External Drive UNMOUNTING On Its Own...

Hello,

I've got the same problem with a hard disk used to contain Aperture and iTunes library plug on my Macbook.
While a Time Machine Backup, a warning message appears to signal that the hard disk was unproperly unmount.
I noticed two different problem :
- Eject my working hard disk instead of the backup drive. I think the backup drive stay mounted.
- While the backup, the working hard drive unmount & the backup failed (cf log below)
.
.
.
.
+20/11/10 23:04:59 com.apple.backupd[1016] Copied 40.5 GB of 44.6 GB, 0 of 87 items+
+20/11/10 23:17:34 ntpd[27] time reset -0.220601 s+
+20/11/10 23:26:09 kernel USBF: 40387. 48 AppleUSBEHCI[0x56b5000]::Found a transaction past the completion deadline on bus 0x24, timing out! (Addr: 3, EP: 1)+
+20/11/10 23:26:15 kernel USBF: 40393. 48 AppleUSBEHCI[0x56b5000]::Found a transaction which hasn't moved in 5 seconds on bus 0x24, timing out! (Addr: 3, EP: 0)+
+20/11/10 23:26:15 kernel disk1s2: I/O error.+
+20/11/10 23:26:15 kernel +
+20/11/10 23:26:15 kernel +
+20/11/10 23:26:16 com.apple.backupd[1016] Error: (-36) SrcErr:YES Copying /Volumes/Aperture/iTunes Music/Mobile Applications/France 1.5.ipa to (null)+
+20/11/10 23:26:46 kernel USBF: 40424. 48 AppleUSBEHCI[0x56b5000]::Found a transaction past the completion deadline on bus 0x24, timing out! (Addr: 3, EP: 1)+
+20/11/10 23:26:46 kernel disk1s2: device/channel is not attached.+
+20/11/10 23:26:46 kernel +
+20/11/10 23:26:46 kernel +
+20/11/10 23:26:46 kernel disk1s2: device/channel is not attached.+
+20/11/10 23:26:46 kernel +
+20/11/10 23:26:46 kernel +
+20/11/10 23:26:46 com.apple.backupd[1016] Error: (-36) SrcErr:YES Copying /Volumes/Aperture/iTunes Music/Mobile Applications/Fruit Ninja 1.5.ipa to (null)+
+20/11/10 23:26:46 com.apple.backupd[1016] Error: (-36) SrcErr:YES Copying /Volumes/Aperture/iTunes Music/Mobile Applications/Google 0.7.1.ipa to (null)+
+20/11/10 23:26:46 kernel disk1s2: media is not present.+
+20/11/10 23:26:46 kernel +
+20/11/10 23:26:46 kernel +
+20/11/10 23:26:46 kernel disk1s2: media is not present.+
+20/11/10 23:26:46 kernel +
+20/11/10 23:26:46 kernel +
+20/11/10 23:26:46 kernel disk1s2: media is not present.+
+20/11/10 23:26:46 kernel +
+20/11/10 23:26:46 kernel +
+20/11/10 23:26:46 kernel disk1s2: media is not present.+
+20/11/10 23:26:46 kernel +
+20/11/10 23:26:46 kernel +
+20/11/10 23:26:47 kernel disk1s2: media is not present.+
+20/11/10 23:26:47 com.apple.backupd[1016] Error: (-36) SrcErr:YES Copying /Volumes/Aperture/iTunes Music/Mobile Applications/Le Monde.fr 2.104.ipa to (null)+
+20/11/10 23:26:47 kernel +
+20/11/10 23:26:47 kernel +
+20/11/10 23:26:47 kernel disk1s2: media is not present.+
+20/11/10 23:26:47 kernel +
+20/11/10 23:26:47 kernel +
+20/11/10 23:26:47 kernel disk1s2: media is not present.+
+20/11/10 23:26:47 kernel +
+20/11/10 23:26:47 kernel +
+20/11/10 23:26:48 kernel jnl: disk1s2: close: journal 0x5776d08, is invalid. aborting outstanding transactions+
+20/11/10 23:26:48 com.apple.backupd[1016] Stopping backupd because a backup source was ejected!+
+20/11/10 23:26:48 com.apple.backupd[1016] Copied 47 files (44.9 GB) from volume Aperture.+
+20/11/10 23:26:51 com.apple.backupd[1016] Backup canceled.+
+20/11/10 23:26:52 com.apple.backupd[1016] Ejected Time Machine disk image.+
+20/11/10 23:26:54 com.apple.backupd[1016] Ejected Time Machine network volume.+
+20/11/10 23:27:02 hdiejectd[783] quitCheck: calling exit(0)+

Message was edited by: Jimbo-77

I'm now sure it's not a Time Machine issue.
In my next post the log indicate a problem on the copy of files in the /Volumes/Aperture/iTunes Music/Mobile Applications/ directory.
I tried to manually copy this directory and I had the same usb error. I think it's an usb driver issue.

People have this problem, your Mac boot in 64 bits mode ?

The log :
25/11/10 20:35:54 kernel USBMSC Identifier (non-unique): 000000000033 0x5e3 0x718 0x9
25/11/10 20:37:22 kernel USBF: 5140. 96 AppleUSBEHCI[0x5680000]::Found a transaction past the completion deadline on bus 0x26, timing out! (Addr: 2, EP: 1)
25/11/10 20:37:27 kernel USBF: 5146. 97 AppleUSBEHCI[0x5680000]::Found a transaction which hasn't moved in 5 seconds on bus 0x26, timing out! (Addr: 2, EP: 0)
25/11/10 20:37:27 kernel IOUSBMassStorageClass[0x91a3000]: The device is still unresponsive after 6 consecutive USB Device Resets; it will be terminated.
25/11/10 20:37:27 kernel disk1s2: device/channel is not attached.
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel disk1s2: device/channel is not attached.
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel disk1s2: device/channel is not attached.
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel disk1s2: device/channel is not attached.
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel jnl: disk1s2: do jnlio: strategy err 0x6
25/11/10 20:37:27 kernel jnl: disk1s2: end_transaction: only wrote 0 of 53248 bytes to the journal!
25/11/10 20:37:27 kernel disk1s2: media is not present.
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel disk1s2: media is not present.
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel disk1s2: media is not present.
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel
25/11/10 20:37:27 kernel jnl: disk1s2: close: journal 0x5776e04, is invalid. aborting outstanding transactions

Just another "me too" to add to this thread.

I have 2 external eSATA RAID 1 drives (Venus DS-DS3RPRO), both connected to a Sonnet ExpressCard34 card in my 2006 MacBook Pro.

I have never seen this behavior until I upgraded to 10.6.5. While I was sleeping, at 7:47 AM today both of the external eSATA disks 'disappeared' to the OS completely, with the following messages:

Dec 4 07:47:22 nightowl kernel[0]: disk3s3: device/channel is not attached.
Dec 4 07:47:23 nightowl kernel[0]: disk3s3: media is not present.
Dec 4 07:47:24: --- last message repeated 4 times ---
Dec 4 07:47:23 nightowl kernel[0]: jnl: disk3s3: do jnlio: strategy err 0x6
Dec 4 07:47:23 nightowl kernel[0]: jnl: disk3s3: write journalheader: error writing the journal header!
Dec 4 07:47:23 nightowl kernel[0]: disk3s3: media is not present.
Dec 4 07:47:23 nightowl kernel[0]: jnl: disk3s3: do jnlio: strategy err 0x6
Dec 4 07:47:23 nightowl kernel[0]: jnl: disk3s3: end_transaction: only wrote 0 of 102912 bytes to the journal!
Dec 4 07:47:23 nightowl kernel[0]: disk3s3: media is not present.
Dec 4 07:47:25: --- last message repeated 8 times ---
Dec 4 07:47:25 nightowl kernel[0]: disk3s3: media is not present.
Dec 4 07:47:28: --- last message repeated 44 times ---
Dec 4 07:47:28 nightowl kernel[0]: disk2: I/O error.

When I woke up almost 4 hours later, the screen saver was in effect and I could not get the system to respond. I could not ping it from my iPhone or my other PowerBook G4. I had to hold the Power button in to turn it off to get the system back.

At the time it died I had a reasonably high I/O rate - I was downloading something overnight at roughly 1 Mbyte/sec. But it had been doing that for a good 6-8 hours already, no reason why it should just stop seeing the disks and go off-radar completely. Very disconcerting.

I'm having a similar problem; my Western Digital external drive (which I use for Time Machine backups) just started unmounting itself from the desktop (improperly) today, and then remounting itself, then unmounting, then remounting ... lather, rinse, repeat, about every thirty seconds to a minute.

Earlier today I finally agreed to a bunch of system updates I'd been putting off, including 10.6.6 and the new iTunes, among others. It didn't start happening immediately after the upgrade, but it did start happening this afternoon.

Plugging the drive into another iMac (3.06 Intel Core Duo, running 10.5.8), using the same USB cable and power cord, the drive works just fine.

drmk wrote:
. . .
Plugging the drive into another iMac (3.06 Intel Core Duo, running 10.5.8), using the same USB cable and power cord, the drive works just fine.

Your description does sound like a problem with the connection. Try different ports, cables, and combinations of the two, and make sure all connections are snug and secure: a plug that works fine in one port may not make good contact in another one.

I did try different USB ports in the computer where I was having the problem, and none of them worked. The cable worked just fine on the other iMac.

The plot has thickened, though, as the drive is now hiccuping on the other iMac. I think the drive might just be failing. I've already ordered a new backup drive. Sigh.

It could also be the cable. You'd be surprised how often that's the case -- a pin is just slightly bent, or a tiny fraction too short, or has a speck of foreign matter on it, or a wire inside it is broken, etc. They'll work fine in some cases, intermittently in others. Sometimes just nudging the cable, several inches away from either plug, will cause intermittent problems.

Same problem for me. I'm on OS X 10.6.6, and I noticed that my Western Digital external hard drive had unmounted itself 10 days ago-- it's used for Time Machine, and I hadn't noticed. I formatted it, and used a brand new WD 1 tb drive as my new time machine (the old one became my iTunes library).

So here I am. The old one worked for a day, unmounted and won't remount, at least in Finder. It shows up in disk utility, which assures me it's working fine. I've tried different USB cables, a firewire cable (it has both), and different ports to no avail.

And now my brand new hard drive, the time machine one, has unmounted too.

Help!

Another me too. I have a Newer Technology 1.5TB Raid external which was connected to a PowerMac Dual G5 running OSX 10.5.8 via Firewire 800. This setup worked seamlessly for approximately 9 months. The main use in this setup was for Time Machine backups of the G5 and other networked Macs.

Early this month the drive and cable setup were connected (also via Firewire 800) to a new iMac i7 running Mac OSX 10.6.6. In this setup the drive is mainly used as a Time Machine destination. Approximately every 1 to 2 days the drive spontaneously un-mounts and disappears from the Mac desktop and to Disk Utility. Only after a reboot and power reset of the drive will it remount. I have not definitively pinpointed it, but it seems that the unmount occurs when Time Machine finishes up. I will try to monitor specifically.

I will also try some cable swaps, but don't feel it is the problem as it worked well with the G5 Mac.

I had issues with LaCie drives failing to mount ... turns out the power supplies were failing, and now i have a bag of faulty power supplies as each one was replaced. I switched to G-Tech anticipating more reliability, and everything fine until this same unmounting problem appeared. I'd had the drives going for days as i was doing massive data transfer, and then poof one day three out of four drives daisy chained on fw800 just disappeared, even without a "failure to eject" message. I powered them down, powered up again, they came back, but disappeared again later. I've now put two drives per fw800 port of the macpro (2.63 ghz, dualcore, osx 10.6.6) to see if that helps and will report back here. Could be the G-tech?

I should have done a search and found this thread before posting about the same issue.

I am another 'me2'. I recently destroyed a Maxtor 500M USB disk that I used for TM backups that was detaching itself nearly every day.

Then yesterday I bought a LaCie 1T USB disk to do the job. Only a few hours after writing my files into the TM format I slept my machine. When I awoke my iMac I got the bad dismount message and warning and indeed my LaCie disk was missing from the desktop. The cable that came with the disk is very short with a noise filter ferrite core attached. I really don't think the cable is the issue here.

I also have a WD Studio disk attached via FW800 (through a Belkin FW800 hub) with automatic backup software (WD SmartWare) and it does not eject itself like the TM disk.

If my computer is cutting power to the USB ports intermittently (which is my guess right now) I will find out. I am about to attach another drive to one of the other USB ports and see if it ejects at the same time as the TM drive - and report back.

Yesterday I got a tip from Kappy that may or may not be a solution. So far I have had nothing but success, so here's hoping.

His suggestion is to go into the "Energy Saver" preference pane in System Preferences. Check to see if you have enabled the item: "Put Hard Disk(s) to sleep whenever possible." If it is enabled, disable it. This item is enabled by default but it is inappropriate for Time Machine use, so it should be disabled.

I just wish that Apple had built in a warning that would come up when Time Machine is started up telling the user that this setting should be done and provide a link that when clicked would take the user to that particular setting.

I'm have the same issue as well.
I purchased an Iomega 1TB eGo Helium 3 weeks ago.I'm new to Apple.I've spent the last 2 weeks researching why the external HD keeps ejecting.I'm using it solely as Time Machine back up.I thought it was me.I took my iMac to the Genius Bar.The HD appeared to be okay.But after getting home I have the same issues again.

I'll try the fix mentioned above.I thought I was going crazy,until I discovered all these threads.

No love!

I got up today and fired up my iMac and was greeted with the alert saying my TM drive had been ejected improperly. So with all the suggestions by the experts here, it still happens.

OK, nothing seems to work. I would suggest that this is a problem with the OS. Does anyone know any way to alert Apple that this is happening increasingly to its users (besides sending a message to the "black hole" called "Apple Feedback?")

I should have spotted this thread before launching my own (entitled "Disk not ejected properly").

I'm getting the same behavior (unmounting during an extended backup) from a pair of 500 GB Iomega drives that worked fine with +Time Machine+ on my old Desktop G5 running Mac OSX 10.5.x. I'm on a new iMac now, running 10.6.6, and both drives misbehave. Have tried reformatting & swapping cables. Pattern continues. I have posted the error codes from Console on my thread.

This feels like an Apple problem connected to Snow Leopard's implementation of TM. Too many posts here for random disk or cable failures, especially as some posters can link the behavior to an update.