Cisco routers allow us to configure Unicast Mac address on routers interface by using mac-address command. However you will have notice that router reaction to this command is random. Sometimes it accepts this command and change Mac address but quite a few times it give us error.

Few Accepted Mac Addresses
mac-address bcde.9988.22aa
mac-address dcaa.4433.0066

Rejected / Malformed hex Mac Addresses
mac-address a1a1.bb11.ccee
mac-address ddaa.4433.0066
mac-address 4ba4.4433.0066
mac-address 8fa5.4433.0066
mac-address E3a6.4433.0066

You will get following error
% Malformed hex mac address

Explanation
This happen due to I/G bit. The Individual/Group (I/G) bit is used to indicate whether the destination address is a unicast (individual) or multicast (group) address. For a unicast address, the I/G bit is set to 0. For a multicast address, the I/G bit is set to 1. The broadcast address is a special case of multicast, and its I/G bit is set to 1. a Cisco router allows the user to configure a unicast MAC address, but not a multicast MAC address, on an interface, using the mac-address command.

 I/G bit is the rightmost bit in the leftmost byte or first byte or second nibble (A nibble consists of 4 bits or it constitutes to a hexadecimal unit). After explaining all that stuff, we have to come to know that if I/G bit will be 1 then we will have issues with our mac address. Let’s examine our given ip addresses and see if the same happens to them or not.

Verification:

Look at our malfunctioned mac addresses (a1a1.bb11.ccee , ddaa.4433.0066 , 4ba4.4433.0066 , 8fa5.4433.0066 , E3a6.4433.0066 ). The value of first byte is as follows
Look at the right most bit, In every case it is 1 and it shows that we are trying to give multi-cast address and that’s why rejected.

I hope that I have conveyed the matter.
 
It is understandable that nothing is bad or good in fact it is the capacity of the capable (The device) and/or understanding of the manager (The user or administrator) that defines and map things into good or bad realm. Turning philosophy into networks; both solutions are feasible, important and viable but design need special consideration upon requirement.  A query bakes requirement, Requirement need analysis and Analysis resulted into design and Design can lead to final success or failure. Let’s break and discuss both scenarios in a simplest manner.

Most businessmen are reluctant in buying a solution in which solution to a problem is holding their money and practically giving no benefit or once in a blue moon benefit. Whenever you speak to this sort of customers; they will show their concern upon the expense on active/standby failover solution as standby device will sit idle until active will get into trouble. The notion that we are paying double for an extra device doing nothing and sitting idle. Normally customers get interest in Active-Active solution.

In my opinion, this concern is genuine but its solution is not just to throw away A/S by adopting A/A solution; and one should handle it with good technical knowledge of a solution. One clarification to a point is that Redundancy and load-balancing are two different concepts. Load balancing a redundant solution means you are actually breaking redundancy rules; actually compromising 100% redundancy thing. Technically speaking, we add device so that if one device fail other will take its role but we don’t realize that both devices should have equal capability to handle data/traffic load. Considering 80% utilization of one device will face no problem if active gets down as other device with same capability will handle that 80% utilization with its 80% capacity or so and administrator will have opportunity and time to start working on a solution for upcoming increasing load.

In case of having A/A scenario, administrator will not even realize if both devices are giving 70% making total of 140% as everything went smoothly but what if one device fail at that time; Eventually total load will shift toward other device. Active device will only be able to handle 100% data or traffic and straight forwardly drop 40% of the data or malfunction. In practical, you would keep the load under 100% at all times even in A/A situation

I hope, I have made a point and you have understood it.

 
Picture
1. Don't be confused by seeing symbol. Both are routers, I have just changed router symbol in gns3 so that they look like switch.

2. GNS3 version 0.8.3



Add switching module in router so that router will now behave as switch too.
  • Right click on router and click configure
  • Click Slots
  • Add NM-16ESW into any one available slot
  • Apply and OK
Repeat this process on all routers you wanna make or let them behave as switch

Lets make a trunk link
First look at the image
R1 configuration

R1#show interfaces trunk


R1#show interfaces fastEthernet 1/0 switchport
Name: Fa1/0
Switchport: Enabled
Administrative Mode: static access
Operational Mode: static access

Administrative Trunking Encapsulation: dot1q
Operational Trunking Encapsulation: native
Negotiation of Trunking: Disabled
Access Mode VLAN: 1 (default)
Trunking Native Mode VLAN: 1 (default)
Trunking VLANs Enabled: ALL
Trunking VLANs Active: 1
Protected: false
Priority for untagged frames: 0
Override vlan tag priority: FALSE
Voice VLAN: none
Appliance trust: none


R1(config)#interface fastEthernet 1/0

R1(config-if)#switchport mode trunk

*Mar  1 00:01:49.023: %DTP-5-TRUNKPORTON: Port Fa1/0 has become dot1q trunk



Configuration Explanation

Initially when you give this command; Nothing will be shown as there is no trunk link

This command show following output


Currently its in access mode















Entering into interface mode

Changing interface mode to trunk

This is the output message showing that now port is in trunk mode



Let's see whats going on at R2 after doing above
R2#
*Mar  1 00:01:50.563: %SPANTREE-7-RECV_1Q_NON_TRUNK: Received 802.1Q BPDU on non trunk FastEthernet1/0 VLAN1.
*Mar  1 00:01:50.563: %SPANTREE-7-BLOCK_PORT_TYPE: Blocking FastEthernet1/0 on VLAN1. Inconsistent port type.PVST+: restarted the forward delay timer for FastEthernet1/0

Message is saying that DOT1Q BPDU are recieved on my non-trunk F 1/0

Repeat same process on R2 to create its required ports trunk
R1#show interfaces trunk

Port      Mode         Encapsulation  Status        Native vlan
Fa1/0     on                 802.1q         trunking          1


R2#show interfaces trunk

Port      Mode         Encapsulation  Status        Native vlan
Fa1/0     on                 802.1q         trunking          1




Done :)