From owner-freebsd-hackers Thu Mar 9 7:22: 0 2000 Delivered-To: freebsd-hackers@freebsd.org Received: from aaz.links.ru (aaz.links.ru [193.125.152.37]) by hub.freebsd.org (Postfix) with ESMTP id 8AD0837B69C for ; Thu, 9 Mar 2000 07:21:50 -0800 (PST) (envelope-from babolo@links.ru) Received: (from babolo@localhost) by aaz.links.ru (8.9.3/8.9.3) id SAA06611; Thu, 9 Mar 2000 18:27:16 +0300 (MSK) Message-Id: <200003091527.SAA06611@aaz.links.ru> Subject: Re: ahc0 errors In-Reply-To: <20000309141136.A23000@sindigit.pt> from "[Jos_ Monteiro]" at "Mar 9, 0 02:11:37 pm" To: jm@sindigit.pt Date: Thu, 9 Mar 2000 18:27:16 +0300 (MSK) Cc: freebsd-hackers@FreeBSD.ORG From: "Aleksandr A.Babaylov" MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG [Jos_ Monteiro] writes: > i've been experiencing several problems with an adaptec 2940u2w. > > the disk and controller are detected as: > > ahc0: rev 0x00 int a irq 11 on pci0.17.0 > ahc0: aic7890/91 Wide Channel A, SCSI Id=7, 16/255 SCBs > da1 at ahc0 bus 0 target 3 lun 0 > da1: Fixed Direct Access SCSI-3 device > da1: 80.000MB/s transfers (40.000MHz, offset 31, 16bit), Tagged Queueing Enabled > da1: 8761MB (17942584 512 byte sectors: 255H 63S/T 1116C) > > the kernel messages invariably are: > > proxy /kernel: ahc0: Data Parity Error Detected during address or write data phase > proxy /kernel: ahc0: Received a Master Abort > proxy /kernel: swap_pager: indefinite wait buffer: device: 0x20409, blkno: 7488, size: 4096 > proxy /kernel: swap_pager: indefinite wait buffer: device: 0x20409, blkno: 7488, size: 4096 > proxy /kernel: (da1:ahc0:0:3:0): SCB 0x1d - timed out in dataout phase, SEQADDR == 0x5d > proxy /kernel: (da1:ahc0:0:3:0): BDR message in message buffer > proxy /kernel: (da1:ahc0:0:3:0): no longer in timeout, status = 353 > proxy /kernel: Unexpected busfree. LASTPHASE == 0xa0 > proxy /kernel: SEQADDR == 0x163 > > i already tried: > > - added extra fan near the drive > - shortened the cable distance between the disk and the controller, > trying to avoid near sources of elecromagnetic interference > - lowered the transfer rate on the u2w bus to 40 and 20MB/s > - connected the disk to the uw bus and forced the bus termination on the > disk itself > > my crash test has been cp'ing a 200M file to another dir in the same > filesystem, although these errors usually appear randomly a couple of times per > hour. > > i would like to have a definitive opinion about this before turning to > the hardware supplier. I have the same problem with 2940UW controller and 9G Seagate UW disk FreeBSD are 3.3 and 3.4. High activity of several disk transfers in parallel is more dangerous than one transfer. Soft updates is more dangerous then sync mount. Can you try copy big file (600 Mb for examle) to your disk and then cmp copy with original? I have usually about 3 regions about 10 bytes long difference for each CDROM image when I use SCSI to store copy. Every copy has different difference (:-) -- @BABOLO http://links.ru/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message