Just read you last post. Hadn’t refreshed my browser. No sda6. /dev/sda4 is correct.
As per last post, just having issues unmounting.
That’s bad, the partition should be unmounted instantly after you detect data loss. Otherwise the system might write something and overwrite “empty” space. Try to find what process uses sda4 (google for commands) and kill those processes. Otherwise maybe just force unmount. Also, I hope you’re recovering files to a different external drive?
That’s what I thought. I have not used the drive since loss 5 days ago. But because it needs to be connected, there are processes running, ie wd stuff. Yes backing up to large usb drive 3tb. I delete what is not necessary. Will keep working with photorec for the meantime. Last resort is to kill processes.
I could unmount sda4 with no problem. External drive shouldn’t affect it. You shouldn’t worry about the running WD services if you want to recover your data. Longer the partition is mounted more risk that those files will be permanently overwritten. Extundelete is able to recover files with their full name and meta data, photorec is only recognizing files by their file start. Some files might be incorrect if they were partial (it means that some data was placed at the different disk sectors etc). So I would recommend to try extundelete as first and then if it fails - photorec. Once again, extundelete documentation says that the partition must be unmounted before it will work.
You could use the lsof command. This will list files opened by processes.
I’m not sure whether you think the ‘6’ bit in the sda6 is anything to do with the size.
But, in case you do, it isn’t. sdax are partitions on the disk, whatever size it is. The user data partition is always sda4, and only its size changes to support different disk sizes.
When i try to unmount i get the reply: “umount: /DataVolume: device is busy. (In some cases useful info about the processes that use the device is found by lsof(8) or fuser(1)”
I am concerned if i force the unmount, i may not be able to remount without loss of existing data.
Well, if the worst comes to the worst…
Whoa…! mine was a simple delete human error, yours truly. I knew something was up when it took so long to delete…wrong bloody folder. I didn’t take to time to set up the recycle bin or back up restore program that comes with WD Mycloud NAS…
The implications are huge and the chance for data loss…well lets just say this will become a very popular forum.
Partition unmounting won’t cause any data loss… Simply it won’t be available to read/write by other processess until My Cloud reboot or manual partition re-mount. You should read a little bit more what mounting/unmounting is on Linux.
Thanks Mona, I have reading about the linux processes all day. Just doing a total backup before I attempt using extundelete.
Hi Mona,
Still trying to unmount the sda4. I have attached a screenshot. Do you have any other ideas as extundelete will reportedly only work when the /dev/sda4 is unmounted. Thanks for you persistence.
Hey Ged. Please type: “lsof /dev/sda4” and show the output.
1 Like
Sorry for the delay. I turn the device off when I am not using it so it won’t be running processes. When I restart, it takes forever to initialise. Now I am in via ssh.
Okay, type:
/etc/init.d/wdmcserverd stop
and then try to unmount (umount) it.
This will only stop the service, not disable, so in case of problems this service will be restarted after My Cloud restart. I personally have this service stopped by my startup script, because I don’t use remote access to MyClould, so I’m sure it’s safe to stop this service.
1 Like
Thanks. Worked. Now I will go back to your original post and run extundelete (i have it on the attached external drive to the wd cloud.
Great! Good luck with extundelete. If you remember the folder name, try to recover only this folder (I think I posted a command for this in my big post). Should be quicker and hopefully folder structure can be recovered.
1 Like
