I agree with Hal that this problem is quite complicated. My suggestion was
just meant to deal with a very easy case.

On Thursday, March 31, 2016, Hal Murray <[email protected]> wrote:

>
> [email protected] <javascript:;> said:
> > I suggest that NTS should encourage users to write the current time to a
> > persistent file that is available upon reboot.  This should be
> overwritten
> > on every clock update.    Then, when the client reboots, it can check
> that
> > the expiry time of the certificates is no earlier than the last time time
> > written to the presisent file.  This limits the impact of attackers that
> use
> > old compromised certificates to break the security of NTS.
>
> I think the issue of getting started is much more complicated than that.
>
> It probably deserves a separate document to collect all the ideas.
> Individual documents like NTS should be explicit about what they are
> assuming.  Do they need valid (how close?) time or valid certificates or
> ???
>
> The above paragraph assumes the system has a writable file system.  The
> "every clock update" may be too expensive for some systems.
>
> What happens the first time?  Even if you assume that file was sanely
> initialized at the factory, the unit may have sat on a shelf for a long
> time.
>
> What do you do if your clock gets set far into the future?
>
>
> --
> These are my opinions.  I hate spam.
>
>
>
> _______________________________________________
> ntpwg mailing list
> [email protected] <javascript:;>
> http://lists.ntp.org/listinfo/ntpwg
>
>

-- 
Sharon Goldberg
Computer Science, Boston University
http://www.cs.bu.edu/~goldbe
_______________________________________________
TICTOC mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/tictoc

Reply via email to