Documentation/filesystems/overlayfs.rst

21 In many cases, an object accessed in the union will be indistinguishable
25 While directories will report an st_dev from the overlay-filesystem,
27 upper filesystem that is providing the object.  Similarly st_ino will
30 tools ignore these values and will not be affected.
33 filesystem, all objects will report an st_dev from the overlay
34 filesystem and st_ino from the underlying filesystem.  This will
36 overlay objects will be distinguishable from the corresponding
46 will use the high inode number bits for fsid.  Even when the underlying
52 filesystem will fall back to the non xino behavior for that inode.
102 overlayfs.  The upper filesystem will normally be writable and if it
166 will each have separate caches.  A seekdir to the start of the
167 directory (offset 0) followed by a readdir will cause the cache to be
201 2. If the "redirect_dir" feature is enabled, then the directory will be
281 rename or unlink will of course be noticed and handled).
312 however, the mounting task will have sufficient privileges to perform all
337 that case the overlay will be read-only.
339 The specified lower directories will be stacked beginning from the
340 rightmost one and going left.  In the above example lower1 will be the
347 When metadata only copy up feature is enabled, overlayfs will only copy
349 like chown/chmod is performed. Full file will be copied up later when
365 "trusted." xattrs will require CAP_SYS_ADMIN. But it should be possible
369 conflict with metacopy=on, and will result in an error.
387 though it will not result in a crash or deadlock.
400 lower root origin, mount will fail with ESTALE.  An overlayfs mount with
401 "inodes index" enabled will fail with EOPNOTSUPP if the lower filesystem
407 probably will succeed but expect the unexpected later on. So don't do it.
412 the copied layers will fail the verification of the lower root file handle.
439 will fail with EXDEV ("Invalid cross-device link").
447 up, then this will "break" the link.  Changes will not be propagated to
458 enough free bits in the inode number, then overlayfs will not be able to
460 value of d_ino returned by readdir(3) will act like on a normal filesystem.
475 the behavior of the overlay is undefined, though it will not result in
488 to by the "trusted.overlay.redirect" extended attribute, will verify
492 will not be merged with the upper directory.
539 copy_up of that disconnected dentry will create an upper index entry with
556 handles, so exporting with the 'subtree_check' exportfs configuration will
564 read-write mount and will result in an error.