Bridging ethernet over Wi-Fi?

Matt Minuti matt.minuti at gmail.com
Mon Nov 30 23:15:56 EST 2015


At my previous employer, I set up a wireless Ethernet bridge between two
buildings using some 5.4GHz ubiquiti hardware designed for the purpose,
costing like $200 total for both ends. It worked beautifully, was trivial
to set up, and as far as anyone could tell, there was a cable between the
buildings. Nanostation bridge, or something like that?

That said, I know nothing about network layers and all that. Say OSI model
and my eyes gloss over, no matter how many times I read about it. But it
seemed plenty transparent to me, however it worked!
On Nov 30, 2015 11:08 PM, "Bill McGonigle" <bill at bfccomputing.com> wrote:

> How much "ethernet" on the bridge is working?  I happened to run across
> this topic yesterday where it would have been really handy to bridge a
> VLAN trunk across a wifi bridge, but almost everything that says that
> it's "layer 2 transparent" is really just doing ARP proxy/masquerading
> of a sort, which works OK for stuff that happens to fit into WiFi
> framing but not other stuff.
>
> To do real layer 2 bridging seems to require doing L2 over L3 (L2TPv3,
> EoIP, VPLS, or eoMPLS) but that requires a cooperating endpoint, which
> you don't have, so it can't be that.  Aside: when I get a chance I'd
> like to get a config worked out to have an easily deployable L2TPv3
> bridge on a pair of openwrt boxes, for when the next thing breaks.
>
> -Bill
>
>
> On 11/13/2015 11:39 AM, Joshua Judson Rosen wrote:
> > I've noticed that VirtualBox and some other VM packages include
> > functionality to bridge in-VM network interfaces onto the external
> > network through a network interface on the host machine. I get how this
> works
> > for ethernet, but it seems like they make it work for Wi-Fi, too;
> > given that Wi-Fi frames don't actually carry enough information to
> > to do bridging at the station (client) end, how the heck does this
> > actually work?
> >
> > The only technique with which I'm really familiar, for bridging ethernet
> > through an 802.11(b|a|g|n) client interface, involves running the
> > radio in a non-standard 4-address mode, which requires support
> > on the AP at the other end of the wireless link; I'm fairly certain
> > that's not what the VM systems are doing, because they appear to
> > work with bog standard APs. So what _are_ they doing? Creating
> > a second (hidden?) interface on VM host with identical MAC address
> > to the interface inside the VM, and mirroring traffic between them?
> > Faking it by sniffing and relaying packets at layer 3? Something else?
> > Maybe something about this actually did get into a companion IEEE
> > standard that I'm not familiar with?
> >
>
>
> --
> Bill McGonigle, Owner
> BFC Computing, LLC
> http://bfccomputing.com/
> Telephone: +1.855.SW.LIBRE
> Email, IM, VOIP: bill at bfccomputing.com
> VCard: http://bfccomputing.com/vcard/bill.vcf
> Social networks: bill_mcgonigle/bill.mcgonigle
> _______________________________________________
> gnhlug-discuss mailing list
> gnhlug-discuss at mail.gnhlug.org
> http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.gnhlug.org/pipermail/gnhlug-discuss/attachments/20151130/8db734d5/attachment.html 


More information about the gnhlug-discuss mailing list