[ntp:hackers] Protocol specification modification for MS-SNTP

Danny Mayer mayer at ntp.org
Tue Jul 14 11:38:09 UTC 2009

tglassey wrote:
> Danny Mayer wrote:
>> Andrew Bartlett wrote:
>>> On Sun, 2009-07-12 at 22:47 -0400, Danny Mayer wrote:
>>>> Harlan Stenn wrote:
>>>>> Danny wrote:
>>>>>> The code should not be using Unix domain sockets. It needs to use
>>>>>> either
>>>>>> AF_INET or AF_INET6.
>>>>> Why?  What's the problem you are trying to solve?
>>>>> H
>>>> That's a point solution. The server can be anywhere.
>>> We (the Samba Team) have no need for a broader solution.  We are very
>>> happy with the solution as proposed and implemented.  Our users are also
>>> very happy with the solution.      
> Actually NO they are not. The issue is that they are for the most part
> commercial users, and they now need good-quality digital evidence of
> samba shared file services because of the legal implications. Who are
> happy are the people who dont realize how important this evidence issue
> is, or too stupid to see what the impact of making the evidence models
> around the use of something unprovable.

Do you have evidence that most Samba users are commercial users? Do you
know what files being served require legal evidence? There has yet to be
a single legal case where the timestamp of a file is part of the
evidence and any assertion of such a timestamp is subject to legal
challenge and will most likely fail because it is so easy to fake a
timestamp not withstanding ntpd running.

>> Except if I want to run NTP on a Microsoft Domain Controller which is
>> what I do at home. 
> OK is this W32TIME or a fuller NTP implementation you want to run. It
> makes a difference because the use of SNTP on open Internet connections
> is iffy at best.

Why would I want to run w32time?


This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

More information about the hackers mailing list