Q: Serious external drive problems after installing Mavericks

The drives that were corrupted and unrepairable by Disk Utility (but still mountable) were Time Machine volumes -- the original one and then the replacement (a fresh new drive). I set aside the original one in case I needed to restore data from it later, put in the new one, and it went bad 2 days later.

The other drive -- the one I used to diagnose this problem -- was a drive I was using to edit some video, so the only files that got written were cache files, new exports of finished videos, and compressions. I tracked down the problem because stuff I was uploading to Vimeo wasn't encoding properly, and I finally downloaded the file Vimeo got and compared it to what I *thought* I'd uploaded.

After that I found that I could simply copy a 5GB file off the PROBOX into the MBP's internal drive a couple of times, and sometimes the copies would be silently corrupted... but not always.

Right now I'm running a TM volume (reformatted the second one) and the volume I was using for testing in a 2-drive enclosure and regularly running Disk Utility to check.

I assume you have, until resolution, considered just going the "formatted brick" route? 

This is proven 100% reliable...Just a USB formatted volume and no software control for/of same.

Sorry about your data corruption

You will have to define what you mean by "formatted brick".

Sorry, as meant:...........

Formatted brick is just parlance for referring to a HD either bare or in a USB enclosure that is formatted in (whichever format needed) Mac OSX extended journaled.

Meaning: no RAID, ...no control software, ...nothing but a formatted HD and its little SATA bridge card inside its USB enclosure.

...Either that or a naked HD inside a HD dock.

such as:

Ah, ok. Well, the ProBox (as I use it) is just an enclosure for 4 formatted bricks. My fallback is a couple of 2-slot docks but they are USB2 .

I did some more testing last night, and managed to get one silent file corruption when copying a 9GB file PROBOX -> MBP Internal, but not PROBOX ->(through MBP)-> USB2 dock or USB2 dock -> MBP Internal.

I also did repeated attempts to get a corruption copying from the PROBOX to a MACPRO running 10.6.8 but could not get it to error.

Overnight I ran the MBP Hardware check; no issues found.

Drives that are corrupted by this report a SMART status of 100% perfect.

I have experienced the same problem.  I've lost 22 TB of data on multiple external drives (firewire, eSATA, USB) and some multiple times just as I rebuilding them.  I  disconnected a 6TB WD MyBook. 

I just now lost an INTERNAL 4TB hard drive (replaced with an empty MyBook partition) which has a Drobo NAS backup.  I just called Apple support and they wanted me to take my bulky MacPro into the store.  I decided not to for the moment.  I will use a spare Mac Mini to rebuild my Mac Pro drive from a CCC backup prior to installing Mavericks and rebuild my backups. 

I first thought the problem was my eSATA PCIe cards (a WiebeTech 2 port eSATA and a CalDigit 2 port eSATA and 2 port USB3).  I eliminated those and still had external drives lost.  I used firewire 800 instead completely.  I also added back the rear eSATA extender cable that connects to the unused CD eSATA port.  That worked to create a time machine backup and seemed faster after disconnecting the MyBook 6TB drive.  Unfortunately, you have to restart to change the drive with extender.

I see the error codes on the corrupted RAID drives-----  Writing lastMountVersion as FSK..........journal magic is bad (0x0 != 0x4a4e4c78)

I've seen that before and it's discussed here a bit

http://www.sooperarticles.com/technology-articles/data-recovery-articles/how-res olve-mac-journal-magic-bad-error-340304.html

MAC operating system has a feature of journaling the file system due to which the chances of fault flexibility together with providing protection to the file system preventing from the hardware failure incidents and sudden power turn off. The journal plays vital role comprising all the confidential information that are required by the MAC operating system so that it could be returned to the previous working station.

These journals being vital component of the MAC operating system if gets corrupted, leads into serial results. It may turn the whole data inaccessible by means of turning the volume unmountable. Once user undergoes such disaster the only thing that leads to recover the situation is the valid backup as the data can be restored and the corrupted can be removed from the hard disk without being worried about the data loss issue.

While the corruption is encountered in the HFS volume of the MAC operating system then error that is encountered with the user is stated below:-

HFS(3): Journal replay fail. Writing lastMountVersion as FSK!

jnl: is_clean: journal magic is bad (0x1fd17 != 0x4a4e4c78)

HFS: late jnl init: failed to open/create the journal (retval 0).

jnl: open: journal magic is bad (0x1fd17 != 0x4a4e4c78)

While the corrupted HFS + volume is attempted to be mounted, the above shown error triggers on the screen. The corruption of journal being the major cause turns often turns to smash up the entire MAC volume to an extent along with damaging the command line. Although you may find the disk utility option as well as the command line application facilitated in the MAC operating system which you may use in order to get the issue resolved. If possible try disabling journal and further turning it on.

I had the same trouble:

i've got a 2013 Western Digital MY  BOOK STUDIO EDITION II (4tb in  raid 1).

after 2 days i upgrated my imac 27 late 2011 to mavericks, suddenly my WD was considered totally empty. the folder indicate 2TB available (because of it has partitioned in RAID 1).

during the upgrading to maverick I disconnected the external hard disk.

all functions very well, but it is like i had formatted all my files in the WD.

it strange because the light in the case of the external WD, that indicates the capacity, is showing that the hard disk is not empty, but indicates the same quantity of files that was indicated also before the "resetting"...

(sorry for my english.)

what do i have to do?

thank's

andrea (venice, Italy)

After four long days of deep scanning of my wiped out drives, I finally managed to rescue 3 TB out of 5 TB. I used Data Rescue 3. It's not free, and it's worth every penny.

I wasn't able to recover the 2TB drive where used to keep my Time Machine backups and loads of work related reference documents and media.

I believe that my problem was caused right after Mavericks upgrade by some strange behaviour of Mavericks+Time Machine+External Drives...

• After upgrade, somehow Mavericks+Time Machine found my 90% full external drive as a blank drive...
• Automatically, made a full Time Machine backup - and the drive name never changed!
• And, kept on making backups every hour during the upgrades of the upgraded applications.
• That corrupted my data irreversibly.

When I plug-in the larger drive, similar thing happened (both drives are of diffferent makes)

• It's been wiped and renamed as the Time Machine backup device!
• This time, I was lucky. When I realized this change and I directly pulled the USB cable.
• Since, it has not been written, I was able to recover the data.

I disabled Time Machine during recovery. When recovery is complete, I started automatic Time Machine backups to a new external drive.

I'll be observing for a few days if anything odd happens. My advice to those of you upgrading is, to turn off Time Machine before upgrading to Mavericks and start it fresh to a new external drive. Best of Luck and Patience.

Unfortunately my USB connected WD HDD got erased like ithe Firewire 800 connected WD RAID drives.

So no difference if the HHD is connected via USB or Firewire 800.

Flavio Muscetra Italy 

Unfortunately my USB connected WD HDD got erased like ithe Firewire 800 connected WD RAID drives.

Yes, Ive seen a dozen or so reports of same, USB or not,  use of the WD software caused data corruption and loss of all metadata.

Common factors appear to be:

All RAID arrays

Any RAID area, including seagate (Lacie, same thing) that has WD installed

USB non-RAID single drives utilizing WD software

and a minor "sleep issue" on connected Firewire and Thunderbolt drives of any variety.

My main problem is that on the WD RAID I have 3 years of high resolutions photo archives indexed with Lightroom 5.

Now I'm looking for a solution to recover at least the photo archive.

I was thinking maybe with a disk recovery tool I could recover the files with no names (I've read this in the thread) and maybe I could re-import the photos in the Lightroom DB even if without metadata.

Really i would like to wait and see if a solution exists to recover completely the contents as they were before this issue.

What happened is really unbelieveble. I bought an external disk with RAID just to avoid this nightmare.

I have gotten the corruption on USB non-raid devices that DO NOT use the WD software, specifically a PROBOX 4-drive enclosure hosting 4 independent drives, each formatted with one volume.

PROBOX technical support has been unable to replicate this on a unit with the same (latest) firmware but using Seagate drives.

However, I have not gotten any corruption when connected via a STARTECH 2-drive dock.

Yes, I noted same on another thread on USB HD with corruption and loss of all metadata

same drives using non RAID , and no control software

which leaves eliminatively only SATA bridge interface and possible firmware on same, ...OR Mavericks itself

However bare docked formatted bricks (naked HD formatted Mac OSX ext. jour.) have shown such corruption and Ive tested a pile of same and have a few on mavericks for past few days.

I know the mechanical HD and its mfg. has no role in any of this, regardless of quality of Mfg.  the actual HD itself, be it Toshiba, WD, Hitachi, or Seagate is not at fault for same.