Date: Fri, 7 Mar 2003 11:10:39 +0100 From: Andre Albsmeier <andre.albsmeier@siemens.com> To: Andre Albsmeier <andre.albsmeier@siemens.com>, freebsd-stable@FreeBSD.ORG Cc: fcasadei@inwind.it Subject: Re: (da0:ahc0:0:0:0): Unexpected busfree in Data-in phase and other weirdness Message-ID: <20030307111039.A76900@curry.mchp.siemens.de> In-Reply-To: <20030307094353.GA860@goku.kasby>; from fcasadei@inwind.it on Fri, Mar 07, 2003 at 10:43:53AM %2B0100 References: <20030228235031.GE5033@grummit.biaix.org> <20030301094934.GA547@goku.kasby> <20030306153355.A12907@curry.mchp.siemens.de> <20030306225837.GA1277@goku.kasby> <20030307081541.B71305@curry.mchp.siemens.de> <20030307094353.GA860@goku.kasby>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 07-Mar-2003 at 10:43:53 +0100, Francesco Casadei wrote: > On Fri, Mar 07, 2003 at 08:15:41AM +0100, Andre Albsmeier wrote: > [snip] > > I am no expert in these things, but: I have never seen a bad Adaptec > > controller (and I used lots of them from 2940 to 39160). I _have_ seen > > many bad cables/terminators. I _have_ seen many bad cables/terminators > > that used to work without a glitch and suddenly started to become bad. > > The possibility that this happens is reciprocally proportional to the > > price you pay for them :-) > > > > I have SCSI systems that worked for years and started to spit out bus > > error messages when changes were commmitted to the driver. In most > > cases it turned out that these changes only triggered bad cable/ > > terminator errors or even firmware problems with the drive. > > > > Don't you have a friend who can give you a spare AHA2940 for > > testing? > > > > -Andre > > > [snip] > > end of the original message > > Yeah, it's hard to think about an Adaptec 2940 failing, but I'm just > guessing the cause of the error. > > I had ATA 'fallback to PIO mode' errors with three IBM 40G, and > I found that one of the three hd fans installed had failed. I removed it and > the the problem disappeared (system uptime is 31 days now). > Could it be an electrical problem due to a failing hd fan attached to the DNES > power cable? It could, but I don't think so. Why don't you measure the temp of the DNES with the following script. The DNES usually doesn't get very hot... #!/bin/sh shit () { if [ ! $sflag ]; then echo >&2 echo "$@" >&2 echo >&2 fi exit 1 } sflag="" if [ "x$1" = x-s ]; then sflag=1 shift fi for d in "$@"; do set -- `camcontrol cmd $d -c "4d 0 6f 0 0 0 0 0 0C 0" -i 12 "i1 *i1 i2 i2 *i1 i1 i1 i1 i1 i1"` [ "$1" != 47 ] && shit "Byte 0 is not 47" [ "$2" != 8 ] && shit "Bytes 2/3 are not 8" [ "$3" != 0 ] && shit "Bytes 4/5 are not 0" [ "$4" != 4 ] && shit "Byte 7 is not 4" if [ $sflag ]; then echo -n "$7 " continue fi echo echo "Device $d" echo "---------------------------------------" echo "SMART Sense Code Byte : $5" echo "SMART Alert Reason Code : $6" echo "Most Recent Temperature Reading : $7" echo "Vendor HDA Temperature Trip Point: $8" done echo To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030307111039.A76900>