Ask the Expert: High Availability on Wireless Lan Controller (WLC)

[ciscomoderator]

With Madhuri C.

Welcome to the Cisco Support Community Ask the Expert conversation. This is an opportunity to learn and ask questions of Cisco expert Madhuri C.about the new High Availability (HA) feature (that is, AP SSO) set within the Cisco Unified Wireless Network software release version 7.3 This feature allows the access point (AP) to establish a CAPWAP tunnel with the Active WLC and share a mirror copy of the AP database with the Standby WLC. The APs do not go into the Discovery state when the Active WLC fails and the Standby WLC takes over the network as the Active WLC.

Madhuri C. is a customer support engineer at the Cisco Technical Assistance Center in Bangalore, India. During her four years of experience she has worked on a wide range of Cisco wireless products and technology such as autonomous IOS (aIOS) access points, wireless routers, wireless LAN controllers, wireless VoIP phones, wireless control systems, network control systems, prime infrastructure, and mobility services engines. She has also worked in LAN switching technology.

Remember to use the rating system to let Madhuri know if you have received an adequate response. 

Madhuri might not be able to answer each question due to the volume expected during this event. Remember that you can continue the conversation on the Wireless Mobility sub community discussion forum shortly after the event. This event lasts through March,22 2013. Visit this forum often to view responses to your questions and the questions of other community members. 

More Information : http://www.cisco.com/en/US/products/ps10315/products_tech_note09186a0080bd3504.shtml

[Madhuri C]

Hi Janesh,

Both the WLCs in HA setup keep track of gateway reachability. The Active WLC sends an Internet Control Message Protocol (ICMP) ping to the gateway using the Management IP address as the source, and the Standby WLC sends an ICMP ping to the gateway using the Redundancy Management IP address. Both the WLCs send an ICMP ping to the gateway at a one-second interval.

Standby by default will be exchanging keepalives with primary and also parallely checking gateway reachability. Gateway reachablility check if for network related issues and keepalives for box failure issues.

The above algorithm will slightly change in event of failure. That is, it introduces additional ICMP to active WLC (not gateway) if keepalive is unanswered.

This is new ICMP to active WLC and gateway reachability check from standby will run parallel asusual.

Lets run through the algorithm for box failure from doc :

• The Standby WLC sends keep alive to the Active WLC and expects and acknowledgment within 100 msec as per the default timer. This can be configured in range from 100-400 msec.
• If there is no acknowledgment of keep alive within 100 msec, the Standby WLC immediately sends an ICMP message to the Active WLC via the redundant management interface in order to check if it is a box failover or some issue with Redundant Port connection.
• If there is no response to the ICMP message, the Standby WLC gets aggressive and immediately sends another keep alive message to the Standby WLC and expects an acknowledgment in 25% less time (that is, 75 msec or 25% less of 100 msec).
• If there is no acknowledgment of keep alive within 75 msec, the Standby WLC immediately sends another ICMP message to the Active WLC via the redundant management interface.
• Again, if there is no response for the second ICMP message, the Standby WLC gets more aggressive and immediately sends another keep alive message to the Standby WLC and expects an acknowledgment in time further 25% of actual timer less from last keep alive timer (that is, 50 msec or last keep alive timer of 75 msec - 25% less of 100 msec).
• If there is no acknowledgment of the third keep alive packet within 50 msec, the Standby WLC immediately sends another ICMP message to the Active WLC via the redundant management interface.
• Finally, if there is no response from the third ICMP packet, the Standby WLC declares the Active WLC is dead and assumes the role of the Active WLC.

Network Failover

In the case of a Network Failover (that is, the Active WLC cannot reach its gateway for some reason), it may take 3-4 seconds for a complete switchover depending on the number of APs in the network.

You can find stats using command :

show redundancy statistics 

        Redundancy Manager Statistics

        Keep Alive Request Send Counter          : 16
        Keep Alive Response Receive Counter      : 16

        Keep Alive Request Receive Counter       : 500322
        Keep Alive Response Send Counter         : 500322

        Ping Request to Default GW Counter       : 63360
        Ping Response from Default GW Counter    : 63360

        Ping Request to Peer Counter             : 12
        Ping Response from Peer Counter          : 3

        Keep Alive Loss Counter                  : 0
        Default GW Loss Counter                  : 0

        Local Physical Ports 1...8               : 10000000
        Peer  Physical Ports 1...8               : 10000000

Let me know if this answered your question.

Regards,

Madhuri

[janesh_abey]

Hi Madhuri,

Thanks for taking time to answer our queries.

It is all clear  to to me now.

Keep up the good work.

cheers,

janesh

[Madhuri C]

Glad to hear that. Thanks Janesh

[skhalil]

Hi Madhuri,

We are looking at deploying two 5508 in HA (Active/Active), is there a document or design blue print that would help us with the basic deployment and configuration of the units. Those two units would be homed in on two HA 6509E switches interconnected to each other using etherchannel trunks and running HSRP between them.

We are also looking at deploying two 5760 units in the exact same scenario as above. Your feedback would be much appreciated.

Thanks

Samer Khalil