Answer: Switch Forwarding to Default Gateway

certskills
By certskills May 21, 2014 09:05

Here’s the answer to the question I posted late last week. Ask’em if you’re got’em! Details below to fold to avoid spoiling it.

Answers: B, C, D, E

The key to this question is that the scenario and the question tell us that the switches use all default configuration, and that the switches have not learned any information yet. That means that the switches have no dynamically-learned MAC table entries, and no statically-configured MAC table entries. Based on these facts, you can know that both SW1 and SW2 will flood the frame.

Flooding means that a switch forwards the frame out all ports in the same VLAN, except for the port in which the frame arrived. In this case, SW1 does no send the frame back out its F0/1 port, because the frame came in that port. SW1 does send it out it’s other three ports, F0/2, G0/,1 and G0/2 (answers B, C, and D in this case).

SW2 receives the frame on its G0/1 port. SW2 uses the same flooding logic, because based on the scenario and question, SW2 has no MAC table entries. So, SW2 does not forward the frame out G0/1, but does forward it out F0/9.

Router R1 does not forward Ethernet frames, because routers remove the Ethernet header and trailer as part of the routing process. Just based on wording, routers in general do not forward Ethernet frames, but do forward the packets that might arrive at the router inside Ethernet frames. So, R1 will not forward the Ethernet frame, although R1 might forward the packet inside the frame. Note that this logic may seem a bit picky, but it is intended to help you notice the use of the words frame and packet, and recall that switches forward frames, and routers forward packets.

Switch Forwarding of Frame From Host to Gateway
Question: Ping Doesn't Work in a Simple Network
certskills
By certskills May 21, 2014 09:05
Write a comment

4 Comments

  1. Jason T June 19, 16:58

    Is it wrong that I got this right because I saw the packet stopping at the R1 f0/0 interface, simply because the question said that is where it was headed?

    Reply to this comment
  2. CCENTSkills June 20, 06:16

    Hi Jason,
    Well, I’m not quite sure what you mean by “wrong that I got this right”… but if you’re talking about the picky wording on frame versus packet, probably yes. Say the host was sending a packet to some destination IP address far past the router, which is more typical of user traffic. That is, the user doesn’t typically send a packet with a destination IP address for one of the router’s IP addresses, so it’s probably destined for some other IP host past the router. The frame that holds that packet still has a destination MAC address of the MAC of the default router, which is where this particular question picks up the story.
    I probably should’ve given a little more detail on the background – sorry about that!
    Wendell

    Reply to this comment
  3. w951afh March 10, 17:40

    Hi Wendell,

    Quick question, or perhaps just an observation. In your question you stated that ” Switch SW1 receives an Ethernet frame from host A. The frame lists a destination MAC address of R1’s F0/0 MAC .” If PC A already has R1’s MAC address that would kind of imply that SW 1 has already Floodedd an ARP request from PC A to R1 and SW1 should already know R1’s MAC address as well.

    Reply to this comment
    • CCENTSkills March 14, 10:52

      Hello,
      That’s a great point. Bear with me on my wording as a make what I hope is a useful answer!

      First off, as worded, I agree, the question needs a tweak. But maybe in a slightly different way than you’d expect. I need to say that there are no dynamic entries in the table right now…

      You might infer (or believe the text implies) that soon before this event that either host A or the router had sent an ARP broadcast for the other, causing switch SW1 to learn both device’s MAC addresses. I agree that you might infer that from reading the question.

      Or…
      Maybe the ARP happened some seconds or even minutes ago. (The question didn’t state that particular detail.) Independent from that, the switch might have removed its MAC table entry. The ARP table timeouts on the host and router, as well as how a switch manages when it removes switch table entries, are all independent. So that leaves space for the idea that host A knows R1’s MAC address (no ARP timeout yet) while the switch has cleaned up that entry from its MAC address table. EG, an engineer could have flushed the mac table via a command., or the switch table could have filled with fresher entries, forcing the entries related to this question out of the MAC table.

      I definitely agree that statistically speaking, it’s more likely that the MAC table has those entries.

      Thanks for the note!
      Wendell

      Reply to this comment
View comments

Write a comment

Leave a Reply to w951afh Cancel reply

Subscribe

Subscribe to our mailing list and get interesting stuff and updates to your email inbox.

Thank you for subscribing.

Something went wrong.

Search

Categories