From owner-freebsd-scsi Wed Mar 31 10:51:10 1999 Delivered-To: freebsd-scsi@freebsd.org Received: from server.cirdan.iae.nl (server.cirdan.iae.nl [212.61.28.34]) by hub.freebsd.org (Postfix) with ESMTP id 7D15B14E4B for ; Wed, 31 Mar 1999 10:50:54 -0800 (PST) (envelope-from graaf@cirdan.iae.nl) Received: from cirdan.iae.nl (cirdan.iae.nl [212.61.28.35]) by server.cirdan.iae.nl (Postfix) with ESMTP id 8DEBCD98A; Wed, 31 Mar 1999 21:38:04 +0200 (CEST) Date: Wed, 31 Mar 1999 20:54:53 +0200 (CEST) From: Edwin de Graaf Reply-To: graaf@iae.nl Subject: Re: CAM timeout in dataout phase (3.1-STABLE) To: gibbs@narnia.plutotech.com Cc: admin@iae.nl, scsi@FreeBSD.org In-Reply-To: <199903311605.JAA11461@narnia.plutotech.com> MIME-Version: 1.0 Content-Type: TEXT/plain; CHARSET=US-ASCII Message-Id: <19990331193807.8DEBCD98A@server.cirdan.iae.nl> Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi Justin, Thanks for your quick response! On 31 Mar, Justin T. Gibbs wrote: > In article <19990331075950.7691794F1@surf.iae.nl> you wrote: >> DTC RAID controller boot: >> ------------------------- >> da1 at ahc0 bus 0 target 1 lun 0 >> da1: Fixed Direct Access SCSI-2 device >> da1: 40.000MB/s transfers (20.000MHz, offset 16, 16bit), Tagged Queueing Enabled >> da1: 25500MB (52224000 512 byte sectors: 255H 63S/T 3250C) > > Were the number of tags ever reduced? You'll probably have to boot > with '-v' to know for sure as Jordan moved the message that indicates > when this occurs behind "if (bootverbose)". We will run with -v the next time we reboot the system to check this. >> Timeout error messages: >> ----------------------- >> Mar 30 10:24:30 iaehv /kernel: (da1:ahc0:0:1:0): SCB 0x2a - timed out >> in dataoutphase, SEQADDR == 0x5d Mar 30 10:24:30 iaehv /kernel: >> (da1:ahc0:0:1:0): BDR message in message buffer Mar 30 10:24:31 iaehv >> /kernel: (da1:ahc0:0:1:0): SCB 0x2a - timed out in dataoutphase, SEQADDR >> == 0x5d Mar 30 10:24:31 iaehv /kernel: (da1:ahc0:0:1:0): no longer in >> timeout, status =34b Mar 30 10:24:31 iaehv /kernel: ahc0: Issued Channel >> A Bus Reset. 255 SCBs aborted > > Perhaps the controller is getting overloaded by too many overlapped > commands. It could also be that your cabling or termination is > inadequate causing a REQ or ACK to be lost during some data transfers. > Are you using 'Forced Perfect' terminators? Is your cable < 1.5m in > length? There are two SCSI connectors on the motherboard, one wide SCSI and one narrow connector. We have the boot disk on the narrow connector with an active terminator on the cable. The DTC RAID controller is connected to the wide connector; there is also an active terminator on the SCSI cable with the RAID disks. I am not entirely clear if these are two separate SCSI busses, or it it is one bus with wide and narrow devices. I think it is one SCSI bus, and we have disabled termination on the Adaptec controller. The cable length of the two cables combined might be a little over 1.5m Again thanks for your help. Best regards, Edwin de Graaf -- "O Oysters, come and walk with us!" The Walrus did beseech. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message