What is the reason for the nofail option in /etc/fstab?
Without it the boot process hangs if the drive is not plugged in.
Any reason you can't format the drive as ext4?
I would have to buy another drive to backup the data first.
Performance is adequate, the bottleneck is the 100Mb/s ethernet i/f.
I'm doing similar on a Raspberry Pi2 with Jessie and have had no such issues, but I have a weird one in that the system immediately crashes back to "emergency" mode if I try to run it headless (no mouse/keyboard).
I don't know what emergency mode is. The BBG has no video output, so
it's always headless.
Is the USB drive powered separately or from the USB bus?
Yes, it has its own power brick. I have also a USB printer connected
to the same hub and it doesn't seem to have this problem.
One more thing, since you are mounting by UUID why/how does the code writing to the drive "know" its sda1? As long as its mounted at /mnt/usbdrive it shouldn't matter if its sda1 or sdb1. you could try mounting /dev/sda1 instead of UUID=. Have you verified that the usb/drive is really mounted at /mnt/usbdrive after it "switches" to /dev/sdb1?
It is because the uuid is nothing but a symlink:
root@beaglebone:/dev/disk/by-uuid# ls -l
lrwxrwxrwx 1 root root 10 Jun 8 10:35 F250C93950C904F7 -> ../../sda1
lrwxrwxrwx 1 root root 15 Jun 8 08:54
c1d24650-2c0c-4272-8762-56158468bbfb -> ../../mmcblk0p1
The only code using the device file is mount.ntfs (I think that is a
hard link to ntfs-3g). mount.ntfs runs as a background process,
keeping the device file locked. As far as the file system is
concerned, the drive stays mounted on /dev/usbdrive, but any attempt
to access it gives an I/O error.
I have been banging my head against a wall, trying to write a udev
rule. Then I realised that udev is not the problem: the /dev/sdx block
dev files are managed by the kernel. This means that even if I
uninstall udev, ntfs-3g and try to mount it as ntfs the kernel will
still change the dev files.
I have since found this topic:
I have connected the drive to my Windows laptop and used Seagate's s/w
to disable sleep mode. It can't be that, either.
My unix experience dates back to the days of floppy disks and parallel
ports. This has me stumped.
Thanks for trying to help, Wally.