known to be great reinventors of the wheel, we may suspect these devices to have some questionnable behaviour. Anyway, without a clear description of your SCSI BUS, I cannot really help you a lot. If we suppose that SUN guys didn't break SCSI too much here (??), the problem may be due to your SCA-to-SCSI2 tinking machines not connecting and/or terminating all signals as needed for WIDE SCSI. Another possibility could be that you are using a 8 bit SCSI BUS, but that your devices are claiming and accepting a wide negotiation. Etc..., etc... > After switching to the new motherboard, the old adaptec was unusable > because it was an ISA card (new board only supports PCI). > The CDROM was the last device in the chain and was terminating. I have > changed this now, and have terminated the chain with an active terminator > at the end of the SCSI cable. > > But, the errors keep showing up and the devices are not useable. > > Can you perhaps give me any hints on that problem? You may first check that, on paper, all SCSI BUS signals are properly, either connected, or terminated when they have to be, as expected by the 53c875 chip. G=E9rard. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message