Allied Telesis AT-9900 用户手册

 
 

When the master node detects an outage somewhere in the ring, using either of 
the detection methods previously described, it will:

■

declare the ring to be in a “failed” state

■

unblock its secondary port to enable the data VLAN traffic to pass between 
its primary and secondary ports.

■

flush its own forwarding database (FDB) for (only) the two ring ports

■

send an EPSR “Ring-Down-Flush-FDB” control message to all the transit 
nodes, via both its primary and secondary ports

The transit nodes respond to the “Ring-Down-Flush-FDB” message by flushing 
their forward databases for each of their ring ports. As the data starts to flow in 
in the ring’s new configuration, each of the nodes (master and transit) re-learn 
their layer 2 addresses. During this period, the master node continues to send 
health check messages over the control VLAN. This situation continues until 
the faulty link or node is repaired. 

 shows the flow of control frames 

under fault conditions.

Figure 1-2:  EPSR Fault Detection Messages

For a multi domain ring, this process will occur separately for each domain 
within the ring.

Master 

Node

Transit 

Node

1

Transit 

Node

4

Transit 

Node

2

Master Node Hello Message

Control VLAN

Ring-Down-Flush-FDB Message

Transit 

Node

3

EPSR 2.eps

Transit Node Links Down Message

P

S

(1)

(2)

(3)