1. 23

  2. 2

    Onwards my triple boot

    1. 2

      Can I write to the filesystems? Yes! Because FAT and ext2 reserve the sectors used by ZFS’s labels, and FAT reserves all space used by ext2, and ZFS doesn’t write to the boot space, all three exist in harmony.

      Can somebody explain this to me? I don’t understand how the file system can ensure that e.g. the FAT mount during a write operation won’t write a chunk of the file to sector 12345, even though ext2 already has stored a part of one of its files on sector 12345 (to FAT sector 12345 looks unused). From what I understood all file systems think that they have the full storage of the image available as their own space (expect for the reserved metadata sectors for each FS)?

      By using a program that creates an ext2 filesystem that marks many of its blocks unusable, all three of these can be made to not overlap their metadata, allowing them to live in harmony.

      From my understanding this section is only about ensuring that ext2 will not overwrite the ZFS metadata at the 8k marker (by using mkext2 mentioned later in the README). But it does not explain how the file systems can manage the actual data space.

      1. 2

        At least on FAT I believe it is possible to mark clusters of sectors as “bad blocks” in the File Allocation Table (the FAT, in fact), so maybe they mark the ext2fs area as a series of bad blocks?

      2. 1


        I liked the original work because the two filesystems could be quite large, allowing you to make one invisible to the casual observer. Mount will find one first, I suspect across Windows, BSDs, Linux, and macOS. This limits the sizes, and a 3 mb mount on a 16gb flash drive looks just wrong.

        But it’s cool that we can do this. No btrfs love?