Table of Contents
Advertisement
So what happened to the expected doubling of the aggregate throughput?
As it turns out, even with bridging activated, only one ATM port is allowed
to send and receive data. The second port is blocked, as can be seen in
the following screen shot:
grf16:/root brstat
Bridge Group bg1
Spanning Tree:
Enabled
Designated Root: 32768 00:c0:80:84:8c:eb
Bridge ID:
32768 00:c0:80:96:38:68
Root Port: ga010, Root Path Cost: 10
Topology Change Detected: No
Root
Max Age: 20, Hello Time: 2, Forward Delay: 15
Bridge Max Age: 20, Hello Time: 2, Forward Delay: 15, Hold Time: 1
Interface Port ID Con State
--------- ------- --- ---------- ----- ----- ----------------------- -------
*ga010
128 1
Yes Forwarding 10
ga0180
128 2
Yes Blocking
Dump snapshot finished at Mon Jun 15 20:01:33 1998
This bridging environment is useful, nevertheless. It provides an inherent
redundancy and failover mechanism. When port 0 is unplugged, the
up-to-then blocked port 1 gets activated and takes over the traffic with a
delay of about 10 seconds. With port 1 having the same IP address, this is
fully transparent to any clients.
See the next screen shot for details:
Path Desig Desig
Cost Cost Bridge
0
32768 00:c0:80:84:8c:eb 128 1
10
0
32768 00:c0:80:84:8c:eb 128 2
Multiple RS/6000 SPs and Multiple GRFs
Desig
Port
221
Advertisement
Table of Contents
