> 
> I have been reading about floating device support in PF and the
> documentation seems to suggest that its main use is to enable you to
> connect a switch or AP (to which multiple other devices may be
> connected) to an access port on any existing PF managed switch, and
> can be moved at will to an access port on other PF managed switches
> without the need for isolation/registration on each move, either for
> the device or any of the devices connected to that device. (I'm a
> newbie so I'm open to correction here).

The goal is to allow infrastructure devices (APs, switches) to be
movable anywhere without the limitation of handling per-port
configuration exceptions.

The implementation is to reconfigure the port where the floating device
is plugging in so that it no longer has a limitation of one device per
port (imposed by port-security) and to allow reconfiguration of the
allowed VLANs to that port.

> 
> I was wondering if it matters whether a floating device is an endpoint
> device (eg: PC, printer, phone) rather than another connectivity
> device (eg: switch, hub, wireless AP).

I see no case in configuring printers, PCs or phones to be floating
devices and I see security and scalability problems doing.

> 
> The reason I ask is that it would be useful if I could configure
> certain (but not all) devices to never undergo isolation/registration
> no matter where they are connected on our network (an all Cisco
> network) even if they are moved from time to time around the network.
> We are a college, and basically, the devices connecting to our network
> are one of two kinds.....college owned and end user owned.

Port movement will never have the consequence of sending a user back to
isolation/registration. Once registered a device will stay so unless you
manually de-register it.

> 
> Ideally what I would like to be able to do is:
> 
> * Have all college owned devices (PCs, printers etc) configured as
> floating devices so they can be registered once, and then moved
> whenever we desire. Ideally the access port in most cases would be
> configured only with a primary access VLAN and no tagged VLANs.
> 
This is how packetfence works. No need to use floating devices for that.

To completely avoid registration even the first time, pre-register the
MACs by putting switches in registration mode (search archives for doing
so) or import the list of MACs using the node import feature.

> * Have all non college owned devices (end user laptops) connecting to
> either the wired or wireless network undergo the usual
> isolation/registration process at the network edge.
> 
> This has to allow for the possibility that an end user might
> disconnect one of our (floating) devices, and plug in their own.
> 
> Is this doable/feasible with PacketFence?
> 
Allow me to repeat myself: This is how packetfence works. No need to use
floating devices for that. ;)

> And lastly, for floating devices, is it still possible to use the
> custom VLAN assignment hooks via the perl code to assign a VLAN based
> on our local criteria?

No but you don't want floating devices ;)

Cheers!
-- 
Olivier Bilodeau
[email protected]  ::  +1.514.447.4918 *115  ::  www.inverse.ca
Inverse inc. :: Leaders behind SOGo (www.sogo.nu) and PacketFence
(www.packetfence.org)

------------------------------------------------------------------------------
EditLive Enterprise is the world's most technically advanced content
authoring tool. Experience the power of Track Changes, Inline Image
Editing and ensure content is compliant with Accessibility Checking.
http://p.sf.net/sfu/ephox-dev2dev
_______________________________________________
Packetfence-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/packetfence-users

Reply via email to