0

I have a kubernetes cluster of some virtual machines running on top of Vsphere. I deployed Calico IP-IP plugin for connectivity and every thing is Ok. My questions are : 1) As calico performs BGP routed between the nodes , Does that also happen in the IPIP case or the flow is only Full mesh IPIP tunnels. 2) If there is BGP routes --- how can i check them ? and how were they advertised through the hypervisor ?
3) If not... how does the packet flow go between pods in my case ?

I tried to check the calicoctl results but no peers. Some of Ip route results : 192.168.0.0/24 via 172.16.5.226 dev tunl0 proto bird onlink 192.168.2.0/24 via 172.16.5.228 dev tunl0 proto bird onlink

Sam
  • 21
  • 1
  • 2

1 Answers1

0

1) If by IPIP you mean node-to-node peering, then yes, it's configurable

calicoctl config set nodeToNodeMesh off

For large-scale deployments, or for deployments where you require a more specific BGP topology (e.g., peering with ToR switches) the full node-to-node mesh should be disabled and explicit BGP peers configured for your Calico nodes. A BGP peer may be configured in your Calico network as a global BGP peer or a per-node BGP peer

2) List a routing table route -n The direct routes are set up by a Calico agent named Felix when it is asked to provision connectivity for a particular workload. A BGP client (such as BIRD) then notices those and distributes them – perhaps via a route reflector – to BGP clients running on other hosts

3) Calico packets flow

A_Suh
  • 3,727
  • 6
  • 22
  • Sorry. I mean in my case where the virtual machines are communicating through hypervisor, and IPIP is enabled . Is there still BGP peering between the Virtual machines or not ? – Sam May 22 '19 at 11:16
  • @Sam yes, but it's consistent with Calico only when host's route next hop IP represents the local hypervisor. – A_Suh May 22 '19 at 11:40