Date: Thu, 20 Sep 2001 20:00:32 +0100 From: ian j hart <ianjhart@ntlworld.com> To: tenebrae_BSD@niceboots.com Cc: freebsd-stable@FreeBSD.ORG Subject: Re: ad0: WRITE command timeout Message-ID: <3BAA3CD0.D23EBA1D@ntlworld.com> References: <Pine.BSF.4.21.0109200853250.81750-100000@steeltoe.niceboots.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Tenebrae wrote: > > I found a couple messages in the freebsd-stable archives about this error, > but no real resolution... > I'm using freeBSD 4.3-STABLE (uname -a below). > FreeBSD steeltoe.niceboots.com 4.3-STABLE FreeBSD 4.3-STABLE #0: Thu May > 17 22:58:45 PDT 2001 root@steeltoe.niceboots.com:/usr/src/sys/compile/GENERIC i386 > > I occasionally get a dmesg that says: > > ad0: WRITE command timeout tag=0 serv=0 - resetting > ata0: resetting devices .. done > > I'm not sure what's causing it. > One of the messages in the archives said to do this and send the results: > > sysctl -a hw.atamodes -a means show all, you didn't need this > > Well, the results were huge, but this part appears to be the relevant > part: > > hw.machine: i386 > hw.model: AMD-K6(tm) 3D processor > hw.ncpu: 1 > hw.byteorder: 1234 > hw.physmem: 130170880 > hw.usermem: 105263104 > hw.pagesize: 4096 > hw.floatingpoint: 1 > hw.machine_arch: i386 > hw.ata.ata_dma: 1 > hw.ata.wc: 0 > hw.ata.tags: 0 > hw.ata.atapi_dma: 0 > hw.atamodes: dma,---,pio,---, > hw.availpages: 31613 > > One guy said, "I had a similar problem with an old chipset which > did not support dma, so the controller (ata0) made a fallback to pio." > I'm afraid I'm a newbie tard and don't quite understand that and don't > know what FM to R. man ata > I have an ASUS P5A motherboard that's supposed to support UDMA 33. From > ASUS's web site: IIRC this is an ALI chipset. I'm sure someone will tell you if there are any known problems. Anyway, goto: http://www.storage.ibm.com/hdd/support/download.htm Get the IBM feature tool. This should let you set the UDMA mode of the HDD to match the controller. Try it, it might do the trick, and it will be faster than plan B... Try /sbin/sysctl -w hw.atamodes=pio. If this fixes the problem you need to put this somewhere in /etc/rc. At least that's as close as I can get to an 'official' location. A good quality UDMA33 cable is worth a try. Of course if the problem isn't UDMA related none of this will help ;) > > 2 x PCI Bus Master IDE ports (Support up to 4 IDE devices) Support: > PIO Mode 3 & 4 : 17MB/Sec. DMD Mode 2: 17MB/Sec. (Max.; The Actual > Transfer Speed Depends On HD) > ATAPI IDE CD-ROM and LS-120 Supported > Ultra DMA/33 : 33MB/Sec.(Synchronous DMA Mode) > > It has a 27GB IBM Deskstar HDD in it. > I believe these are the stats (don't have the model number handy, but I am > pretty sure it was a 7200RPM 27GB drive: > > Rotational Speed- 7200 RPM > Interface - Ultra-ATA/66 > Sustained data transfer rates - 22.9 to 13.8 MB/sec > Average seek time - 9.0 ms > > Any ideas for me? I do so hate unresolved errors. > > And while we're at it, dmesg.{yesterday,today} don't appear to clear out > and rotate the way I'd expect them to. I keep getting old partial dmesg > output e-mailed to me in my daily security output. Is this a FAQ as well? > Thanks. > -Tenebrae. > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-stable" in the body of the message -- ian j hart 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?3BAA3CD0.D23EBA1D>