I’ve got her machine set up as Windows native, installed on SSD. Her old Windows drive, with her old files, is hooked up to it. She can see it fine from Windows. But while she was using Linux, she’d been writing to an LVM drive, which was originally a backup, but now has some changes on it from the old drive.
I’m running Fedora as a virtual machine, and it seems to be running fine (so far). I’ve attached the old Windows drive to the physical machine, and it shows up in Nautilus and other file managers. But when I try to access it, I get a message that it’s an NTFS drive with problems, and I can only mount it read-only. It suggests I repair.
So, is this caused by the fact that its already mounted and in use by Windows? Seems a little strange, since the virtual machine probably wouldn’t know that. Windows doesn’t have a problem with it. Bigger question: Can/should I try to repair it as an unmounted drive from the virtual machine using e2fsck? That is, does e2fsck repair NTFS drives? And what is the risk if I don’t attempt to back it up first?
I think by default Linux mounts NTFS drives read only.
This link gives directions on how to mount it read write. http://linuxconfig.org/how-to-mount-partition-with-ntfs-file-system-and-read-write-access
Don’t use filesystem repair tools without backing up first.
If it’s already mounted by Windows, I presume Windows will have marked it ‘dirty’, and Linux will see it as requiring repair. Windows will probably only mark it ‘clean’ when it cleanly unmounts the disk.
Letting two operating systems mount the same drive at the same time is a bad idea. You should be able to configure the VM to see the disk as a network drive; that’s what I do with the Windows VM to read the native Linux disks at work.
I think that’s the problem. What I probably need to do is unmount the disk from Windows before I try to access it.
I was thinking similar. If a partition is to be mounted by both the host system and a VM, the VM should not be accessing it as a raw partition but as a shared drive. Two systems trying to operate on a raw partition at once is asking for collisions and corruption.
That’s probably right. I had this fantasy that both the host OS and the virtual OS could simultaneously access the same drive, to make it easier to talk, but that’s probably a forlorn hope.
Definitely possible, but you’ll need another technique. Some VM software has special device/filesystem drivers for “pass-through filesystems”, or as suggested above, use samba to mount it as a networked filesystem.
Perhaps it’s actually in hibernation state? : http://askubuntu.com/questions/145902/unable-to-mount-windows-ntfs-filesystem-due-to-hibernation
This is especially common in recent versions of Widows due to the ‘Fast Start’ thing.
No, that’s not the problem. I think the issue is that it’s mounted in Windows. I’d have to take it off line in Windows, then I could probably mount it in Linux. I may try later today.
Do not ever mount the same partition from a base OS and simultaneously with a VM. The disk will be corrupted.
If you need to communicate, then use a samba mount so that the partition is native to one machine and a remote mount to the other.
Well, no worries. It won’t let me.