I have set up two vMX routers with dual routing-engines as well as multiple FPCs, activated the trial license, configured lite-mode for all FPCs, and set network-services to enhanced-ip.
I have come across a very strange issue whilst setting up a LAG - when a LAG has member interfaces spread across multiple FPCs, I am unable to ping across to the other router. I have run a capture on all member interfaces between both vMX routers, and I can see icmp-request making it to the other end, but the other end doesn't respond.
As soon as I disable all member interfaces on other FPCs, so that only 1 FPC is active for a LAG, icmp starts flowing back and forth with no issues.
Has anyone experienced this behaviour before, or perhaps is it well known issue?
I may not be able to resolve your issue. But provide with a few more details so that we can further look into it
1. What kind of hypervisor is used? VMWARE or OPENSTACK or KVM?
2. What kind of NIC / driver is used ?
3. Is it running on Virtio or SR-IOV?
Thanks for your reply.
To update this thread, I have now found a solution to this issue.
It seems that when multiple linecards get connected with dual REs, 3rd interface becomes internal interface just like the 2nd one which connects FPC to RE.
After bridging all 3rd interfaces on all FPCs together, ping across multiple FPCs is now working.
Usually, the 3rd interface would be a ge-0/0/0, however, that now shifts to the 4th available interface.
Hope that makes sense!