Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

MacRumors

macrumors bot
Original poster
Apr 12, 2001
68,179
38,960
Apple posted a response to concerns about file corruption on External Firewire Drives.

Apple has identified an issue with external FireWire hard drives using the Oxford 922 bridge chip-set with firmware version 1.02 that can result in the loss of data stored on the disk drive. Apple is working with Oxford Semiconductor and affected drive manufacturers to resolve this issue which resides in the Oxford 922 chip-set.

In the interim, Apple recommends that you do not use these drives. To stop using the drive, you should unmount or eject the disk drive before doing anything else. Please check this web page for further updates.

Both Firewire 800 and 400 drives are reported to have been affected by user reports.
 
Hmm... good to see Apple finally taking action on this, but I wonder if it's as simple as just Oxford 922 controllers--there've been a lot of reports from people using FW400, and unless someone is confused or some vendors are putting Oxford 922 controllers in FW400 drives, it looks an awful lot like there's more to the problem.

Or, maybe, it's more than one problem, but each has the same symptom--munged FW drive after restart.
 
So I wonder if it's just a driver that needs to be re-coded or what? Because I hate to have bought fw800 ex-drive and not being able to use it.
 
Good to hear that Apple have recongised the problems.

HOWEVER, looking through the Apple discussion boards, ppl have also had FW400 drives trashed.

Hope Apple soon releases a statement about those.
 
Wow- glad I read that
Thanks MR!

I was just about to buy 4 FW800 enclosures.

Anybody know ehre to get some good SATA stuff? (raid card, drives, enclosures, cables)
 
Well this is ***** super.

So does this mean they will do something about my 200 GB of data that I lost? Any recomendations of how I can get my data back?
 
Try mounting your FW drive on another version of MAC OS - 8,9,10 etc.

There have been some reports that other versions can read the disk, but Panther cannot.

Drive X is also popular for recovering trashed FW disks.

If you can't retrieve your data, then your stuffed. You could try phoning Apple for support to see what else they suggest.
 
Originally posted by Stella
Try mounting your FW drive on another version of MAC OS - 8,9,10 etc.

There have been some reports that other versions can read the disk, but Panther cannot.

Drive X is also popular for recovering trashed FW disks.

If you can't retrieve your data, then your stuffed. You could try phoning Apple for support to see what else they suggest.

I've tried all of those things.

I will make a trip to the Apple Store in Tampa this weekend. Hopefully they will help me.

And yeah, what on earth is positive about this?
 
Lacie Already has a patch on their site

Lacie has a fix, but you need to boot up in 9. which sounds smart anyway.
 
I am not quite clear on how the Lacie fix works. Will it just give me a working drive again? Or will it actually recover the data?
 
Re: Lacie Already has a patch on their site

Originally posted by usingmac
Lacie has a fix, but you need to boot up in 9. which sounds smart anyway.

HOw do you boot up nine on a newer Mac that doesn't support nine booting?
 
so does this only happen if you updates the firmware? i've been using my 200gb fw800 drive with panther since i installed and have had no problems. does the data corruption happen consistently?
 
I knew there was a reason my copy of Panther is sitting on my shelf waiting to be installed.

And from my digging, it seems as if its FSCK that's causing the problem. So if you have a problem DO NOT RUN FSCK ON IT :)

For those who have had data loss, i'd recommend running either Diskwarrior, Drive X, or Datarecovery.

I'd also try seeing if Open Firmware can get contents on the drive.

(I've had huge fails where the system wrapped has bee mightly f***ed and I knew it was recoverable cause OF could still see the data)
 
FSCK and UFS

The problem, at least on the surface, seems to be that FSCK reads the HFS and HFS+ formatted drives as UFS formatted, with severe directory damage. FSCK attempts to fix the "UFS" volume, and in so doing, overwrites the functional, undamaged HFS directory with UFS data, rendering the disk all but useless.

And yes, experience (not mine, personally) has proven that both FW400 and FW800 drives are susceptible.
 
Ok, what I don't get is how none of this was relized with all the monthes of developer testing. I am sure that the developers who had panther monthes ago had external hard drives. Also I think that people wanted panther so badly that apple rushed the process and didn't work out the bugs. Personally, I could have waited another month, if Apple could have fixed the problems before the Gold Master.
 
I was running 10.2 on an externaml FW disk, and now I copied over my 6K mp3 to my PowerBook and now many of the songs are cut in half. They are full length, but in iTunes they play the first 10 seconds, then switch songs. Anyone know how to fix that? THanks
 
Hmmmmm.......

It seems like this only affecting folks with Oxford 922 bridge chip-set and not any other chip-set. So whatever you do not buy ex-drive with Oxford 922 bridge in it.
 
Originally posted by SiliconAddict
Why are there 6 positive votes? This is a major issue that could, and most likely has, cause data loss.
The positive thing is that Apple is aknowledging it, and hopefully doing something about it. I'd rate this item positively for that reason, too, since that's the story, not that the problem exists--I knew that since last Friday, since I always scan troubleshooting reports of early adopters before leaping in myself.

As for how this didn't get caught earlier, it's a bit of a mystery, but I don't think it's quite as guaranteed a problem as it seems. The people who have it (and note that there are a couple of similar data loss problems being reported that aren't as repeatable or severe, just as with every major system upgrade) seem to find it very repeatable, but there are people with almost identical hardware setups that don't have the same problem, so it would seem that it's not as cut and dried as 922=bad, in which case surely Apple would've caught it.

For reference, the problem that's causing the most headache is one where, after restarting (via the menu command or a software install restart), the contents of an attached firewire drive are killed in a big way. It's only after a restart, and definitely doesn't affect all attached drives (or even, apparently, all partitions on the attached drive), so never restarting with a FW drive mounted is protection, so long as you're careful.
 
Originally posted by Genie

I had a firewire 400 drive get corrupted in Jaguar a few weeks ago.

Me too! a Lacie 120Gb FW400, just suddenly trashed itself on restart.

I recovered it with Discwarrior but lost a 13Gb image file, not to bad as it saved the remaining 40Gb.

Viv
 
Originally posted by Makosuke
The positive thing is that Apple is aknowledging it, and hopefully doing something about it. I'd rate this item positively for that reason, too, since that's the story, not that the problem exists--I knew that since last Friday, since I always scan troubleshooting reports of early adopters before leaping in myself.

As for how this didn't get caught earlier, it's a bit of a mystery, but I don't think it's quite as guaranteed a problem as it seems. The people who have it (and note that there are a couple of similar data loss problems being reported that aren't as repeatable or severe, just as with every major system upgrade) seem to find it very repeatable, but there are people with almost identical hardware setups that don't have the same problem, so it would seem that it's not as cut and dried as 922=bad, in which case surely Apple would've caught it.

For reference, the problem that's causing the most headache is one where, after restarting (via the menu command or a software install restart), the contents of an attached firewire drive are killed in a big way. It's only after a restart, and definitely doesn't affect all attached drives (or even, apparently, all partitions on the attached drive), so never restarting with a FW drive mounted is protection, so long as you're careful.

This is a good thing yes, but it's not Apple acknowledging their problem. They have found the problem in the Oxford chipset and it may exist in the 911 chipset also.

Again I repeat this is not Apples fault. If you want to scream to someone go to Oxford Apple is trying to help them fix their chipset since Apple developed the 1394 standard that is Firewire.

Also since all data isn't overwritten there should be a way to write a patch to temporarily fix the bad data on the drive so that it can be read and everything removed so it can be reformatted and fixed properly. If this isn't doable a good hard drive data recovery company should be able to recover the lost data.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.