New backup hard drive visible from bootable rescue CD but not visible on desktop.

After much searching I’m not able to come up with a solution.
I have a dual boot system - Windows 7 and openSUSE 13.1 with KDE and kernel version 3.12.67-61.

Last night I plugged in a 1TB Seagate Backups plus slim (BUP slim) USB drive into a USB2 port on my linux system - but it did not appear to be detected. I then booted into Windows 7 - the drive was detected there. It had been preformatted with NTFS and had a few MB worth of Seagate specific files for registration etc. I could access the files and I did the Seagate registration from Windows without any problems. However, try as I might, I could not find the drive on the Linux system via the device notifier, KDE desktop configuration or in dolphin.

The following is a summary of where the drive is detected/undetected:
0. Dolphin: NOT present

  1. KDE desktop configure: NOT present - all automatic mounting turned on.
  2. YAST partitioner - NOT present.
  3. YAST Hardware Information: PRESENT - turns up as “BUP Slim BK”, manufacturer: “Seagate” and various other details.
  4. fdisk -l : NOT present
  5. fdisk -l from bootable installation/rescue disk: PRESENT - turns up as sdh/sdh1 (I have 3 other physical hard drives sda, sdb and sdc)
  6. YAST partitioner from bootable installation/rescue disk: PRESENT (sdh/sdh1)

What is interesting is that the drive is being found in rescue mode but not in the current system - does that indicate a kernel issue?

I tried a suggestion made elsewhere. Unplug and replug the disk and look at the output of dmesg:


~> dmesg | tail
 1822.467306] lp: driver loaded but no devices found
 1822.586274] BIOS EDD facility v0.16 2004-Jun-25, 0 devices found
 1822.586284] EDD information not available.
 2494.860268] usb 1-2.2: USB disconnect, device number 6
 2512.695092] usb 1-2.2: new high-speed USB device number 9 using ehci-pci
 2512.838207] usb 1-2.2: New USB device found, idVendor=0bc2, idProduct=ab24
 2512.838221] usb 1-2.2: New USB device strings: Mfr=2, Product=3, SerialNumber=1
 2512.838228] usb 1-2.2: Product: BUP Slim BK
 2512.838234] usb 1-2.2: Manufacturer: Seagate
 2512.838239] usb 1-2.2: SerialNumber: NA96AN6G

So the device is indeed being detected in the running system.

My next thought was to try to run YAST from the rescue prompt and reformat and mount it. I did the following based on some information found elsewhere for repairing GRUB2.
(my root partition is sda6).


# mount /dev/sda6 /mnt
# mount -o bind /dev /mnt/dev
# mount -o bind /proc /mnt/proc
# mount -o bind /sys /mnt/sys
# mount -o bind /dev/pts /mnt/dev/pts
# chroot /mnt
# yast

I then tried to reformat and mount the drive via YAST but it yielded an error (error # 305 I think - there was no other information).

I don’t know what to do next. Perhaps I should try to reformat with fdisk on the rescue command prompt and then mount it. I’ve not done these sorts of things before so I’m hesitant. I want the drive in NTFS format since I may also use it from my Windows system. Is there an easy way out?

Thank you.

The most informative of what you post is the dmesg output. It should mention (amongst other lines due to you connecting the device) a line telling that it is mass-storage. When that is not the case, You will never see any steps that will follow the detection of a mass-storage device, like it getting a device file (like sdX) and of course no mount (of a not existing device) will be possible.

Thus first and foremost we have to find out why the device is not seen as being mass-storage.

In Yast Hardware Information, it is reported as “Mass Storage Device” both under “Disk” as well as under “USB” - the details are identical under both.

Bus: USB
Class (spec): Disk
Class: Mass storage device
Device Identifier: 240420
Device: BUP Slim BK


Model: Seagate RSS LLC BUP Slim BK


Vendor Identifier: 199618
Vendor: Seagate RSS LLC
hotplug: usb

I tried plugging it into several USB ports - including directly into one of the ports at the back of the desktop computer - but dmesg gives the same result.
Currently it is plugged in to a USB2 port on the Dell monitor.
Oddly enough, the vendor identifier seems to change every time it is plugged in. [Edited: Sorry ignore that, the vendor identifier is the same]

It also shows in the “Block Devices” section in Yast Hardware Information as “Mass Storage Device” with the same parameters.

Hm, strange. The fact that the hardware info tells you it is disk/mass-storage should be the result mentioning of logging of the detection as such in dmesg.
How can it not be in dmesg and nevertheless in the hardware info???

Your dmesg also does not say anything about creating a device file (/dev/sdX). Is there nevertheless one created?

I no longer have a 13.1 system to test this.

I do have a Seagate 2T slim backup drive, and it is working fine (with opensuse 42.2).

Note that the KDE defaults are to not automatically mount. Instead, it should show up in the device notifier in the tray. From the device notifier, you should be able to mount.

The live rescue system (assuming that this is the opensuse rescue system) uses XFCE as a desktop, and XFCE does automatically mount.

I have. And it is up-to-date. And the device notifier in KDE pops up.

Thus I guess it is specific to the Op’s device/system/whatever. Not easy to solve I am afraid.

I tried:

tail -f /var/log/messages

Then unplugged and replugged the drive from the usb port (directly at the back of the computer). I get:

2016-12-13T20:46:19.823672+05:30 linux kernel:   162.979496] usb 1-1: USB disconnect, device number 2
2016-12-13T20:46:24.846113+05:30 linux kernel:   167.997858] usb 1-1: new high-speed USB device number 7 using ehci-pci
2016-12-13T20:46:24.964104+05:30 linux kernel:   168.115991] usb 1-1: New USB device found, idVendor=0bc2, idProduct=ab24
2016-12-13T20:46:24.964135+05:30 linux kernel:   168.116005] usb 1-1: New USB device strings: Mfr=2, Product=3, SerialNumber=1
2016-12-13T20:46:24.964142+05:30 linux kernel:   168.116012] usb 1-1: Product: BUP Slim BK
2016-12-13T20:46:24.964148+05:30 linux kernel:   168.116018] usb 1-1: Manufacturer: Seagate
2016-12-13T20:46:24.964153+05:30 linux kernel:   168.116023] usb 1-1: SerialNumber: NA96AN6G
2016-12-13T20:46:24.970030+05:30 linux mtp-probe: checking bus 1, device 7: "/sys/devices/pci0000:00/0000:00:12.2/usb1/1-1"
2016-12-13T20:46:24.971282+05:30 linux mtp-probe: bus: 1, device: 7 was not an MTP device

If I do it again, I get exactly the same results except device number 7 is changed to device number 8.
Not sure if that is useful.
Thanks.

Looking up “MTP device” on-line tells me that the drive is being treated as “Media Transfer Protocol” device - perhaps like a drive on a mobile phone. Why is that?
There is a somewhat old thread on this forum:

https://forums.opensuse.org/showthread.php/479571-Can-t-mount-usb-disk

with the same problem - but the original poster on that thread had other issues so it was not resolved. A suggestion from this thread (that I don’t quite follow)

It appears that the the mtp-probe (from libmtp udev rules) is incorrectly grabbing your device. Do you have any MTP devices that you connect to? If not, one workaround I can think of is to remove libmtp9 for now. A bug report should certainly be considered.

It may also be possible to write a udev rule to grab this this device (by its vendor and product ID’s) ahead of the problematic MTP rule(s). I could try and assist if necessary.

Thanks.

Hm, it looks as if we are getting somewhere, but it is still a bit of a riddle.

It tests for MTP and then concludes it is not. But why does it not handle it as mass-storage.

It reminds me a bit of cameras that are set to PTP (picture transfer protocol) and that you then can switch to mass-storage so the system will see it as such. But I think the comparison will stops here.

I found an expert piece of hackery at the link below:

Yet another Seagate quirk for unusual_uas.h
http://www.spinics.net/lists/linux-usb/msg137992.html

Following advice on the web I created file ‘/etc/modprobe.d/ignore_uas.conf’ containing the line “options usb-storage quirks=0bc2:331a:u”.
The disk could then be mounted and used without any problems.

I’m wondering if I should implement it. I just need to change the device id from 331a to ab24 for my device.

Well, try it. When it does not work for you, it is easy to remove the file again.

Of course allays make it easy to revert to the old situation when testing things. But in this case you are not altering a file, so no need to create a backup of it.

It worked!

Now fdisk -l gives the additional output:


Disk /dev/sdd: 1000.2 GB, 1000204885504 bytes, 1953525167 sectors                                                                                                                      
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk label type: dos
Disk identifier: 0x21911dd8

   Device Boot      Start         End      Blocks   Id  System
/dev/sdd1   *        2048  1953525166   976761559+   7  HPFS/NTFS/exFAT

Also the drive can be seen in YAST partitioner with a single partition - NTFS formatted but not mounted.

So I guess all I have to do now is to mount the drive - say via the YAST partitioner and I’m ready to go?

Also it got mounted at sdd instead of sdh, befitting its status as a hard drive I guess. Also, YAST hardware information has some additional fields such as “Device Names” and “Device Numbers”.

It now being recognised as mass storage it should give the loged in KDE user a popup where you can choose what you want to do with it (e.g. open with Dokphin).

Does that happen?

I do not uderstand why you think it should be /dev/sdh. Do you have so many devices already?

It depends on how you want to use it.
If it is only to be used by the end-user from the desktop, leave it to the desktop to mount it.

When it must be there always, use YaST to create an fstab entry.

I shouldn’t have used the word “mounted”. I mean it got assigned to sdd.
Before it got recognized as a mass storage device it got assigned to sdh - which seemed a bit strange - the reason maybe that there are many other USB devices - Mouse and keyboard dongles, Mic/Headphone, a small hub, scanner etc. But now that it has been recognized as a Mass storage device, it seems to be placed appropriately next to the three other hard drives.

Sorry I did not see this sooner.

In openSUSE 13.1, the newest external HDs (I have observed this with an external Seagate 3-TB) call for the fast UAS.

I suggest that you temporarily remove that file you added. You can always add it back.

When you have done that, try plugging in the device. Again, it will not show up at this point, but then issue the following command:

modprobe -v uas

The drive should suddenly show up. If it does, you need to go to /etc/modprobe.d/50-blacklist.conf and remove the line:

blacklist uas

That was put there to solve an earlier bug that has long since been patched in 13.1, but when it was patched the blacklist line remained because it was forgotten.

Hm, that sounds as a better solution to me, because the other one looked more like a bypass.

Yes. I tested both ways, and in my case the data transfers blazingly fast using UAS.

Just saw this post. Thank you. It did work. I’ll check the speed on my next backup.

@Henk: Thanks for your help.