[ntp:hackers] [ntpwg] Pending NTP WG Last Call on Autokey

David L. Mills mills at udel.edu
Fri May 16 17:10:50 UTC 2008


Danny,


Your proposal would break the parsing rules that determine if an 
extension field or MAC is present. It also wastes precious header space.

Dave

Danny Mayer wrote:

> This is the way it should look according to my new scheme:
>
>
>  0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
>
> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>
> |   Extension Identifier        |1|          Length             |
>
> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>
> |R|E|   Code    |Reserved | VN  |   Reserved - MUST be 0        |
>
> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>
>
> For old autokey configurations it would look like this:
>
>
>  0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
>
> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>
> |R|E|   Code    |Reserved | VN  |0|          Length             |
>
> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>
>
> Notice that bit 16 would be 0 anyway since the length is not currently 
> that large.
>
>
> Now we have privitized the autokey private information but still allow 
> it to interoperate with the updated protocol.
>
>
> Danny
>
> David L. Mills wrote:
>
>> Bian,
>>
>>
>> Oops, you found errors, both in the diagram and text describing the 
>> Code field. The first word should be
>>
>>
>>  0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
>>
>> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>>
>> |R|E|   Code    |Reserved | VN  |            Length             |
>>
>> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>>
>>
>> the text should be
>>
>>
>> <t>The 6-bit Code field specifies the request or response operation, 
>> while the 3-bit Version Number (VN) field is 2 for the current 
>> protocol version. There are two flag bits: bit 0 is the Response Flag 
>> (R) and bit 1 is the Error Flag (E); the Reserved field is unused and 
>> should be set to 0. The remaining fields will be described later.</t>
>>
>>
>> To reprise, the intent of my proposal is to use the Reserved field as 
>> a type or class code, with autokey assigned class 0. If other than 0, 
>> the bits other than the Reserved bits are arbitrary, but it is my 
>> hope that they be interpreted in the same fashion in the interest of 
>> code reuse.
>>
>>
>> Dave
>>
>>
>> Brian Utterback wrote:
>>
>>
>>> I hate to enter the fray at this late date, but I only just got a 
>>> chance to look at the autokey draft, and I am confused. During the 
>>> course of this conversation, there was mention of taking a bit from 
>>> the code field or the length field. In the draft it mentions that 
>>> the code field is 8-bits and specifies the operation. But I cannot 
>>> find either the code or the version field in the format diagram. 
>>> Where is the code and version fields located?
>>>
>>>
>>> Brian Utterback
>>>
>>
>> _______________________________________________
>>
>> hackers mailing list
>>
>> hackers at lists.ntp.org
>>
>> https://lists.ntp.org/mailman/listinfo/hackers
>>
>>
>



More information about the hackers mailing list