Other External products

[ChrashM]

The last replacement drive I received around a month ago failed earlier this week. I decided it wasn't worth the hassle to pay them another $20 to replace it with another unit that would ultimately fail. I decided to take the drives out, void warranty, and see if at least one of them was usable. Both were labeled "factory reconditioned" or something like that and BOTH cannot be re-partitioned or mounted.

A good friend of mine took the drives and is going to see if any of low-level utilities can bring them back to life. Both drives were making the two "dink-donk" clicks you described. I will post the results when I hear back from him.

So now I am down to my last of three 1.5TB units. This one is hooked up to my Mac Mini and is serving music and video. We will see how long it takes to fail.

My replacement solution was to order two Hitachi 1TB Deskstar internals for installation in my Mac Pro. Both were installed today and appear to be working fine. Let's hope Hitachi knows how to design a reliable storage solution.

[Didrik]

ChrashM,

   Take steps to protect the stuff on that last 1.5TB drive.  It is not to be trusted.

    I'm presently trying to recover the contents of a 2nd-time failed, out-of-warranty, Maxtor OneTouch2.  The MFT and partition table were scrambled in such a way that very early in the recovery scan unexpected $MFT records occurred in very high sectors ("unexpected MFT record 116934 at 3260071424...."). [Please note that I do not intentionally use smiley faces in my posts.  That last one is automatically placed by who-knows- what.]

    Cautionary messages that if this is a RAID volume, then check it for consistency, (it's not; it's a 279.5 GB single drive I think), suggest to me that perhaps the software / firmware that permits drive pairs to be mirrored or striped, and its interaction with the controller, may be the underlying cause.

    This current case is as puzzling as any of those directly concerning OneTouchIII and IVs, because in some of the reported cases, the systems were simply shut down normally, and failed to find the drives when restarted the next day or at some other time. This is exactly what happened for this OneTouchII.

     The recovery efforts have been revealing: there are several GB of files which should be in a space using 512-byte allocation units (the one I thought I was using throughout the disk), but which the recovery software tells me are actually in 4K allocation units.  This would indicate, based on the history of this drive, that the "format" Maxtor's utility employs -- or the one that Microsoft uses through Disk Management -- somehow did not properly reformat the drive after the last such "reinitialization" when I lost my data on this drive the second time.

     Tests are underway using the default Maxtor initialization utility on this drive, then running some recovery software.  The Maxtor utility presumably does something like a "Quick Format," which I believe adjusts pointers and perhaps replaces the partition table. But it probably doesn't change a huge amount of what is on the disk(s). (Recall I suggested that users receiving "refurbed" units should immediately try running a recovery software on the replacement drive for interesting results ....)

[Didrik]

   A followup to my previous post, with results of reformatting.

   Right.  What follows is entirely in the Windows XP Pro SP2 environment on a Maxtor OneTouchII (279.5GB) drive. Similar results are expected on OneTouchIIIs.

   From a data recovery point of view, once the drive is recognized to the point that it can be mounted in the file system, and be seen by Disk Management, I can see little difference between (1) the drive as formatted by Maxtor's s/w and (2) the drive formatted with a Quick Format, and (3) the same drive formatted with a "Full" format (i.e., without "Quick Format" checked).  There are a few differences.

    The Maxtor autorun file and the mxoicon file are removed by the Full Format.

    There are many still unexpected MFT records at specific locations.  These are not corrected by a Full Format. For example, MFT record 11776 is still located at both sector 105,545,728 and 3,258,646,528, as before.

    There appears to be no difference as regards successfully recovering 206 GB of Retrospect 7.5 datafiles.  They recovered properly and were usable by Retrospect.

     The additional 25 GB of miscellaneous files temporarily parked on this drive which was principally used for backup, were trashed but identifiable.  That is, the recovery software tells me the name, the supposed length and location of most of those files, but when further steps are taken to actually recover them to another location, they are unreadable by the application that created them, and are shown as "unrecoverable." PDFs, JPGs, ZIPs, and TIFFs are good examples of those types of files.

     Forum participants may be interested to know that with a suitable disk editor, however, it is clear that the file headers have been trashed, at the very least.  Perhaps much more is corrupt, maybe not.  The header part of the file that would tell an app what type of file it is, is clearly missing.  A skillfull patient person could put in a suitable header and make the files readable by the app that created them, or a similar one.

     In other cases, such as email that includes lots of text, even though an email program may not open the files properly, the disk editor clearly can read through (and past) any application-level "password protection."   The disk as recovered has all sorts of confidential yet readable stuff on it, clearly visible to the "imprudently curious." Remember this is after a Maxtor re-init, a MS quick format, and a MS "full" or regular format.  Ollie North probably remembers this, still.

    Files encrypted at the app level appear to remain unreadable.  Email Bodies which are encrypted, appear to remain so.

    Recovery software is typically a fix-after-fail (FAF) approach. It need not be only FAF.

    For the courageous and desperate who continue to use these drives, I suggest investing in a good recovery software and running it on the reformatted drive which Seagate / Maxtor will eventually suggest you start again with (if they don't simply tell you to return it).  However, after re-initializing but before re-loading the drive with data, run the recovery software on it, get the partition layouts, bad sectors, if any, and so on.  Save it to a thumb drive or some safe place.  If necessary this info may later prove helpful.  There is also software out there that gets and keeps this information even though it's not disk file recovery software, per se.

    Hope this helps someone, or gives someone better ideas.

[AlanM]

Hi all,

It looks like there are 3 different issues going on here.
1) Taking a drive between Mac and PC and expecting it to write correctly.  That's a Mac and PC OS compatibility issue.
See this KB article.  (That's for you, brainspin.)

2) The drive doesn't show up at all.  I don't know what to tell you on this one - it's pretty inconceivable that 5 drives in a row would fail unless some kind of bad USB port or power surges or something else that has not been disclosed is behind it.  

3) Disappearing partitions.  Our guys haven't been able to reproduce the issue without violating normal safe removal procedures, after multiple tests on multiple drives of this kind.
There may yet be something found, but it hasn't been yet.  I apologize that it's caused so many problems, and I take this opportunity to remind everyone that you should have a backup of all important data, which means at least 2 copies of the data on at least 2 different storage media.

I'm going to close this thread since the tone has degenerated and it's become mostly "me too" pile-ons, with little useful information shared, which is what the forum is for.



Message Edited by STM on 03-25-2008 10:00 AM