ciopfs is one, obviously you’ll need to build it with libfuse armhf. The other only way I could think of is VFAT mounts (loop won’t work on our kernel) but that’s not a bright idea as you’ll need to mess with the partitions more tricky with the exisiting md raid.
I’m not sure what is your backup strategy. Are you trying to ignore certain paths/files? If yes, use find regex to build your dynamic exclude vars and pass it to rsync (rsync = no regex). If most of them are static, it’s even easier to use exclude txt file.
I simply doing a rsync between two WD cloud drives. I’m pretty much syncing all shares, not excluding anything.
I guess I’ll have to live with this…
The reason why I brought up the question was that brcause if on the source side even if I rename a folder from “MOVIES” to “Movies”. That in return results in rsync syncing everything under that folder. Which could be like 100 Gb of data…
Nite that I use rsync without checksum calculation and utilize file size and time stamp comparison for syncing files.
if I rename a folder from “MOVIES” to “Movies”. That in return results in rsync syncing everything under that folder. Which could be like 100 Gb of data…
I guess you have three choices:
i) don’t rename folders after you’ve backed them up…
ii) if you do rename folders, accept that it will take time to resolve the change
iii) access your backup and change the folder name to match. Depending on the nature of your backup, of course; rsync is a relatively simple mirroring tool (in concept, if not in how it does it), so it should be possible to change the name, unlike a more sophisticated version-holding backup tool.