47
Because the member switches are connected with a ring connection, when a single IRF link fails, the
connection turns into a daisy chain connection and the IRF fabric is not partitioned. When both the IRF
link between Device A and Device C and the IRF link between Device B and Device D go down, however,
the IRF fabric will break into two IRF fabrics: one (IRF 1) with Device A and Device C, and one (IRF 2) with
Device B and Device D. In addition, IRF link status error and card failure messages will be displayed. The
following takes IRF 1 for example:
#May 7 09:13:42:388 2010 HP STM/4/LINK STATUS CHANGE:
#May 7 09:13:42:720 2010 HP DEVM/1/BOARD STATE CHANGES TO FAILURE:
Because the master member ID of IRF 2 is bigger, when the MAD mechanism detects errors, the state of
IRF 2 changes to recovery and all ports on the member devices except for the reserved ones are shut
down.
# If IRF 1 also fails, log in to the console port of Device B or Device D and use the
mad restore
command
to restore IRF 2 to active state and bring up the disabled ports on the member devices.
<Sysname> system-view
[Sysname] mad restore
This command will restore the device from multi-active conflict state. Continue? [Y/N]:y
Restoring from multi-active conflict state, please wait...
[Sysname]
#May 7 09:23:16:050 2010 HP IFNET/4/INTERFACE UPDOWN:
Trap 1.3.6.1.6.3.1.1.5.4<linkUp>: Interface 277872640 is Up, ifAdminStatus is 1,
ifOperStatus is 1
%May 7 09:23:16:069 2010 HP IFNET/3/LINK_UPDOWN: GigabitEthernet2/3/0/2 link status is
UP.
#May 7 09:23:16:302 2010 HP LAGG/1/AggPortRecoverActive:
Trap 1.3.6.1.4.1.2011.5.25.25.2.4<hwAggPortActiveNotification>: Aggregation Group 2:
port member 277872640 becomes ACTIVE!
%May 7 09:23:16:322 2010 HP LAGG/5/LAGG_ACTIVE: Member port GigabitEthernet2/4/0/1 of
aggregation group BAGG2 becomes ACTIVE.
%May 7 09:23:16:370 2010 HP IFNET/3/LINK_UPDOWN: Bridge-Aggregation2 link status is UP.
%May 7 09:23:16:381 2010 HP IFNET/3/LINK_UPDOWN: Vlan-interface1 link status is UP.
%May 7 09:23:16:391 2010 HP IFNET/5/LINEPROTO_UPDOWN: Line protocol on the interface
Vlan-interface1 is UP.
The output information shows that IRF 2 is recovered. You can repair IRF 1 and the IRF links.
# When IRF 1 and the IRF links recover, reboot IRF 1. IRF 2 prompts that the IRF port is up and a new card
is inserted, which means that the two IRF fabrics has emerged.
%May 7 09:30:12:122 2010 HP STM/6/STM_LINK_STATUS_UP:
#May 7 09:30:36:566 2010 HP DEVM/1/BOARD INSERTED:
# Execute the
display irf
command. You can see that the IRF fabric is recovered, and Device B is the
master.
<Sysname> display irf
Switch Slot Role Priority CPU-Mac Description
1 1 Slave 1 00e0-fc0f-8c02 -----
*+2 1 Master 1 00e0-fc0f-8c20 -----
3 1 Slave 1 00e0-fc00-5801 -----
4 1 Slave 1 00e0-fc00-3583 -----
--------------------------------------------------
* indicates the device is the master.
+ indicates the device through which the user logs in.
Summary of Contents for A7503
Page 4: ...ii Index 56 ...
