Switching loops

Sky747

Hi everyone!

I'm studying for the Network+ exam and I'm having trouble with switching loops.

I'm going to use the following YouTube video as an example to illustration my question - http://youtu.be/P04gaoq53FU

I understand the basic concept of loops having watched a few videos on the subject. Looking at the video above, shouldn't the loop be broken once "Computer B" replies with its MAC address? The MAC address would then be stored in the connected switch, which should stop further broadcasts from looping, no?

Obviously, I"m am missing a concept here and I appreciate anyone help!

IIIMaster

A switching loop is a behavior that occurs when the switch keep forwarding packets because it gets the packet from more than one port. If stp was enable it would only receive packets from one port and move it out in one direction so the packet does not come its way again preventing a loop. The computers has nothing to do with it,as it is similar to a child watching grown ups fight about nonsense.

I have included some documentation in regards to this matter.

Understanding and Configuring Spanning Tree Protocol (STP) on Catalyst Switches - Cisco

mistabrumley89

I think what he means is.... If computer A sends out an ARP request to the switch and that switch broadcasts a request out to the other two switches, which eventually gets out to Computer B. Shouldn't Computer B's reply be stored in the ARP table, which would end the loop because new broadcasts would no longer have to be sent out since the IP/MAC information should now be stored in the ARP table eliminating the need for another broadcast.

Sky747

Thank you for the link, IIIMaster! It will come in handy when I focus on STP.

mistabrumley89, that is exactly what I was trying to convey and still wonder about.

EdTheLad

ARP uses the broadcast mac address, this means the switches will flood this broadcast out all ports except the receiving port. As per the video PC B will receive the arp request and send an arp reply. At the same time the original arp request is still being flooding around and around by the switches due to its broadcast l2 destination address. The arp reply may not make it back to PC-A as the switches are too busy processing the arp request which will consume all the bandwidth on the core links.
If only there was a ttl on a layer two frame!