We have two Nexus 7706s as our core. Standard-sized pings over the VPC peer link between the two are experiencing a small amount of packet loss - less than 1%. The peer link is a 2x10Gb port-channel. I tried isolating each physical link to see if the packet loss still existed, but it was still there. The isolation I performed was to shut down the physical interface on switch #1, then run the pings over the peer link. I did the same for the other physical interface in the port channel. In both cases, I still lost 3-4 packets out of 1000.
I was pinging an SVI on the opposite 7706 switch and verified via the MAC address table that it saw that address via the peer link.
7706 Switch 1 Eth1/33 -----> 7706 Switch 2 Eth1/33
7706 Switch 1 Eth2/33 -----> 7706 Switch 2 Eth2/33
The physical links are 10Gb running through our in-house fiber runs.
We have a TAC case open for this, but since I wasn't able to isolate the physical links as the issue, what could be the cause? I'd be happy to provide more details, if necessary.