I have 2 external hard drive usb connected that I use to rotate my system backups. This weekend one of the drives is no longer recognized. The drive is a WD My Book 8TB drive and has been working fine for approximately a year. The drive is formatted as NTFS. It doesn’t show up in Files application, however if I plug drive into Windows 10 is seems to be fine.
In the Disks application the drive shows up but shows unknown partition type, unknown contents, and Label is default Partition 1, but in Gparted application the drive appears to be fine in that it shows the File System type NTFS, drive Label, drive size,used space and unused space.
Problem is that I still can use the drive.
Another thing I have noticed for the last few years is that in Files app the option to “Safely Remove Drive” only appears in about half of the kernel releases. It is just my guess that it is the kernel release and not something else. This is something that changed from update to 5.12.11-200.fc33.x86_64 from 5.12.10 which I applied Saturday morning. Without the Safely Remove option Files just unmounts the drive. With USB drives I have found that a lot of the time there is still date in buffers that hasn’t been written. I try to use the sync command a couple of times but am not sure if it has any affect on USB drives.
Any help resolving this would be appreciated…
Edward
@edwardc01:
I guess at this point I’m wondering if this is just a GUI thing, or if linux actually can’t mount ntfs for some reason. So, try it from the terminal and see if it can be identified as a drive at all and THEN if it’ll successfully mount, GUI-related stuff aside…
When it’s plugged into the linux box, does it show up when you run lsblk in the shell?
If that output is a little weird to parse through, since you know it’s a WD: sudo fdisk -l | grep -i “model”
This is my output, for example…
Disk model: Samsung SSD 970 EVO Plus 1TB
Disk model: My Passport 259F
If that works, try mounting it manually… Per best practices, make sure a folder exists under /mnt ls /mnt ← if no folder exists, or specify one:
mkdir /mnt/my_wd_drive mount -t ntfs /dev/sdx /mnt/my_wd_drive ← (assuming that works without returning an error) ls -lht /mnt/my_wd_drive
Since the drive is ntfs formatted I would use the windows disk manager to check that all is well. The tool in windows is called chkdisk I believe and it will clean up things that cause linux to hiccup. Once that is done, if there are still problems in linux then more trouble shooting may be needed, but that is the first step.
Use the native OS tools to check and fix the drive.
The drive shows up in lsblk and I can run fdisk. It shows the proper drive Model. I an also manually mount the drive and see all the files, however it tells me the drive was unclean and was not safely removed from Windows. However the drive does not auto mount when it is plugged in as it did before last weekend. I have an external Seagate drive and another WD MyBook drive that both work.
Could this be related to the Safely Remove option that sometimes appears in Files and then when a new kernel is loaded the Files option to Safely changes to unmount? Just asking since this has been happening for a couple of years now but I haven’t seen miss behavior on the part of my other NTFS drives.
You may want to try ntfsfix -n
to dry run ntfsfix
or ntfsfix -d
to clear the dirty flag
Unmount the drive and then see if it automounts.
You could also try doing things by hand (in a terminal) and see if you have the same issue. sync udisksctl unmount -b /dev/sdXXXXXXX udisksctl power-off -b /dev/sdXXXXXXX
I don’t know if it’s related to the safely remove option or not.
Failure to properly disconnect a device in windows will leave the dirty flag set on the drive. That will cause it to not auto-mount in linux.
Reconnect it to the windows machine.
Run chkdisk. Then properly disconnect it from windows.
I think the problems will then go away.
This is only one example of why it is always a good idea to properly dismount & disconnect removable drives regardless of what OS you are using them in.
The problem was resolved. Windows quit recognizing the drive but I could still see it in Fedora and mount the ntfs partition so I copied the data from the drive to other drives and then reformatted the drive in Fedora. Last weekends backup ran to the drive with no problems. Thanks for all the assistance. I learned a few things about how to troubleshoot this problem with assistance from the community, Thanks.
Still unresolved is the fact that the safely remove option in Fedora is sometimes there and other times the only option is unmount. Still not sure if that had anything to do with my problem but it came up after the drive was mounted and the “safely remove” option was not available.
In fedora, if the umount command works, or the file manager unmounts it, that means no writes are pending and it is clean. The caveat with the file manager is that it really gives no indication that anything is being done so you have to wait until the icon disappears in the locations column which indicates it is cleanly unmounted. If the icon is still there then something is still active with it and it has not cleanly unmounted.