"Copyright notice: Original translation articles, translation level is limited, errors are unavoidable, The Translator is not responsible for the consequences of errors or omissions in the article, please be careful to reprint them. Reproduced please retain this statement and source: blog.csdn.net/shallnet, download the English version of the book "
since TCN BPDUs are so simple, how can he play such an important role? Before you answer this question directly, consider the subtle side effects of a topological change. The following discussion refers to the scenario depicted in figure 6-17. Figure 6-17. TCN BPDUs is Required to Update Bridge Tables more Quickly
Imagine that Host-d is playing Doom game with Host-e, as discussed earlier in Figure 6-12, then data traffic is going from host-d directly through Cat-b to HOST-E (first step). Assuming that the Ethernet transceiver on the CAT-B:PORT-1/2 fails, as previously stated, CAT-C:PORT-1/2 takes over the role of the specified port within 50 seconds. However, without TCN BPDUs, the game will continue to break for another 250 seconds (4 minutes and 10 seconds). Why would that be? Before the failure, the entries for MAC address Ee-ee-ee-ee-ee-ee in the Bridge Address table for all three switches are as follows in table 6-7:Table 6-7. Bridge Table Values before topology change
bridge Table |
port Associated with Ee-ee-ee-ee-ee-ee |
cat-a |
port 1/1 |
cat-b |
port |
cat-c |
port 1/1 |
That is , before the failure occurs, all frames to HOST-E must be transmitted in the inverse clock direction due to cat-c:port-1/2 blocking. When the CAT-B:PORT-1/2 fails, the CAT-C:PORT-1/2 becomes the specified port, which allows data traffic to reach Host-e in a clockwise direction.
However, the Bridge Address table for the three switches still points to the wrong direction, in fact the bridge Address table still needs to be updated if the HOST-E position in the network changes. One method is to wait for the bridge Address table to naturally age, but since the Address table default aging time is 300 seconds, this is the unfortunate result of 5 minutes of interruption before calculation. TCN BPDUs are a fair and simple way to improve the convergence time (allowing us to continue playing Doom) . The TCN bpdus and configuration bpdus are tightly matched as follows:
- The bridge begins sending TCN BPDUs in two cases:
- The port status becomes forwarding state and has at least one specified port.
- The port transitions from the forwarding state or learning state to the blocking state.
these are scenarios in which the active topology changes to notify the root bridge. Assuming that the current bridge is not a root bridge, the bridge will start to advertise processing by sending TCN BPDUs from the root port, and will continue to send TCN BPDUs per hello time until the TCN message is confirmed (note: The Hello is a locally configured Hello, Instead of the root bridge is distributed from the configuration BPDUs to the Hello time). 2. The upstream bridge receives TCN BPDUs. While some bridges receive TCN BPDUs (those that are directly connected to the root port link), only the port is specified to accept and process TCN BPDUs. 3. The upstream bridge sets the topology change acknowledgement flag in the next configuration BPDU that is sent downstream (emitted from the specified port). Acknowledgement of receiving TCN BPDUs in the previous step causes the originating bridge to terminate sending TCN BPDUs. 4. The upstream bridge propagates TCN BPDUs from the root port (TCN BPDUs are further away from the root bridge). 5. Repeat steps 2nd through 4th until the root bridge receives TCN BPDUs. 6. The root bridge sets the topology change acknowledgement flag in the next outgoing configuration BPDU (in order to confirm the network bridge that sent TCN BPDUs previously) and the topology change flag. 7. The root bridge continues to set the topology change flag in all sent configuration BPDUs (forward latency + Max age seconds (default 35 seconds)). This flag commands all bridges to shorten their address table aging process, shortening it from the default of 300 seconds to the current forward delay value (default = 15 seconds).
Figure 6-18 summarizes the above 7-step TCN process that corresponds to BITS usage (the sequence number of the steps is circular):
Figure 6-18. Sequence of flows in topology change Processes
applying these steps to the topology diagram in Figure 6-17 (these steps are not shown in the diagram for the sake of simplicity), Cat-b and cat-c send TCN BPDUs from their PORT-1/1 (the first step), because the upstream bridge is the root bridge, Steps 2nd and 5th occur simultaneously (allow skipping 3rd and 4th steps). In the next configuration Bpdu sent, the root bridge sets the TCN ACK flag to confirm receipt of the TCN from both downstream catalyst switches. Cat-a also sets the topology change flag for 35 seconds (assuming that the default forwarding delay and Max age are used) to make the Bridge Address Table update faster (6th, 7 steps). All three switches accept the topology change flag and age their address table within 15 seconds. Note that shortening the aging time to 15 seconds does not refresh all address tables, only accelerating the aging process. The device will not leave the Address table for a 15-second aging period and continue to "keep calling". However, if host-d tries to send a frame to host-e within 20 seconds (assuming HOST-E has not already sent the packet), the frame floods all the switch links, because the Ee-ee-ee-ee-ee-ee MAC address is no longer in the Address table. table 6-8 shows that three switches converge in the new topology and the data traffic is also restored after the entry of address Ee-ee-ee-ee-ee-ee in the MAC address table.
Table 6-8. Bridge Table Value after topology change
Bridge Table |
Port associated with Ee-ee-ee-ee-ee-ee |
Cat-A |
Port 1/2 |
Cat-B |
Port 1/1 |
Cat-C |
Port 1/2 |
at this time, host-d and host-e re-establish the connection, Doom Deathmaatch game can be restored. Note that the TCN BPDUs take the fault time from 5 minutes to 50 seconds.
As described in the previous section, "Configuring BPDUs," the two flag domains are stored in the same eight-bit bytes in the configuration BPDUs. This eight-byte arrangement is shown in Layout 6-19.
Figure 6-19. Layout of Configuration BPDUs Flag fields
as described in the previous section, the TCA flag is set by the upstream bridge to tell the downstream bridge to stop sending TCN BPDUs. The TC flag is set by the Root bridge to shorten the Address table aging period from 300 seconds to forward delay seconds.
Translation of the classic Cisco Lan switching, chapter sixth (10): Topology change Process
