Wireless Access

Hello all,

We have multiple class-C VLANS bind to a single VAP in bridge mode and we tagged all VLANs to all areas within the company.  It works fine so far, however I do not think it is very effective approach.  Should I tag different VLAN base on its location?  Say Area A got VLAN 10, Area B got VLAN12?  The DHCP scopes have been set up correctly for different VLANs. Would Aruba be able to assign correct IP address to the workstation based on different VLAN? 

In some areas with more than 256 workstations, I need to tag VLAN 13 and VLAN14.   I am curious how the IP addresses will be assigned. Will it use up one pool in one VLAN then the other or randomly try one after the others?  What happens if one pool is full and the other still have some free addresses?

Thanks

Ian

Question (s)

- What is your topology (how many controllers/aps)?

- Why are you using bridge mode?

3 controllers and about 100APs. Is that matter ?

The vendor who set up for us. He told us it is more efficient in our case.

How are all of your sites connected?

It does matter because if the design is complicated and it consumes alot of time managing it, it is not efficient.  Bridge mode makes sense in some circumstances, but becomes very difficult to manage in others.

one site with many blocks, all interconnected with 1G fibre/cat6 link.

If that is the case, make all the client traffic tunnel from the access points to the controller, instead of bridged.  That way all wireless clients have an ip address from a single range instead of selectively putting them on different wired VLANs.  This would significantly decrease complexity because they would all draw from a single pool of ip addresses.  As long as all clients can access anything they want from a single VLAN on the controller, they will be fine.

Many thanks for your advice. I totally agree your approach is much simpler.

However, I am still interested to know how the ip addresses are assigned in our existing setup.

Would it be in the documentation ?

In Bridge Mode, a client's VLAN is determined by the VLAN parameter in the Virtual AP profile and the Native Vlan Parameter in the AP system profile.

When the Virtual AP's traffic is set to bridged, if the Vlan parameter in the Virtual AP profile and the Native Vlan Parameter in the AP system profile are the same, it will bridge the client traffic to the same vlan as the physical interface of the AP that the client is on.  If they are different, the client traffic is bridged, but tagged with the Virtual AP VLAN, expecting the AP to be on a trunk.  Most people make it the same, so that they are merely bridging the traffic on the same VLAN as the wired AP.  Others who want to configure the AP's switch interface as a trunk and have client traffic end up on a different VLAN configure the Virtual AP Vlan different:

Please see here:  https://arubanetworkskb.secure.force.com/pkb/articles/FAQ/What-is-the-purpose-of-native-vlan-in-the-ap-system-profile-on-the-Aruba-Controller for details.

Thanks a lot.