Re: [trill] TRILL and IEEE 1588 PTP

Right, 1588 uses both multicast and unicast messages.
There are a couple of other issues to consider, to guarantee that time is distributed accurately:
- 1588 messages between every {master,slave} pair should be forwarded symmetrically, i.e., the forward direction and the reverse direction should use the same physical path.
- ECMP should not be used on 1588 messages.

I believe quite a few of the principles of draft-ietf-tictoc-1588overmpls are applicable to TRILL as well.

Cheers,
Tal.

-----Original Message-----
From: trill [mailto:trill-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Monday, January 26, 2015 7:47 AM
To: Jon Hudson; Donald Eastlake
Cc: Tal Mizrahi; Nandakumar Natarajan; trill@ietf.org
Subject: Re: [trill] TRILL and IEEE 1588 PTP

Yes, 1588 has more clocks syncrionization accuracy than NTP. I also think 1588 over TRILL seems to be very similar to BFD, because PTP message is a multicast message, it can be conveyed using multicast RBridge channel.
Thanks,
weiguo
________________________________________
From: Jon Hudson [jon.hudson@gmail.com]
Sent: Sunday, January 25, 2015 5:56
To: Donald Eastlake
Cc: Nandakumar Natarajan; Tal Mizrahi; trill@ietf.org
Subject: Re: [trill] TRILL and IEEE 1588 PTP

Speaking personally I would LOVE to see this.

Recent NTPd issues in the Wild have shown that 1588 might be a great tool to have.

Cheers
Jon

> On Jan 24, 2015, at 8:30 AM, Donald Eastlake <d3e3e3@gmail.com> wrote:
>
> If there is interest in IEEE 1588 over TRILL, I would think it would 
> be quite straight forward. Just assign an RBridge Channel protocol 
> number to it as was done for BFD.
>
> Thanks,
> Donald
> =============================
> Donald E. Eastlake 3rd   +1-508-333-2270 (cell)
> 155 Beaver Street, Milford, MA 01757 USA d3e3e3@gmail.com
>
>
>> On Sat, Jan 24, 2015 at 9:09 AM, Tal Mizrahi <talmi@marvell.com> wrote:
>> Hi Nanda,
>>
>>
>>
>> The IEEE 1588 standard does not include a mapping for TRILL, nor is there
>> currently a plan to add a TRILL mapping in the next revision of IEEE 1588.
>>
>> Personally, I am not aware of any work in the IETF on 1588 over TRILL. On a
>> (somewhat) related note, there was  some discussion about 1588 over MPLS
>> (see draft-ietf-tictoc-1588overmpls, draft-mirsky-mpls-residence-time).
>>
>>
>>
>> Cheers,
>>
>> Tal.
>>
>>
>>
>> From: trill [mailto:trill-bounces@ietf.org] On Behalf Of Nandakumar
>> Natarajan
>> Sent: Friday, January 23, 2015 2:50 AM
>> To: trill@ietf.org
>> Subject: [trill] TRILL and IEEE 1588 PTP
>>
>>
>>
>>
>>
>> Hi,
>>
>>
>>
>> Is there any existing documentation on supporting IEEE 1588 PTP protocol in
>> a TRILL network? Thanks for any pointers.
>>
>>
>>
>> Thanks,
>>
>> Nanda
>>
>>
>> _______________________________________________
>> trill mailing list
>> trill@ietf.org
>> https://www.ietf.org/mailman/listinfo/trill
>
> _______________________________________________
> trill mailing list
> trill@ietf.org
> https://www.ietf.org/mailman/listinfo/trill
_______________________________________________
trill mailing list
trill@ietf.org
https://www.ietf.org/mailman/listinfo/trill