[kwlug-disc] recommendations for wireless ethernet bridge?
unsolicited at swiz.ca
Sun May 16 14:30:11 EDT 2010
Lori Paniak wrote, On 05/16/2010 1:09 PM:
> On Sun, 2010-05-16 at 12:55 -0400, unsolicited wrote:
>> zixiekat at gmail.com wrote, On 05/16/2010 11:59 AM:
>>> I currently do this with a Linksys WTG54GL and dd-wrt. The router
>>> connects to a remote wifi point and provides net access to clients
>>> connected to the switch on the router (and in my case, clients can
>>> also use a virtual wireless network, so I don't have to use the
>>> wired ports if I don't want to.)
>> 'Virtual wireless'?
>> Sounds like you're having one radio connecting to another's, and at
>> the same time providing (relay) connectivity for wi-fi computers. Am I
>> reading that correctly?
>> I take it from the comments in this thread that it's hw possible to
>> both 'receive' (accept clients to 'my' wi-fi SSID) / 'transmit'
>> (connect to another AP) / 'relay' (have same SSID as that other AP) on
>> one radio. That's not something I see much on retail packages - so is
>> that due to sw limitations / complexity then? (vs hw limitations.)
> I actually use WDS. Each satellite router connects to the gateway
> router wirelessly. There is one SSID in the system and one DHCP server.
> The satellite routers serve as wireless repeaters and offer wired
> connections with system DHCP via their switch ports. The WDS
> functionality is apparently an undocumented feature of the Linksys WRT
> routers. It is much better documented in OpenWRT.
> The more simple case is to use the router as a client to connect to the
> existing system wifi network. Then you can adjust the switch
> configuration so the LAN ports on the device serve system DHCP.
That far makes / made sense to me. What startled was the switch
'functionality' being extended (in turn?) to the wi-fi as well.
I always thought it strange that somewhere along the way the marketing
speak stopped distinguishing between bridge and router (pre-wi-fi),
and they 'admitted' that routers inherently did bridging (if that's
all you wanted.) To that point, bridges were stupid expensive compared
to routers - which irked when they were exactly, or substantially, the
same guts. Still are, as Bob's messages points out.
I never understood why that (bridging/relaying) didn't extend to
wi-fi, so assumed one radio could only do one SSID (which, if the same
SSID as the gateway, caused it to automagically turn into merely a
repeater). My guess was the hardware doesn't change, so why it didn't
extend to wi-fi I didn't understand. Thinking about that, with the
prior message, I assumed the same 'radio' would have to split its
broadcast time between two SSIDs, which doesn't strike me as intuitive.
Thanks for the info about WDS. It makes me guess, given the above
paragraph, that the functional restriction had more to do with the
limited web interface of stock routers, rather than hardware presence
or platform capability.
The 'virtual wi-fi' term startled me, but remembering, now, that
virtual provides duplicate functionality over the same physical
hardware, makes it go down a little easier.
> Clear as mud? Maybe Rob should bring in a router for a 20 minute
I'm suspecting, now, that the back end / front end different wi-fi
networks seems strange, or new these days, really goes back to limited
web interfaces. I'd bet, even today, most stock router interfaces have
no facility for filtering (ip rules) between the two SSIDs. The
hardware may be present, but not the software interface, so the
capability went unmentioned? And the interface has since been
expanded? [Ignoring that OpenWRT may have been doing this from day one.]
The Netgear WNDR3700 I have on order startled me in that the marketing
blurb, for the first time I'd seen, talks about presenting both a
public and a private wi-fi SSID. It will be interesting to see if its
web interface allows for specifying ip access lists between the two
SSIDs. Let alone between the back end and front end wi-fi's and the
hard links. [No, not a presentation. I would just point you to the
More information about the kwlug-disc