Hey, I am currently working on implementing NVGRE. I had certain questions in my mind which I think could be cleared through discussion. My understanding of NVGRE is that on the sender side when we add the NVGRE header we set a particular VSID for the packet. This VSID is assigned on the basis of the policy which contains a lookup table maintained by the administrator. A lookup is performed and the VSID corresponding to the particular MAC address and IP address is added in the NVGRE header and the packet is sent on the tunnel. On the receiver side the VSID is extracted from the NVGRE header and again a lookup is performed on the table maintained at the receiver side. If the VSID of the receiving packet is found from the table and the corresponding MAC and IP addresses are matched then the packet is further processes otherwise discarded. Please could anyone point out if I am wrong at some point or if I am missing something. I wanted to know that what is the basis of this assignment of VSID, are they just random 24 bit numbers set my the administrator or is there some standard to follow while allotting VSIDs.
Attached is a picture to clarify the scenario.
<<attachment: nvgre tunnel.png>>
_______________________________________________ dev mailing list [email protected] http://openvswitch.org/mailman/listinfo/dev
