10-06-2020 04:22 AM
I'm have a question about CEF's default load-balancing decisions (or maybe it's more of a question about Cisco's traceroute). From what I've read, when two equal-cost links to a destination are available, CEF's default method is per-destination. I also read this from Cisco's documentation: "Packets for a given source-destination host pair are guaranteed to take the same path, even if multiple paths are available".
If that's the case, when I run a traceroute between a given source and destination IP address, shouldn't all the traceroute packets go out the same forwarding interface, since each packet has the same source and destination IP address? However, when I set up this situation in a lab with two equal-cost links, I find the traceroute alternates between the outgoing interfaces (10.1.2.2 and 10.10.10.1 are the two equal-cost next hops):
R1#traceroute 2.2.2.2
1 10.1.2.2 105 msec
10.10.10.1 2 msec *
Can someone explain this behavior for me?
Solved! Go to Solution.
10-06-2020 06:35 AM
Hello @Venison Mogambi ,
traceroute probe packets are not processed by CEF but they are processed in software this is a less efficient way but allows to discover possible alternative paths for each given TTL .
Generally speaking CEF applies to packets that are not sourced or destined to the router itself but to user traffic that just need to be routed via a pair of router interfaces.
With default settings of three probes for each TTL value traceroute can discover up to three alternate next-hops on the path to destination.
Hope to help
Giuseppe
10-06-2020 06:35 AM
Hello @Venison Mogambi ,
traceroute probe packets are not processed by CEF but they are processed in software this is a less efficient way but allows to discover possible alternative paths for each given TTL .
Generally speaking CEF applies to packets that are not sourced or destined to the router itself but to user traffic that just need to be routed via a pair of router interfaces.
With default settings of three probes for each TTL value traceroute can discover up to three alternate next-hops on the path to destination.
Hope to help
Giuseppe
10-07-2020 12:35 AM
Thanks for the response! I did read something about the process switching vs CEF when I was researching this, but I just wanted to confirm.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide