Resolved some dmesg errors (#104)

[Ioa1]

Resolved "upper fs does not support file handles, falling back to index=off" and " fs on 'tmp' does not support file handles, falling back to xino=off"

[ezrizhu]

       index={on|off}
           Inode index. If this feature is disabled and a file with
           multiple hard links is copied up, then this will "break" the
           link. Changes will not be propagated to other names referring
           to the same inode.

       xino={on|off|auto}
           The "xino" feature composes a unique object identifier from
           the real object st_ino and an underlying fsid index. The
           "xino" feature uses the high inode number bits for fsid,
           because the underlying filesystems rarely use the high inode
           number bits. In case the underlying inode number does overflow
           into the high xino bits, overlay filesystem will fall back to
           the non xino behavior for that inode.

           For a detailed description of the effect of this option please
           refer to https://docs.kernel.org/filesystems/overlayfs.html

looking at the docs for xino, I am curious to see what would now break when it's set to off. However, from the looks of it, both seems quite useful(?) I am not sure if we should disable them.

[mgree]

We can turn index off, but we need to dig a little more for xino.

[mgree]

index is being disabled automatically for our upper, so it's safe to turn that off. xino should only be turned off when we know pre-emptively that it's going to fail... can we identify the fs types that's true for, and only disable it for those overlays?

[mgree]

Not a bad start!

1. Would be good to have a more thorough enumeration of which fs types work without a merger.
2. The fs_fail_flag logic doesn't seem right to me.

[mgree]

[[ ... ]] is a bashism. but also, if COND; then return 1; fi return 0 is really just ! COND. To save on processes, we probably want:

Suggested change

	if [[ "$fstype" = "msdos" || "$fstype" = "exfat" || "$fstype" = "hfs" || "$fstype" = "hfs+" ]]; then
	case "$fstype" in
	(msdos|exfat|hfs|hfs+) returrn 1;;
	(*) return 0;;
	esac

[mgree]

Is this an exhaustive list? Can we not do it for ext3/4?

[mgree]

Suggested change

	check_fstype() {
	mountable_without_mergerfs() {

[mgree]

Suggested change

	if check_fstype "$pure_mountpoint"; then
	if ! mountable_without_mergerfs "$pure_mountpoint"
	then
	needs_mergerfs=1
	fi

And then... we ought to break out early. If any mount requires a mergerfs, we should give up early and just use mergerfs. Right?