Please help! USB pen drive is fscked... any advice for recovery?

[Feathers McGraw]

The USB pen drive that holds my Pi's root filesystem has given up the ghost.

When I try to boot from it I get messages like this on screen:

usb-storage 1-1.3:1.0: USB mass storage detected
scsi0 : usb-storage 1-1.2:1.0
scsi 0:0:0:0: Direct-Access          Corsair      Voyager Mini 3.0 PMAP PQ: 0 ANSI: 6
sd 0:0:0:0: [sda] 60566016 512-byte logical blocks: (31.0 GB/28.8 GiB)
sd 0:0:0:0: [sda] Write Protect off
sd 0:0:0:0: [sda] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA
usb 1-1.3: reset high speed USB device number 4 using dwc_otg
usb 1-1.3: device descriptor read/64, error -110
usb 1-1.3: device descriptor read/64, error -110
usb 1-1.3: reset high speed USB device number 4 using dwc_otg
usb 1-1.3: device descriptor read/64, error -110
usb 1-1.3: device descriptor read/64, error -110
usb 1-1.3: reset high speed USB device number 4 using dwc_otg
usb 1-1.3: device descriptor read/64, error -110
usb 1-1.3: device descriptor read/64, error -110
usb 1-1.3: reset high speed USB device number 4 using dwc_otg
usb 1-1.3: device descriptor read/64, error -110
usb 1-1.3: device descriptor read/64, error -110
sd 0:0:0:0: Device offlined - not ready after error recovery
sd 0:0:0:0: [sda] Unhandled error code
sd 0:0:0:0: [sda] Write Protect off
Result: hostbyte=0x01 driverbyte=0x00
sd 0:0:0:0: [sda] CDB:
cdb[0]=0x28: 28 00 00 00 00 00 00 00 08 00
end_requestL I/O error, dev sda, sector 0
Buffer I/O error on device sda, logical block 0
  sda: unable to read partition table
sd 0:0:0:0: [sda] Attached SCSI removable disk
usb 1-1.3: new high-speed USB device number 5 using dwc_otg
...

and so on.

It doesn't show up when I plug it into my kubuntu laptop, and doesn't appear if I do lsusb...

I'm really hoping I can recover the data from the drive, it had my email server and website on it! The irony is that if I have to redo it all I can't use my own tutorials, because they're on the USB... lol.

I have had a look through this: https://help.ubuntu.com/community/DataRecovery but most of the commands seem to require pointing at /dev/sdX and I can't get Kubuntu to recognise the device at all, so I'm not sure they are going to work... any ideas?

Feathers

[oshunluvr]

dmesg will show some activity when you plug in the stick. If it doesn't - it's dead. If it does, dd can make a copy, but that may not help you.

You do have a backup, right?

[Feathers McGraw]

[ 1434.443043] usb 3-3: new SuperSpeed USB device number 4 using xhci_hcd
[ 1434.460735] usb 3-3: New USB device found, idVendor=1b1c, idProduct=1a0c
[ 1434.460745] usb 3-3: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 1434.460751] usb 3-3: Product: Voyager Mini 3.0
[ 1434.460755] usb 3-3: Manufacturer: Corsair
[ 1434.460759] usb 3-3: SerialNumber: 070839F75F1DDF20
[ 1434.461732] usb-storage 3-3:1.0: USB Mass Storage device detected
[ 1434.461874] scsi5 : usb-storage 3-3:1.0
[ 1435.461017] scsi 5:0:0:0: Direct-Access     Corsair  Voyager Mini 3.0 PMAP PQ: 0 ANSI: 6
[ 1435.461578] sd 5:0:0:0: Attached scsi generic sg1 type 0
[ 1435.461778] sd 5:0:0:0: [sdb] 60566016 512-byte logical blocks: (31.0 GB/28.8 GiB)
[ 1435.462124] sd 5:0:0:0: [sdb] Write Protect is off
[ 1435.462131] sd 5:0:0:0: [sdb] Mode Sense: 45 00 00 00
[ 1435.462752] sd 5:0:0:0: [sdb] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA
[ 1466.041777] usb 3-3: reset SuperSpeed USB device number 4 using xhci_hcd
[ 1471.063217] usb 3-3: device descriptor read/8, error -110
[ 1471.167291] usb 3-3: reset SuperSpeed USB device number 4 using xhci_hcd
[ 1476.188797] usb 3-3: device descriptor read/8, error -110
[ 1476.349071] usb 3-3: USB disconnect, device number 4
[ 1476.349078] sd 5:0:0:0: Device offlined - not ready after error recovery
[ 1476.349084] sd 5:0:0:0: [sdb] Unhandled error code
[ 1476.349095] sd 5:0:0:0: [sdb]  
[ 1476.349097] Result: hostbyte=DID_ABORT driverbyte=DRIVER_OK
[ 1476.349099] sd 5:0:0:0: [sdb] CDB: 
[ 1476.349100] Read(10): 28 00 00 00 00 00 00 00 08 00
[ 1476.349107] end_request: I/O error, dev sdb, sector 0
[ 1476.349110] Buffer I/O error on device sdb, logical block 0
[ 1476.349158] ldm_validate_partition_table(): Disk read failed.
[ 1476.349181] Dev sdb: unable to read RDB block 0
[ 1476.349210]  sdb: unable to read partition table
[ 1476.349306] sd 5:0:0:0: [sdb] Attached SCSI removable disk
[ 1476.351204] xhci_hcd 0000:00:14.0: xHCI xhci_drop_endpoint called with disabled ep ffff88030967ee80
[ 1476.351208] xhci_hcd 0000:00:14.0: xHCI xhci_drop_endpoint called with disabled ep ffff88030967eec0
[ 1476.581294] usb 3-3: new SuperSpeed USB device number 5 using xhci_hcd
[ 1481.602878] usb 3-3: device descriptor read/8, error -110
[ 1481.706925] usb 3-3: new SuperSpeed USB device number 5 using xhci_hcd
[ 1486.728418] usb 3-3: device descriptor read/8, error -110
[ 6896.325321] usb 3-3: new SuperSpeed USB device number 7 using xhci_hcd
[ 6896.342992] usb 3-3: New USB device found, idVendor=1b1c, idProduct=1a0c
[ 6896.343002] usb 3-3: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 6896.343007] usb 3-3: Product: Voyager Mini 3.0
[ 6896.343011] usb 3-3: Manufacturer: Corsair
[ 6896.343015] usb 3-3: SerialNumber: 070839F75F1DDF20
[ 6896.344043] usb-storage 3-3:1.0: USB Mass Storage device detected
[ 6896.344214] scsi6 : usb-storage 3-3:1.0
[ 6897.343506] scsi 6:0:0:0: Direct-Access     Corsair  Voyager Mini 3.0 PMAP PQ: 0 ANSI: 6
[ 6897.344054] sd 6:0:0:0: Attached scsi generic sg1 type 0
[ 6897.344398] sd 6:0:0:0: [sdb] 60566016 512-byte logical blocks: (31.0 GB/28.8 GiB)
[ 6897.344735] sd 6:0:0:0: [sdb] Write Protect is off
[ 6897.344741] sd 6:0:0:0: [sdb] Mode Sense: 45 00 00 00
[ 6897.345230] sd 6:0:0:0: [sdb] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA

My last backup is pretty old, I made it 3 months ago

Ironically, I was going to make another one once ModSecurity was up and running, which I managed last night. Woke up this morning to find it was dead. I reckon all those writes from ModSecurity log files might have pushed it over the edge...

So, what's the best plan for trying to recover some stuff?

I'm going to have to invest some time learning RAID or something similar when I rebuild.

[Feathers McGraw]

feathers-mcgraw@Hobbs-T440s:~$ sudo ddrescue -r 3 /dev/sdb /home/feathers-mcgraw/websiteimage /home/feathers-mcgraw/websiteimage_log
ddrescue: Can't open input file: No medium found

ho hum

[oshunluvr]

Lets assume the partition table is bad, as the error above states, but the data is intact. One could rewrite/correct the partition table and gain access to the data again or transfer only the data portion to a functioning drive.

[Feathers McGraw]

OK, sounds good. I still don't really know how to reference the drive, I mean is it /dev/sdb like it usually is because that doesn't seem to be working :/

[oshunluvr]

open a terminal, type dmesg, look at output, insert drive, type dmesg again, look at output again, note assigned device label.

If there's isn't a device label, you're SOL.

[GreyGeek]

Back in the days of batch processing. after I explained "backing up" and how to do it, my clients often asked "How often should we back up?" My answer was "How much stuff do you want to re-enter?"
For some once a night was sufficient. For others twice a day, once at noon and then that night. On really big installations full backups were done only once a week, during the weekend, and daily backups were incremental because there wasn't enough time for a full backup.

But, aren't mechanics the ones who drive junker cars?

[SteveRiley]

Ooooh boy! The grim data reaper really came after you, eh? USB thumb drives are not designed for primary storage. This lesson is, of course, best learned when one has no backups. Perhaps this will be helpful: https://web.archive.org/web/*/http://www.samhobbs.co.uk

Dude. So sorry this happened. That's a lot of high quality work you did. I hope you can reconstruct most of it.

[Feathers McGraw]

Thanks guys, looks like I'll just have to redo it. I've grabbed the tutorials from Google's cache of the site because it's newer and more complete than the web archive, and I'll put a version back up when I've rebuilt.

Shame about the comments, there were a lot of useful bits of info people had posted in there.

At least it's a kick up the bum to get started Drupal, hopefully the end product will be better than what I had before!

But, it looks like I'll need some new hardware if I want to run Drupal... so the question is basically this:

Should I buy an Intel NUC or a Chromebox?

The NUC is quite expensive, but available now and I could easily expand and modify it later if I wanted to. Chromeboxes are much cheaper but I'd have to wait, plus they're less easily expanded, and the BIOS seems to have an annoying quirk where it requires you to push space when you power it on if you're running replacement software... or it won't boot, which could be a real pain for a headless unit. I might be able to do away with that by flashing Coreboot or something similar, but I'm not certain.

[Feathers McGraw]

Ordered myself an NUC, the main factor in my decision being expandability, for example it can take 2.5" drives so I can buy a reasonably cheap SSD and shove that in there, plenty of space for RAM etc.

I'm going to rebuild my email server on a Pi and use the NUC for HTTP, that way there's a neat separation of use, if the NUC server is down for maintenance (which ofc never happens ) then my emails will be unaffected.

I should look into virtualisation really, that stuff is pretty cool.

[oshunluvr]

Start a new thread when you get your new NUC. I'm interested in hearing what your experiences are, specs, etc.

I currently have a low-power server for media delivery (mostly) but I'm thinking at next rebuild I might switch to a high power server and low power work stations. Some of the newer NUC's coming out might be just the ticket.

[Feathers McGraw]

Will do. This is the one I've ordered:

http://www.intel.com/content/www/us/...54250wykh.html

[Feathers McGraw]

Hmm...

I had two RasPi servers, one running the website and mail server, the other running OwnCloud.

I just removed the OwnCloud USB drive, reformatted it and isntalled Raspbian again, thinking I would use this USB to store the mail server's root FS. It's really slow, taking absolutely ages to just upgrade a package.

I think it's because this USB drive (the same make, bought at about the same time) is nearing its end of life.

Is there a way I can see the level of wear on a USB flash drive or SD card?

[oshunluvr]

AFAIK, no. Generally, SD cards and USB drives are damaged physically or thrown away before they wear out. You encountered a manufacturing defect IMO. I've got a 5 year old Sandisk 4GB drive that's been through the laundry three times among other careless treatments at my hand and it's still in daily use.

If I had to bet, I'd bet that other drive will last longer than you will want it to.

Curiosity: What file formats are you using on these drives that you use daily?