|
Broadcasts and Layer 2 loops
can be a dangerous combination, as illustrated in Figure .
Assume that neither switch is running STP. Host A begins by sending out
a frame to the broadcast Media Access Control (MAC) address
(FF-FF-FF-FF-FF-FF) in Step 1. Because Ethernet is a bus medium, this
frame travels to both Cat-1 and Cat-2 (Step 2).
When the frame arrives at
Cat-1:Port-1/1, Cat-1 will follow the standard bridging algorithm and
flood the frame out all other ports, including Port 1/2 (Step 3). The
frame coming out Port 1/2 will travel to all nodes on the lower Ethernet
segment, including Cat-2:Port1/2 (Step 4). Cat-2 will flood the
broadcast frame out all other ports, including Port 1/1 (Step 5) and,
once again, the frame will show up at Cat-1:Port-1/1 (Step 6). Oblivious
to the loop, Cat-1 will send the frame out Port 1/2 for the second time
(Step 7). By now you can see the pattern; a substantial loop has
propagated.
Additionally, notice that
Figure
ignored the broadcast that arrived at Cat-2:Port-1/1 back in Step 2.
This frame would have also been flooded onto the bottom Ethernet segment
and would have created a loop in the reverse direction. In other words,
be sure to understand that this "feedback" loop would occur in
both directions.
An important conclusion can be drawn from the Figure: bridging loops are
much more dangerous than routing loops. Why? Suppose this frame was an
Ethernet Version 2 frame, as shown in Figure .
Recall that the V2 Ethernet
frame contains only two MAC addresses, a Type field, and a cyclic
redundance check (CRC) (plus the network-layer packet as data). By way
of contrast, an IP header contains a Time To Live (TTL) field that gets
set by the original host and is then decremented at each router. By
discarding packets that reach TTL = 0, routers prevent
"runaway" datagrams. Unlike IP, Ethernet (or, for that matter,
any other common data-link implementation) does not have a TTL field.
Therefore, after a frame starts to loop in the network above, it will
probably continue until someone shuts off one of the switches or breaks
a link.
As if that is not
frightening enough, networks such as the one illustrated in Figure
or even more complex networks such as the one in Figure
will actually witness the feedback loop grow at an exponential rate! As
each frame is flooded out multiple switch ports, the total number of
frames multiplies quickly. In fact, a single Address Resolution Protocol
(ARP) has filled two OC-12 ATM links for 45 minutes (each OC-12 sends
622 megabits per second [Mbps] in each direction, for a total of 2.4
gigabits per second [Gbps] of traffic)!
|