Dan Peretz <dan.peretz...@gmail.com> wrote:

> Thank you for responding, Theo :)
> 
> On Thu, Aug 13, 2020 at (...):59 AM Theo de Raadt <dera...@openbsd.org> wrote:
> >
> > the FAQ is wrong.
> >
> > Those images don't contain signatures because my build & sign
> > procedure does not have a way to sign something, then continue
> > building, then sign the result.
> >
> >
> > If you looked, you would see there is an unsigned SHA256 file.
> >
> 
> Yes, I opened the install ISO and I see. I also read the INSTALL.amd64
> doc. Maybe change the FAQ to something like this?
> "[...] If someone were to make a rogue installation image, they could
> certainly change the installer to say the files were legitimate.
> Regardless, an unsigned SHA256 file is used by the installation to
> detect accidental corruption in the file sets. If the distribution
> sets do not match their recorded checksums, the installation program
> will complain."

I disagree with your direction.  

You are now explaining to someone how to create invalid images,
rather than 


> > It already uses the SHA256 file to determine which files to install,
> > this is done, but a hash is not a cryptographic signature, so the warning
> > issued is accurate.
> >
> 
> Maybe also rephrase the warning by the installer to something like
> this, to make it clearer to the admin that the installer does not skip
> verification *completely*:
> "Directory does not contain SHA256.sig. Continue without verifying
> authenticity? (The sets will still be verified against accidental
> corruption with SHA256.) [no]"

I disagree.  I think it is good enough the way it is.  I don't think
People need to know all the details behind a warning.

> > Huh.  What you are asking for cannot be done.  And obviously a bogus
> > image would declare that it isn't bogus.
> >
> 
> True, but I meant that if the ISO boots a BSD.RD file, then the
> ramdisk (booted from that exact file) would verify the checksum of
> that file, from the disc. Obviously not from RAM. It's not foolproof,
> and it surely doesn't help against malicious alterations, but I think
> this is better than not doing so...

What you describe cannot be done.  You aren't thinking critically.

Reply via email to