Re: strange vnd disk behavior in -current (with ffs2?)

[Markus W Kilbinger <mk%kilbi.de@localhost>]

>>>>> "Michael" == Michael van Elst <mlelstv%serpens.de@localhost> writes:

    >> Ok, if it's really a limitation of vnd(4) then how to ensure
    >> that it does not run into this sparse file problem?

    Michael> Usually you know that you created a sparse file :)

Is there an easy way to implement into vnconfig to check the file for
sparse content, so you will be warned _before_ running into this
trouble?

Is this vnd(4) limitation documented anywhere (beside source)?
... only found pr #29838 which still seems to be open.

    Michael> vnd is supposed to return I/O errors when accessing
    Michael> unallocated blocks like a real disk device that cannot
    Michael> access a particular block.

    Michael> AFAIK vnd itself doesn't panic nor should the kernel if
    Michael> you access the vnd device. Things are different for a
    Michael> mounted filesystem, some do respond to I/O errors in the
    Michael> underlying device with a panic.

Due to the panic and the missing error messages now in -current I
guess I should send-pr about aggravation of this know
problem/limitation...

Markus.