Configuration Guide Vol. 1


29.3.3 Operation when recovery is detected

The following figure shows operation when recovery from a ring fault is detected for a single ring.

Figure 29-11: Operation during error recovery

[Figure Data]

<Structure of this section>

(1) Master node operation

When a ring fault has been detected, and a health check frame sent by the current node is received, recovery from the ring fault is determined, and the following restoration operations are performed:

The VLAN status of the ring for data transfer is changed.

The ring VLAN status for the secondary port is changed from Forwarding to Blocking. The ring VLAN status when a recovery is detected is changed as shown in the following table.

Table 29-3: Ring VLAN status for data transfer when recovery is detected

Ring port

Before (normal)

After (fault)

Primary port

Forwarding

Forwarding

Secondary port

Forwarding

Blocking

Flush control frames are sent.

Flush control frames are sent from the primary port and secondary port of the master node. Note that during recovery from the ring fault, any flush control frames transferred by each transit node return to the master node, and are discarded.

The MAC address table is cleared.

Clears MAC address table entry.

Clearing the MAC address table entries allows paths to be switched quickly.

The monitoring status is changed.

When recovery from the ring fault is detected, the master node changes from the recovery monitoring status to the fault monitoring status.

(2) Transit node operation

The following operations are performed on a transit node that receives a flush control frame sent from a master node:

Flush control frames are transferred.

Any received flush control frames are transferred to the next node.

The MAC address table is cleared.

Clears MAC address table entry.

Clearing the MAC address table entries allows paths to be switched quickly.

To prevent loops on transit nodes for which a link fault has occurred and for which recovery has then been performed, the ring VLAN status of the ring port is changed to Blocking. This Blocking status is cleared when a flush control frame sent by the master node is received, or a timeout occurs on the transit node for the reception hold time for flush control frames (forwarding-shift-time) of the ring port. The reception hold time for flush control frames (forwarding-shift-time) is set when recovery from a link fault on a ring port has occurred.