MAC hash or round robin

Answered Question
Jun 10th, 2012
User Badges:

I am in the process of revamping our wireless solution. I am interested in VLAN pooling and am wondering if it is done round robin or with a hash. it is important for us to keep the same IP addresses so the MAC hash would be ideal over round robin in my attempts to keep users in the same VLAN.


Sent from Cisco Technical Support iPad App

Correct Answer by Amjad Abdullah about 5 years 2 months ago

Hi Robert,

It all depends on the code version on your WLC.

If you are using 7.0.116.0 or 7.0.220.0 it will use round robin. (need to confirm for 7.0.220.0 though).

If you use 7.2 or 7.0.230.0 (or later) it will use an index and same clients will be mapped to same interface as long as the DHCP process is going fine.

Integration of VLAN Pooling or the VLAN Select feature in release 7.0.116 provided a solution to the restriction where the WLAN can be mapped to a single interface or multiple interfaces using an interface group. Wireless clients associating to this WLAN receive an IP address from a pool of subnets identified by the interfaces in a round robin fashion.

In WLC release 7.2, the VLAN Select feature (which is supported only on the newer WLCs like 5508, WiSM-2, 7500, and 2500) was modified and now supports VLAN Select with a new modified algorithm. In the previous implementation, using the round robin algorithm was causing clients to obtain new IP addresses on every re-association, thus depleting IP addresses fast from the available DHCP pools. The new algorithm is based on the Client’s MAC address and operates in this way:

When a client associates to a WLAN on a controller, an index is calculated based on the MAC address of the client and the number of interfaces in the interface group using a hashing algorithm.

Based on this index, an interface is assigned to the client.

Whenever this client joins the controller, the hashing algorithm always returns the same index and the client is assigned to the same interface.

If the interface is “dirty”, then a random index is generated and the interface is assigned based on that random index.

If that interface is still dirty, then a fall back to round robin implementation occurs.

Note: In order to support the new VLAN Select feature on legacy controllers (such as the 4400 Series, WiSM, and 2100 Series) with the same MAC-based algorithm, the VLAN Select feature was modified in release 7.0.230 and now operates in the same fashion as release 7.2.

Reference:http://www.cisco.com/en/US/products/ps10315/products_tech_note09186a0080bb4900.shtml


Note please in underlined quote above it is metnioend that 7.0.230.0 uses same method as 7.2. having 7.0.220.0 not mentioned then I suppose that it follows same round robin as 7.0.116.0 but this needs to be confirmed in practice because documents are not always 100% accurate.


Also, for your benefit, here is the link specified for hte feature for 7.0.116.0 version:

http://www.cisco.com/en/US/products/ps10315/products_tech_note09186a0080b78900.shtml


HTH


Amjad

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 5 (1 ratings)
Loading.
Correct Answer
Amjad Abdullah Sun, 06/10/2012 - 22:35
User Badges:
  • Red, 2250 points or more

Hi Robert,

It all depends on the code version on your WLC.

If you are using 7.0.116.0 or 7.0.220.0 it will use round robin. (need to confirm for 7.0.220.0 though).

If you use 7.2 or 7.0.230.0 (or later) it will use an index and same clients will be mapped to same interface as long as the DHCP process is going fine.

Integration of VLAN Pooling or the VLAN Select feature in release 7.0.116 provided a solution to the restriction where the WLAN can be mapped to a single interface or multiple interfaces using an interface group. Wireless clients associating to this WLAN receive an IP address from a pool of subnets identified by the interfaces in a round robin fashion.

In WLC release 7.2, the VLAN Select feature (which is supported only on the newer WLCs like 5508, WiSM-2, 7500, and 2500) was modified and now supports VLAN Select with a new modified algorithm. In the previous implementation, using the round robin algorithm was causing clients to obtain new IP addresses on every re-association, thus depleting IP addresses fast from the available DHCP pools. The new algorithm is based on the Client’s MAC address and operates in this way:

When a client associates to a WLAN on a controller, an index is calculated based on the MAC address of the client and the number of interfaces in the interface group using a hashing algorithm.

Based on this index, an interface is assigned to the client.

Whenever this client joins the controller, the hashing algorithm always returns the same index and the client is assigned to the same interface.

If the interface is “dirty”, then a random index is generated and the interface is assigned based on that random index.

If that interface is still dirty, then a fall back to round robin implementation occurs.

Note: In order to support the new VLAN Select feature on legacy controllers (such as the 4400 Series, WiSM, and 2100 Series) with the same MAC-based algorithm, the VLAN Select feature was modified in release 7.0.230 and now operates in the same fashion as release 7.2.

Reference:http://www.cisco.com/en/US/products/ps10315/products_tech_note09186a0080bb4900.shtml


Note please in underlined quote above it is metnioend that 7.0.230.0 uses same method as 7.2. having 7.0.220.0 not mentioned then I suppose that it follows same round robin as 7.0.116.0 but this needs to be confirmed in practice because documents are not always 100% accurate.


Also, for your benefit, here is the link specified for hte feature for 7.0.116.0 version:

http://www.cisco.com/en/US/products/ps10315/products_tech_note09186a0080b78900.shtml


HTH


Amjad

Actions

This Discussion

 

 

Trending Topics - Security & Network