Allied Telesis SwitchBlade x3100 Series Manual page 438

Introduction
4.6.4.4 Master Node and Transit Node Fault Detected Correction Procedure
When the master node detects a fault in the ring using either of the above described procedures, it takes the
following actions:
Declares the EPSR domain to be in a failed state (from the complete state the EPSR domain was in before
•
the fault was detected).
Unblocks its SP port for the non-control traffic for this EPSR domain
•
Flushes its own forwarding database (FDB) for just the two ring ports
•
Sends an EPSR Ring-Down-Flush-FDB control message to all the transit nodes via both its PP port and SP
•
port.
As the EPSR domain non-control traffic starts flowing, each of the nodes (both master and transit) then re-
learn the layer 2 addresses on the flushed ring ports again to reflect the newly collapsed network topology. The
master node continues to follow the Master Node Polling Fault Detection Procedure and as long as the fault is
still present in the ring, the EPSR domain will continue to remain in the failed state. This newly constructed net-
work topology exists until the fault in the ring is corrected; then the fault restoration procedures take over and
restore the ring to its original normally operating state.
The EPSR stabilized topology under normal operating conditions is shown in
detected between Allied Telesis System A and Allied Telesis System B,
lized topology after the fault detection and corrections procedures have been executed.
4-118
Software Reference for SwitchBlade x3100 Series Switches (Layer Two Switching)
EPSR Protocol
Figure 4-9. For a link fault
Figure 4-10 shows the new EPSR stabi-