Q: Mavericks corrupts external hard drive

I have a Toshiba Simpledrive.

After updating to Maverick I lost the ability to use Time Machine. (I would enter TM but could not reach any files or even confirm that my back ups were still there).

It did appear that files still existed on my SimpleDrive ext drive but I could not confirm that.

I just used Disk Utility to Verify my back up disk. It took about an hour and reported there were no problems but more importantly when I rechecked TM and it is now working fine (as it did before Maverick)

I can now enter TM and get to any file that was backed up. Not only that but all this time that TM appeared to be not working it has been backing up every hour etc. just like it was supposed to.

One other thing of note: The clock looking icon for TM at the top of the screen is NOT spinning during back up like it used to but it IS STILL backing up. This is my only remaining "malfunction".

So running Mac Disk Utiliy Verify function (not Repair) fixed it.

I hope this helps some out there.

Tony

Noticing the same thing with the Time Machine Menu Icon no longer spinning whilst backing up, yet it is still backing up properly to external. Must just be another GUI change in Mavericks. (I have no idea why!)

But all good

Cheers

Pete

Interesting.  Thanks for taking the time to test.  Question for you:  Can you provide some details about how you partitioned the drive, what partition type it is, etc.

So, here's something that used to happen to two of my FW800/USB 2.0 dual interface drives.  Occasionaly on reboot or power down, the drives would not mount on my Mac Pro (all OS X versions since 2009) via the FW800 interface.  If I just unplugged the FW800, then plugged them in via USB, let them mount, then eject and plug in via FW800 they were fine again for a few more times.  There was never any damage, just needed that bump once in a while (I probably had to do it 10 times since 2009 or so).

I'm not using any FW800 drives regularly now but the two I have, have been plugged in numerous times to my 2013 iMac running Mavericks via a TB adaptor daisy chained off of my TB RAID and have not failed to mount yet.  Also worth noting is they are both WD.

GetRealBro wrote:

 

For several days now, I’ve been trying to get my problem Seagate 3T disk (single partition) to fail again while running Mavericks. I began by using a USB3 Interface/base (which only works at USB2 speeds on my MBP). I would eject, un-attach and re-attach the single partition using the Finder or unmount and re-mount it using Disk Utility. When it didn't fail, I started copying GBs of data to it then ejecting or un-mounting and re-mounting. It worked fine.

 

So I switched to a FW800 interface/base, mounted it, copied 40ish GB of data and un-mounted it with Disk Utility and immediately asked DU to re-mount it.  Bingo.... it failed to mount. And, as before, Mavericks Disk Utility could not verify or repair the volume. The repair fails at

“Checking the file system

Volume repair complete.

Updating boot support partitions for the volume as required.

Error: …."

 

I shut down and rebooted into Snow Leopard (10.6.8) with the drive still attached. The drive did not mount and SL’s Disk Utility could not verify or repair the volume (same story as DU in Mavericks). So I ejected the drive using DU and switched back to the USB3 interface/base. When I reattached the drive it mounted and verified!

 

I ejected the volume and restarted in Mavericks with the drive still attached. The drive mounted and verified!

 

So either this brand new FW800 interface/base is bad or Mavericks does not always handle this FW800 interface correctly, leaving the volume slightly damaged after un-mounting it. In either case, I’ll be using the USB interface/bases until this issue is resolved.

 

— GetRealBro

 

BTW the unmount/mount problems I have reported are at the partition/volume level, not the whole drive. While this drive has only one 3T partition/volume, the other problem 3T drive had two partitions. And only one of them had mounting issues.

This is good news, I assume this could be reproducible? You did this without any WD software installed?

It would be worth grabbing logs & noting rough times that you saw the damage occur & other key events like reattaching the disk, what the volume names were etc.

Is this a clean 10.9 install? 

What about the 10.6.8 install?

If you grab all the logs there is no telling what might strings will be in there so ideally you are testing on a clean OS, without any Apple ID details or anything else setup.

I suspect it would be useful to see what errors are produced as it is attached to both OS's.

Does it sound useful to have another pair of eyes look at the logs?

I have a dumb bash script that I use to grab all logs on a system, is that any use?

It dumps them into a folder next to the script & datestamps the folder.

I don't mind taking a look if you need help, I'm just not sure how to filter everything that may be revealing.

bigd_pdx wrote:

 

Interesting.  Thanks for taking the time to test.  Question for you:  Can you provide some details about how you partitioned the drive, what partition type it is, etc....

The drive I tested in my previous post is a 3TB Seagate Backup Plus USB 3 drive that I partitioned in April of 2011 with 10.6.8 Disk Utility when I purchased my MBP. The Partition Map Scheme is an Apple Partition Map with a single 3TB Mac OS Extended (Journaled) partition. I used it as my external Time Machine drive on the MBP until May of 2012 when I replaced it with a 4TB Seagate Backup Plus USB3 drive. During that time period I probably never used a FW800 base with this drive, since our only FW800 interface/base was being used on a similar Seagate Backup-Plus drive that was the external Time Machine drive for our iMac running 10.6.8.

After a clean install of Mavericks on my MBP and the iMac, I needed more work space to archive the 10.6.8 Time Machine backups to .dmg and clean up/organized the flotsam and jetsam on these TM drives for the transition to Mavericks on both Macs. So I purchased a 3TB Seagate GoFlex Desk For Mac with a 2nd FW800 interface/base and an Apple Thunderbolt-FW800 adapter to make the process go faster. That was when I began having problems mounting 2 of the volumes on 2 of our 6 Seagate drives.

-- GetRealBro

p.s. After nealry 3 decades of being at the bleeding edge of Mac OS upgrades I have learned to wait for the dust to settle. But the iMac had just come back from the Apple Store with a new GPU which cured the overheating. So it seemed like a good time to go ahaed and do a clean install of Mavericks -- wrong! I wasted the btter part of a week fighting with this mounting issue while cleaning up these drives.

Drew Reece wrote:

 

This is good news, I assume this could be reproducible? You did this without any WD software installed?

 

It would be worth grabbing logs & noting rough times that you saw the damage occur & other key events like reattaching the disk, what the volume names were etc.

 

Is this a clean 10.9 install? 

What about the 10.6.8 install?

 

If you grab all the logs there is no telling what might strings will be in there so ideally you are testing on a clean OS, without any Apple ID details or anything else setup....

I ALWAYS do clean installs of a new OS (e.g. 10.6.8 in April of 2011 and 10.9 a few weeks ago). And I don't use WD drives*.

This mounting problem is reproducible in the sense that it has happened before. Note: I reported the same “spontaneous recovery” on this same disk a few days ago when I was finally able to archive the Time Machine backups to .dmg after it mounted in 10.6.8 using the USB interface. That is why I began trying to get it to fail again.

I’ll glean through the system logs and report if I find anything interesting.

— GetRealBro

*I do own one WD drive that was attached to our Dish DVR before it began losing recordings:( Now it just sits there unplugged reminding me not to buy another WD drive.

Tony808 wrote:

 

I just used Disk Utility to Verify my back up disk. It took about an hour and reported there were no problems but more importantly when I rechecked TM and it is now working fine (as it did before Maverick)

How can "Verify Disk" repair a disk?

I thought "Verify Disk" is just a read/check and report process.

GetRealBro wrote:

The drive I tested in my previous post is a 3TB Seagate Backup Plus USB 3 drive that I partitioned in April of 2011 with 10.6.8 Disk Utility when I purchased my MBP. ...

After checking finacial records to determine exactly when I had purchased these 6 Seagates I realised that both of the 3T Backup Plus drives were purchased 1 week apart in 10-2012. The 3T problem drive contians a TM backup dating back to when I purchased the MBP (4-2011). So I must have transfrred the TM backup to this drive. That leaves me unsure of why I would have used an Apple Partition Map Scheme instead of GUID.

Which got me remembering that while cleaning up these drives I noticed that one of them had a Master Boot Record partiton scheme. When I repartitoined that drive I changed it to GUID. Unfortunately this was early on in the clean up process. So I don't remember whether this was the other Seagate that had a partion that wouldn't mount in Mavericks but would in Snow Leopard.

All of these Seagate GoFlex and Backup Plus drives (except the most recent GoFlex FW800) were formatted for use on PCs when I purchsed them. So I would have erased and partioned them with Disk Utility.

--- GetRealBro

Haven't this problem been sorted out?.... Any patch?

Despite the FW800/USB2 interface/base being brand new, I decided to verify that the firmware was up to date. I downloaded seagate-firmware-updater-1.2.2.dmg from the Seagate site and followed the instructions.

APPLIES TO:

Software Version:1.2.2

COMPATIBLE OPERATING SYSTEMS

At Step 5 after plugging in the FW800 cable…. the MBP crashed!

Here is the report that I sent to Apple...

Sat Nov 23 07:58:42 2013

panic(cpu 0 caller 0xffffff8023adc19e): Kernel trap at 0xffffff7fa5fa0d9d, type 14=page fault, registers:

CR0: 0x000000008001003b, CR2: 0x0000000000000000, CR3: 0x0000000026624000, CR4: 0x00000000000606e0

RAX: 0xffffff8024069060, RBX: 0xffffff8024102618, RCX: 0xffffff80240e2b20, RDX: 0x0000000000000006

RSP: 0xffffff80a356be50, RBP: 0xffffff80a356bea0, RSI: 0x0000000000000020, RDI: 0x0000000000000000

R8:  0x0000000000000000, R9:  0x0000000000000001, R10: 0x0000000000002630, R11: 0xffffff8024089ce8

R12: 0x0000000000000000, R13: 0x0000000000000000, R14: 0x0000000000000000, R15: 0xffffff8030fa5fa0

RFL: 0x0000000000010282, RIP: 0xffffff7fa5fa0d9d, CS:  0x0000000000000008, SS:  0x0000000000000010

Fault CR2: 0x0000000000000000, Error code: 0x0000000000000000, Fault CPU: 0x0

Backtrace (CPU 0), Frame : Return Address

0xffffff80a356bae0 : 0xffffff8023a22f69

0xffffff80a356bb60 : 0xffffff8023adc19e

0xffffff80a356bd30 : 0xffffff8023af3606

0xffffff80a356bd50 : 0xffffff7fa5fa0d9d

0xffffff80a356bea0 : 0xffffff8023e92266

0xffffff80a356bf30 : 0xffffff8023e8e119

0xffffff80a356bf70 : 0xffffff8023e93093

0xffffff80a356bfb0 : 0xffffff8023ad6aa7

BSD process name corresponding to current thread: kernel_task

Mac OS version:

13A603

Kernel version:

Darwin Kernel Version 13.0.0: Thu Sep 19 22:22:27 PDT 2013; root:xnu-2422.1.72~6/RELEASE_X86_64

Kernel UUID: 1D9369E3-D0A5-31B6-8D16-BFFBBB390393

Kernel text base: 0xffffff8023a00000

System model name: MacBookPro8,2 (Mac-94245A3940C91C80)

System uptime in nanoseconds: 25815734185040

loaded kexts:

Model: MacBookPro8,2, BootROM MBP81.0047.B27, 4 processors, Intel Core i7, 2.2 GHz, 4 GB, SMC 1.69f4

Graphics: Intel HD Graphics 3000, Intel HD Graphics 3000, Built-In, 384 MB

Graphics: AMD Radeon HD 6750M, AMD Radeon HD 6750M, PCIe, 1024 MB

Memory Module: BANK 0/DIMM0, 2 GB, DDR3, 1333 MHz, 0x80CE, 0x4D34373142353737334448302D4348392020

Memory Module: BANK 1/DIMM0, 2 GB, DDR3, 1333 MHz, 0x80CE, 0x4D34373142353737334448302D4348392020

AirPort: spairport_wireless_card_type_airport_extreme (0x14E4, 0xD6), Broadcom BCM43xx 1.0 (5.106.98.100.22)

Bluetooth: Version 4.2.0f6 12982, 3 services, 15 devices, 1 incoming serial ports

Network Service: Wi-Fi, AirPort, en1

Serial ATA Device: TOSHIBA MK7559GSXF, 750.16 GB

Serial ATA Device: MATSHITADVD-R   UJ-898

USB Device: Hub

USB Device: BRCM2070 Hub

USB Device: Bluetooth USB Host Controller

USB Device: Apple Internal Keyboard / Trackpad

USB Device: USB Laser Mouse

USB Device: FaceTime HD Camera (Built-in)

USB Device: Hub

USB Device: IR Receiver

FireWire Device: Backup+ Desk, Seagate, Up to 800 Mb/sec

Thunderbolt Bus: MacBook Pro, Apple Inc., 22.1

So, not being too bright, I tried it again

This time I plugged in the USB2 cable and…. no drama. Apparently the interface firmware (at least for the USB part of it) is up to date.

But what about the FW800 part? Has anyone used the Seagate firmware updater while running Mavericks?

— GetRealBro

BTW the "Disk Description" in Disk Utility for this drive (with the same FW800/USB2 interface/base) is quite different when mounted using the FW800 cable vs the USB2 cable.

For completeness I tried running the Seagate updater via Firewire in 10.6.8, where it also caused a kernel panic    I won't bore you with the details.

-- GetRealBro

Same problem with me. After upgrade to Mavericks the first problem start with Time Capsule in which sometimes mount normally and the backup was working and sometimes was unable to be mount correctly and it was mounted as EFI & TimeMachine and getting backup failed.

Also for few days after upgrade to Mavericks my external Lacie disk was working properly but suddenly it was repartition and renamed to MyBook and lost all of my data.

Troubleshooting my Macbook I found on my activity monitor a service from the Western Digital which was running in the background name "WDMyBookservices". The service probably was installed by an old external WD disk I was owned, but I was never install by myself any WD software on my Mac.

Upon manual remove WD service my system became stable and over the last few days I don't have any problems with TM and Time Capsule they are working fine or with other external drives.

Till today I haven't managed to recover any of my lost data on the external lacie drive

Got the same problem with 2 WB drives. One with power cord and one without.

The way around it is to

1: recover your hard disk through terminal

2: after the recovery is complete, unplug the power cord or usb cord. Plug them back it and mavericks will say that the hard disk was remove incorrectly.

After that the hard disk should open with all you files with in

How I do recover through terminal. Which command line should I use?

Sorry, not restore but mount it through terminal

sudo mkdir /Volumes/Name of your disk