Re: AMD64 small form factor
Le 6 mars 05, à 00:39, Morten Bo Johansen a écrit : Dale E. Martin <[EMAIL PROTECTED]> wrote: I installed my SN95G5 with very few problems. (During the CD part of the It is a little strange that it works for you and not for me. Maybe the difference lies in the brand of harddisk? I have a Seagate Sata 120 GB sata150 7200.7 Rpm / NCQ disk on the first sata contoller. It seems we are several people stuck with Seagate SATA disks on nforce3. I don't know if Manuele and Zsolt made some progress, but as for me I'm stuck as you are. Exactly the same symptoms. I reported on 02/16 to linux-ide@vger.kernel.org with some interesting tests but got no reaction. I'll try compilling the module in to see if it behaves better. There might be an issue with hotplug (if I manage to use the sata disk very quick after booting from pata it works, if I stop using it after some time it spins down and then is unusable) Thanks for the tip, GoM I tried gentoo and compiled my own kernel. I compiled the sata_nv drier statically into the kernel and then there is no problem. Morten -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] PGP.sig Description: =?ISO-8859-1?Q?Ceci_est_une_signature_=E9lectronique_PGP?=
Re: sata disk failure?
Hi Zsolt and others following this thread, For now I have no solution... exept the trick I tell below. So the problem is stil there, but there is some progress too. I hope this will help the developpers to narrow the problem down. I noticed something stange. If I manage to mount my sata drive and use it fast enough (that is, before the bunch of "device removed" happens), I can use the disk without any problem ! As soon as the drive is idle for some 20-30s, the error comes back and I can't use the drive anymore. During my first test I managed to copy over 3Gigs of .ogg files (823 files total) off of the sata drive to a pata one, and the sata disk went out _after_ the rsync was finished. Not sure about how much time after however. After a reboot I was able to copy 5-6 gigs of .nuv mpeg2 files (yeah this is a mythtv box...) and copy them back to a newly formatted partition on the sata drive before it went out, again _after_ some idle time (_not_ during the test which lasted at least 10-15 mins). Those tests were done on 2.6.10, I can try do do them again with 2.6.11-rc4 if needed. Does it help to narrow the problem down ? I think is some developpers helps me to put some debugging information into the sata_nv.ko we may be able to figure out what's wrong. I'm gom on [EMAIL PROTECTED] Another strange thing is that my last partition on the sata drive (170gigs) was almost full when it was used in my previous mythtv machine (kernel 2.4 with some sort of pata emulation, not on nforce3 but on intel hardware) and now the partition has 74Gigs of empty space. I wonder if there is some 138G barrier as Manuele tells. Later, GoM Le 16 fÃvr. 05, Ã 13:56, Zsolt KOZAK a Ãcrit : Hi Guillaume, I've just noticed your problem on debian-amd64 list. I've just installed a new Debian system and I've run into this problem also Nforce3 motherboard, SATA discs (2 200G Seagate). Hvae you got any solution? Or do you have still the problem? What about driver developers? I noticed you cc-ed them into your messages. best regards, Zsolt On 2005-02-15 23:38, Guillaume LaurÄs wrote: Thanks for your help Jeff. During my experiments I plugged an unpluged the drive numerous times. I also tried another SATA cable, but same thing. Even on the other motherboard SATA plug. I closely examined the drive and it seems right. like new in fact. The motherboard is even newer, and so is the cable. Is your idea consistent with the fact that the drive is _alays_ recognized in the bios, and _always_ recognized in linux ? All the time it fails with the messages "nv_sata: Primary/Secondary device removed" a few seconds after the console promt is displayed. Sometimes I have the time to mount the drive, sometimes not even. Later, GoM Le 15 fÃvr. 05, Å 17:50, Jeff Garzik a Ãcrit : Guillaume LaurÄs wrote: Hello, Here is a small report in attachement, done with 2.6.11-rc4. -- -- * Hardware is nforce3 250GB, booting debian pure64 on pata (Maxtor 30Gb 6E030L0) drive. The goal is to mount data found on a sata disk (seagate 200Gb ST3200822AS). The pata drive holds / and lvm partition with /usr, /var /tmp and /home. The ultimate goal is to switch all this to the sata drive, which already holds a (barely used) swap partition. * Soon after boot the console displays: nv_sata: Primary device removed nv_sata: Primary device removed nv_sata: Primary device removed nv_sata: Primary device removed ... -- Zsolt KOZAK [EMAIL PROTECTED] personal web:http://zso.lt Road To Avonlea: http://www.avonlea.hu -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] PGP.sig Description: =?ISO-8859-1?Q?Ceci_est_une_signature_=E9lectronique_PGP?=
Re: sata disk failure?
Thanks for your help Jeff. During my experiments I plugged an unpluged the drive numerous times. I also tried another SATA cable, but same thing. Even on the other motherboard SATA plug. I closely examined the drive and it seems right. like new in fact. The motherboard is even newer, and so is the cable. Is your idea consistent with the fact that the drive is _alays_ recognized in the bios, and _always_ recognized in linux ? All the time it fails with the messages "nv_sata: Primary/Secondary device removed" a few seconds after the console promt is displayed. Sometimes I have the time to mount the drive, sometimes not even. Later, GoM Le 15 févr. 05, à 17:50, Jeff Garzik a écrit : Guillaume Laurès wrote: Hello, Here is a small report in attachement, done with 2.6.11-rc4. -- -- * Hardware is nforce3 250GB, booting debian pure64 on pata (Maxtor 30Gb 6E030L0) drive. The goal is to mount data found on a sata disk (seagate 200Gb ST3200822AS). The pata drive holds / and lvm partition with /usr, /var /tmp and /home. The ultimate goal is to switch all this to the sata drive, which already holds a (barely used) swap partition. * Soon after boot the console displays: nv_sata: Primary device removed nv_sata: Primary device removed nv_sata: Primary device removed nv_sata: Primary device removed ... nv_sata: Primary device removed nv_sata: Primary device removed nv_sata: Primary device removed nv_sata: Primary device removed nv_sata: Primary device removed nv_sata: Primary device added nv_sata: Primary device added and the sata disk seems to spin down at the same time. This is currently on 2.6.11-rc4, although I got the same behaviour with 2.6.10 and 2.6.11-rc3. Removing power from the drive and pluging it again seems to make te drive work some time , until it spins down again... Seems like the hotplug code is triggering. That signals to me there is some connection problem. Maybe you have a bad cable, or maybe the cable is just seated improperly in one of the two SATA plugs. It's also possible your motherboard or drive SATA plug is faulty, but I would replace the cable first. Jeff -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] PGP.sig Description: =?ISO-8859-1?Q?Ceci_est_une_signature_=E9lectronique_PGP?=
Re: sata disk failure?
Le 13 févr. 05, à 13:07, Manuele Rampazzo a écrit : Hi, Guillaume Laurès wrote: I get this behaviour too ! After some time I get the "abnormal status" messages and the disk is dead until reboot. pure64 is installed on a pata drive (the sata is almost unusable). Is sata unusable because of the bug? so you switch to pata? Yes I actually installed on pata, planning to move to sata later (I have huge video files on the sata I want to keep)... but I'm stuck with this bug. I need to try a pata disk too... I tried the 2.6.10 kernel from pure64 as well as a self-compiled 2.6.11-rc3. Yes, this "bug" doesn't seem related to a specific kernel: I've tried 2.6.8, 2.6.9, 2.6.10... Mmh don't be fooled. with pre 2.6.10 my sata disk was not even detected... GoM PGP.sig Description: =?ISO-8859-1?Q?Ceci_est_une_signature_=E9lectronique_PGP?=
Re: sata disk failure?
Hello,
I get this behaviour too ! After some time I get the "abnormal status"
messages and the disk is dead until reboot. pure64 is installed on a
pata drive (the sata is almost unusable).
nforce 250Gb (sata_nv) + seagate 200 G (dont have the model number
handy)
I tried the 2.6.10 kernel from pure64 as well as a self-compiled
2.6.11-rc3.
I ran the computer case open and realized that at the same time the
disk stops spinning. unpluging power and pluging it back gives another
30s before the behaviour comes back. I'm going to:
- try to check if my bios hasn't some disk sleep settings enabled by
default, or
- see if it a hardware problem by trying to install winxp.
Hope this helps !
GoM
Le 12 févr. 05, à 16:54, Erik Mouw a écrit :
On Sat, Feb 12, 2005 at 02:45:18AM +0100, Manuele Rampazzo wrote:
Hello everybody,
I'm using an updated sid pure64 and I've got some troubles with my
sata
disks, but I'm not sure if the problem is in the hardware (in disks or
maybe in the controller) or in the software... Maybe you can give me a
good suggestion :-)
What kernel, what SATA controller, what driver and what disks are you
using?
Sometimes disk I/O stops and I'm just able to go to a tty and see
something like this:
ata2: command 0x35 timeout, stat 0xd0 host_stat 0x21
ata2: status=0xd0 { Busy }
SCSI error : <1 0 0 0> return code = 0x802
EOM ILI Current sda: sense = 70 7e
ASC= 4 ASCQ=a0
end_request: I/O error, dev sda, sector 215674327
ATA: abnormal status 0xD0 on port 0x977
ATA: abnormal status 0xD0 on port 0x977
ATA: abnormal status 0xD0 on port 0x977
I've seen similar behaviour with certain Seagate disks on sata_sil.
Erik
--
+-- Erik Mouw -- www.harddisk-recovery.com -- +31 70 370 12 90 --
| Lab address: Delftechpark 26, 2628 XH, Delft, The Netherlands
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact
[EMAIL PROTECTED]
PGP.sig
Description: =?ISO-8859-1?Q?Ceci_est_une_signature_=E9lectronique_PGP?=
Re: Netinstall fails to detect SATA disk.
Hello, I had this kind of behavior (sata disk detected but not responding), with netinst and 2.6.10. My solution was in the BIOS. I had to disable all kind of ide detection on empty slots and since then it worked perfectly. I explain: I have an ide disk master on ide0 (no slave), an ide DVD-ROM master on ide1 (no slave), one sata disk on ide2 and no sata disk on ide3. If I prevent the bios from detecting anyway on ide0/slave, ide1/slave and ide3, all is fine ! Hope this helps, GoM Le 1 févr. 05, à 04:01, Dale E. Martin a écrit : I don't know why it is not working with this debian-installer. It seems to me that the modules sata_nv and libata are loaded, but maybe I need others? A bug in kernel 2.6.[789] prevent disk from responding if attached to the nforce SATA port. You've been bitten by this bug. Symptoms are : quering the SATA port finds something, but the query of the diks times out. 2.6.10 works ok. I'm typing this on a Shuttle SN95G5 running 2.6.8 with (only) an SATA hard drive. I don't believe I did anything special - the SATA just worked... (I used the sid-amd64-netinst.iso image.) Let me know if there's any other info I can provide that would be of help. Take care, Dale -- Dale E. Martin - [EMAIL PROTECTED] http://the-martins.org/~dmartin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] PGP.sig Description: =?ISO-8859-1?Q?Ceci_est_une_signature_=E9lectronique_PGP?=
