Jump to content

nForce SATA

thefighter
 Share
102 posts in this topic

Recommended Posts

Gdfath3r
Gdfath3r
Posted
Posted
or maybe easier the differences between nforce3 & 4 because it works without any problem on nforce3

 

That would the difference between sata1 and sata2

Actually, most of the people has incompatibility problems with nVidia onboard SATA controller in NF4 boards, that could be a source of trouble.

tokyovigilante
tokyovigilante
Posted
Posted
That would the difference between sata1 and sata2

Actually, most of the people has incompatibility problems with nVidia onboard SATA controller in NF4 boards, that could be a source of trouble.

I think all of the problems people are having are with nForce4 boards and SATA drives regardless of speed. I think all the nForce4s (CK804 MCP chipset) support SATA 3.0gbps. In my experience I get the same random corruption with both 1.5 and 3.0gbps drives.

 

(As an aside the difference between SATA-I and SATA-II is not the same as the difference between SATA 1.5gbps and 3.0gbps - I vs II is a feature set difference (ie NCQ, hot-plugging etc) and 1.5 vs 3.0 is purely interface speed. They were introduced at roughly the same time and most SATA II hardware also supports 3.0gbps transfer speeds hence the confusion).

jape
jape
Posted
Posted

Motherboard: A8N-E

Chipset: nForce 4

Hard Drive and Hard Drive Type (SATAI or SATAII): Maxtor 6V200E0, SATAII 3gb/s

BIOS: 1011-004

Is Hard Drive under the Silicion Image Controller on windows?: No SIC on Windows Device Manager.

Different Hard Drives on different channels?: Just one hard drive in primary channel

And of course if OSX works for you or not:

 

Random Freezes. With USB Bluetooth dongle plugged in, I get a freeze every 15 minutes. With no USB Bluetooth dongle, I get a freeze every hour.

I have no other kext installed, it is a fresh 10.4.3.

 

I usually get a "Blocking bus" error every 5 minutes. They usually "group" themselves (i.e. for 5 minutes I get no errors, but suddenly 10 Blocking Bus errors appear (slowdown noticeable), and then no errors again during 5 minutes).

 

After around 30 boots, the boot partition is unrecoverable. It's strange, becuase at boot 29, the system still works perfect (no files missing).

AREOS
AREOS
Posted
Posted
On my board (Gigabyte GA-K8NMF-9) the SATA controllers offer two modes 1001 and 1004 (selectable in the bios) 1001 is non-raid and the 1004 is an 'enhanced NVIDIA' mode.

Does anyone else have these options? Do changing them effect the VIA driver operation?

 

Motherboard Gigabite K8N51GMF-9

Chipset Nforce 430

My hard drive is a Maxtor Diamond Plus 9 250GB SATAI

BIOS version F2

Is Hard Drive under the Silicion Image Controller on windows?: No SIC on Windows Device Manager.

Different Hard Drives on different channels?: Just one hard drive in primary channel

 

 

This motherboar suppor for SATAII I edit the AppleVIAATA driver and I put my hardware ID

but after reboot I get the "blocking bus error"

 

So I use a PATA drive for now

jape
jape
Posted
Posted

Well, I correct myself, there is data corruption on boot partition at every boot until it finally dies after a few succesful boots.

 

Some samples:

hfs: remove: strangely enough, deleting truncated file waiting4loginwindow.pid (128309) got err 5i

ntpd[155]: Can't open /var/run/ntpd.pid: Input/output error

DirectoryService[68]: WARNING - dsTouch: file was asked to be opened </var/run/.DSRunningSP1>: (Input/output error)

 

And this error once appeared: diskarbitrationd[48]: unable to mount /dev/disk0s2 (status code 0x0000001C).

disk0s2 is not my boot partition (disk0s2 being located on a PATA disk!!!). However, the next boot the same partition mounted like always.

 

 

Bus blocking errors are mixed between other log lines; however, they don't appear to be near i/o errors.

 

And, sometimes, suddenly... I get this on the log:

kernel[0]: disk1s3: I/O error.

kernel[0]: jnl: do_jnl_io: strategy err 0x5

kernel[0]: jnl: write_journal_header: error writing the journal header!

(disk1s3 IS my boot partition, on the SATA disk)

 

After this error, every disk write operation fails:

$ touch temp

touch: temp: invalid argument

iTunes, iChat,... all fail to start with "error -50".

 

 

And, after a while... complete system freeze. The Caps Lock LED state doesn't change anymore. The mouse sometimes moves, sometimes doesn't.

borez
borez
Posted
Posted

Hi there all,

 

Some questions with regards to the data corruption: does the corruption happen only on the boot partition itself?

 

I'm wondering if there is any corruption on the following instances:

 

1) The drive is not used for booting, but for plainly data storage

2) Whether the other partitions on the boot disk (if the SATA disk is used) are affected.

 

I presume that all the affected corruption disks are using FAT32...

tokyovigilante
tokyovigilante
Posted
Posted
Hi there all,

Some questions with regards to the data corruption: does the corruption happen only on the boot partition itself?

No any read or write to a SATA drive has the potential to corrupt or return corrupt data.

I'm wondering if there is any corruption on the following instances:

1) The drive is not used for booting, but for plainly data storage

Yes this causes corruption.

2) Whether the other partitions on the boot disk (if the SATA disk is used) are affected.

No, only the currently accessed information.

I presume that all the affected corruption disks are using FAT32...

No, my experience has been corruption with both HFS+ and NTFS partitions.

 

I think the problem is a low-level hardware/driver issue rather than a higher level partition or filesystem issue. Unfortunately that is as far as I'd got, planetbeing was doing some interesting investigation a couple of weeks ago, but his conclusion was largely a ground-up driver rewrite was needed.

jape
jape
Posted
Posted

As you can see on the log samples I attached, it seems that corruption only appears when writing -- that's the reason NTFS partitions are not corrupted on my SATA disk, they're mounted read-only.

 

In the SATA disk I just have the HFS+ partition (gets largely corrupted), and a few Ext2 partitions with a NTFS partition. None of the other partitions (other than HFS boot one) have been corrupted.

 

 

I see no reason to believe that data is corrupted while reading. If so, the system would NEVER boot (try to corrupt a single bit in some OS system folder, if it boots you're lucky) . However I have no proof that data is only corrupted while writing.

borez
borez
Posted
Posted
As you can see on the log samples I attached, it seems that corruption only appears when writing -- that's the reason NTFS partitions are not corrupted on my SATA disk, they're mounted read-only.

 

In the SATA disk I just have the HFS+ partition (gets largely corrupted), and a few Ext2 partitions with a NTFS partition. None of the other partitions (other than HFS boot one) have been corrupted.

I see no reason to believe that data is corrupted while reading. If so, the system would NEVER boot (try to corrupt a single bit in some OS system folder, if it boots you're lucky) . However I have no proof that data is only corrupted while writing.

 

Hi jape,

 

Nice observation - could anyone concur with this as well?

tokyovigilante
tokyovigilante
Posted
Posted
Hi jape,

 

Nice observation - could anyone concur with this as well?

No, I have experienced corruption errors with file transfers from a SATA NTFS partition to a PATA HFS+ partition. The read errors are far less frequent than write errors, but I am unable to read more than 1-2gb off a SATA drive without the transfer failing due to I/O errors. This is why I believe there is an nForce4 hardware issue that is not sufficiently addressed with the errorchecking functions of the VIAATA driver. The reason your SATA NTFS partitions are not corrupted is as you rightly stated, they are mounted read-only. However accessing the data returns corrupt data to the system, albeit at a greatly reduced error rate than writing.

 

I can't think of another reason for random errors of the nature I and others have experienced, a pure software issue shouldn't cause random errors - they should be predictable, consistent and reproducable in that case.

 

If anyone could try some large transfers SATA->PATA and corroborate my findings that would be good thanks.

G-Force
G-Force
Posted
Posted

Hello,

 

I have a nForce 3 Ultra motherboard (K8N Neo2 Plat.) and I want to install Mac OS X on a 200GB SATA disk with an extra 8GB partition. The problem is, I have precious data on the second NTFS partition. The disk is connected to the first SATA controller (0x00E310DE). I'm really scared that my downloaded series will get corrupted or something, is it safe to do this?

 

I'm really tired of that slow Western Digital harddisk drive.

jape
jape
Posted
Posted
The problem is, I have precious data on the second NTFS partition. The disk is connected to the first SATA controller (0x00E310DE). I'm really scared that my downloaded series will get corrupted or something, is it safe to do this?
If you're messing with OS and disk partitions, you should always make a backup.

 

Even if probably you won't suffer any corruption since you're using nForce 3.

G-Force
G-Force
Posted
Posted
If you're messing with OS and disk partitions, you should always make a backup.

 

Even if probably you won't suffer any corruption since you're using nForce 3.

Well yes i know that, it's a bit hard to backup 160GB of series to a 100GB harddisk or so...

 

Edit:

Been busy all day long trying to get this to work, it just won't work for me. Is there any possibility to load the JaS modded VIA kext included on the DVD before the setup starts? So I can select and format the partition within the setup?

G-Force
G-Force
Posted
Posted

Well, I installed Mac OS X on a USB drive with 8GB of disk space. Then I used Acronis Disk Director Suite to copy that disk to a spare FAT32 partition (which is now HFS+ Journaled).

 

My system recognises the Mac OS X partition by showing up the Darwin Bootloader at boot. The only problem now is that i get a "Still waiting for root device error"

 

My configuration is:

SATA Controller # 1:

Maxtor DiamondMax 10 200GB SATA

|_ Windows XP (NTFS, Primary)

|_ Data (NTFS, Logical)

|_ Mac OS X (HFS+ Journaled, Primary)

 

SATA Controller # 2: Nothing

 

PATA Controller # 1:

Samsung DVD-ROM

NEC ND3540 DVD-RW

 

PATA Controller # 2: Nothing

 

Does someone know the answer to this problem? ;)

jape
jape
Posted
Posted

I didn't get the installer to install to SATA either. I guess some images out there have the kext just broken.

 

I always use VMware for setup.

 

BTW, my root OS X partition got unbootable once again. It's been nearly 40 boots this time. It doesn't boot into single either, it's just broken.

  • 2 weeks later...
andrewf
andrewf
Posted
Posted

I installed OS X 10.4.6 on an ide disk (A8N SLI Delux, nForce4) with the sata.via option (I guess this essentially uses the method described here) and after a couple of reboots it corrupted 2 out of 11 FAT32 partitions on one of my sata-disks. I'm still trying to recover the data.

 

Has anyone who experienced these data corruptions succeeded? Is there some place where I can find usefull info?

  • 3 weeks later...
  • 2 weeks later...
 Share
Go to topic listing
×
×
  • Create New...