Hello gang,
I'm looking for a clear explanation of PTP commands on NXOS platform.
TL/DR:
which IP I should use for ptp source
1) Grand Master
2) Boundary Clock's own IP addr (loopback/etc.)
The topology is like this
GM(GPS)(192.168.0.1) --- Arista Low-Latency SW (BC) --- Nexus01 (BC) ---Trunk --- Nexus02 (BC)
(BC = Boundary Clock)
Here's the configuration I used for Nexus switches.
Nexus01 feature ptp ptp source 192.168.0.1(GM IP addr) int eth x/x (trunk) ptp ptp vlan xxx int eth x/x (edge ports towards Ordinary clock) ptp ptp vlan xxx Nexus02 (about the same)
What I'm confused about is "ptp source" command
According to Cisco and Arista whitepaper, source IP should be BC's own IP address and it is used for restamp PTP messages, and it's used only in Hybrid profile(unicast ptp).
But my boss says I have to use the IP address of the Grand Master clock.
Now, I checked the PTP counters on both Nexus switches and it seems to be working fine, and I presume that the current network is configured that way(source ip set to GM on every BC device) as well.
We only use multicast ptp only, so I figured ptp source command has no effect on our environment.
Which one is correct for ptp source ip? GM's IP or BC's own IP addr?
No comments:
Post a Comment