Precision Time Protocol (ieee 1588) CRS326-24G-2S+

RouterOS General
1

Hi,
I’m trying to use hardware ptp support with a CRS326-24G-2S+ running routeros 7.12.1. On the eth3 port, the grandmaster clock is a mixer’s dante sound card. what does the uncalibrated status on the eth3 port represent? Is such a high delay normal? Does anyone have any suggestions?

Thanks in advance
f-mik2.png
f-mik1.png

2

I’m guessing Dante is using PTPv1, not PTPv2 offered by RouterOS. So, from PTPv2 perspective, it’s unmanaged.

Audiante has a complex matrix of what supports PTPv2 with Dante:
https://www.audinate.com/learning/faqs/dante-devices-ptp-clocking-support?lang=z

If you put Dante in to AES67 mode, you’d get PTPv2 — but all Dante device would need to switch to use AES-67 mode. Never used DDM (Dante domain manager) but I believe that can enable PTPv2 in the mixer’s Brooklyn2/etc card, too.

One question, does Dante Controller show any clock issues?

3

Hi,
thanks for the reply.
I’m using Dante in AES67 mode, to try to receive audio on Linux. In the test environment I currently have a mixer, a Dante Avio and a PC with the AES67 client
I’m not using DDM (although I would like to try it); in dante controller if as grandmaster clock has priority the audio mixer is all clocked correctly, if instead I assign a higher priority to the ptp clock of the switch the dante devices are clocked correctly but the mixer is not (maybe ptpv1 and ptpv2 clock must be in sync?).
I’ll do more tests in the next few days

4

Some updates after upgrading to ros 7.14.
Now everything is working correctly with both Dante and aes67; I also tried using an external GPS clock and it works perfectly.

5

Nice. FWIW, Mikrotik document Dante’s QoS for the CRS326’s HW QoS recently, suggesting there is full support coming in 7.15:
https://help.mikrotik.com/docs/pages/viewpage.action?pageId=189497483#QualityofService(QoS)-Dante

6

That info was removed from the doc under the comment “Formatting” in the most recent update. You have to roll back to the prior version or diff them to recover the info.

7

Perhaps the docs came before the code on the HW QoS side, dunno. I take it as good news…kinda means they been looking at/testing PTP and AV protocols.