curious about how wireless to wired network bridges work

Discussion in 'Wireless Internet' started by yawnmoth, Mar 31, 2007.

  1. yawnmoth

    yawnmoth Guest

    i'm considering buying a wireless to wired network bridge (WET54G) and
    am not really sure how it'd work. say you're wireless network is
    using WEP or WPA. how do you tell the bridge that?

    the user guide (page 22) suggests that the WET54G has a web interface
    (presumably for this very reason). the default ip address to this web
    interface is 192.168.1.226. what happens, however, if your network
    configuration uses ip addresses of the form 10.*.*.*? i assume
    192.168.1.226 wouldn't work, at that point.

    ....or what if the ip addresses that are assigned by the router aren't
    internal - what if they're directly route-able to the internet? eg.
    what if, when i plug two computers into a hub that's plugged into the
    WET54G's lone ethernet port, i'm supposed to get two IP addresses that
    the outside world can see? 192.168.1.226 obviously wouldn't work,
    then...
     
    yawnmoth, Mar 31, 2007
    #1
    1. Advertisements

  2. yawnmoth

    seaweedsteve Guest

    YES


    the default ip address to this web
    You can change the bridge's address to whatever you want once you are
    in the interface.

    The address of the actual bridge is not the address the attached
    computer(s) will use, though it does have to be in the same subnet.
    If you don't give them static ip addresses, then they can get their
    addresses by DHCP from the central router that the bridge is connected
    to.

    I don't know if you can actually run two computers through the bridge
    or not.

    By the way, the cheaper way to do this is to use a router that can be
    set to serve as a client. The Buffalo WHR-G54S will do this and sells
    for under $50. Many others (Like the Linksys WRT54GL) can also do
    this if running aftermarket firmware like DD-WRT.

    Items sold exclusively as client bridges cost more and do less.

    Steve
     
    seaweedsteve, Apr 1, 2007
    #2
    1. Advertisements

  3. The absolute basics:
    1. ALL wireless is bridging. Wireless knows nothing about IP
    addresses and routeing.
    2. Basic wireless configuration consists SSID, encryption, and
    authentication. Everything else is set by the wireless access point.
    The bridge just follows.
    3. A wireless client adapter is nothing but a marketing term for
    wireless bridge.
    4. Bridges can be point to point or point to multipoint depending on
    configuration and topology.
    Argh. You read the manual. I'll try to undo the damage.

    The IP address for the WET54G is only used to configure the device. It
    has no real purpose in the operation of the bridge, which is all done
    on Layer 2 (MAC address layer) and not on Layer 3 (IP address layer).
    Nothing happens. The IP addresses are handled by the corresponding
    wired or wireless router along with the IP stack in the connecting
    computah. The bridge does NOTHING with the IP addresses.
    Again, the bridge does nothing. Other than the IP address required
    for configuration, the wireless bridge acts something like an ordinary
    CAT5 cable. It passes IP addresses unchanged. (Actually, that's not
    exactly true because a bridge only passes packets that have a
    destination MAC address on the other side of the bridge). However, at
    the IP address layer, a bridge looks like a cable.
    No. The hub is just a repeater. It doesn't do NAT in any way. If
    you need a 2nd IP address on the LAN for a 2nd computah, it will need
    to be supplied by a router.

    With NAT, the outside world sees exactly one IP address. That's
    you're routeable IP addresses and is the one that's reported by
    various internet sites the return your IP address:
    <http://whatsmyip.org>

    One of many features inside your router is NAT (network address
    translation). If you happen to be reading Cisco literature, it's
    really PAT (port address translation). What NAT/PAT does it rewrite
    the IP header on the LAN (local area network) side by port number
    giving you multiple IP addresses on the LAN side. These addresses are
    assigned by another feature called DHCP (dynamic host confusion
    protocol).

    Bottom line is that your questions about IP addresses are really
    handled by your router and that if you want multiple IP addresses, it
    cannot be done with just a wireless bridge or hub.

    Reminder: Wireless is all bridging and does NOT involve IP addresses.
     
    Jeff Liebermann, Apr 1, 2007
    #3
  4. yawnmoth

    seaweedsteve Guest


    Hah! I love it ! Normaly, the reprimand is "RTFM" This is the
    alternate reprimand for those who actually did ! Between the two,
    all possible questioners are covered.

    Cheers,
    Steve
     
    seaweedsteve, Apr 2, 2007
    #4
  5. Well, there's a bit of truth to my cynical attitude about manuals.
    I'll explain (whether you wanted it explained or not):

    1. These daze, manuals are mostly legal disclaimers and wholesale
    repudiation of resposibility. They are also incomplete as the most
    important information is usually on a sticker plastered over the
    connectors or on a "read this first" loose document generally lost
    among the packing material.

    2. Manuals only contain what to do, not how things work. If the user
    would take the time to understand how things work, they wouldn't need
    a manual as the proceedure would be fairly obvious. They also only
    cover the absolute basics, with many of the obscure settings
    remaining... well, obscure.

    3. Manuals no longer contains a "Theory of Operation" section. This
    section is quite common in the better network device manuals, but is
    missing in consumer products.

    4. The manual rarely follows the advances and feature additions in
    the firmware. It's often that the screenshot of a page does not
    resemble the actual page.

    5. None of the manuals ever seem to mention that it's a good idea to
    check for firmware updates on the manufactory web pile. Those that
    do, don't bother to explain how to install the update. Those that
    explain how to install the update, don't bother to explain how to
    decode the revision numbers. Those that actually include all the
    aformentioned, don't bother to explain what to do when the firmware
    update goes wrong.

    6. The troubleshooting section of most manuals are fairly crude and
    border on useless. None bother to suggest the ultimate
    troubleshooting proceedure, which it to punch the reset button and
    start over from scratch. They also never bother to suggest power
    cycleing the router if it seem to be hung or malfunctional.

    7. The very simplicity of the manual implies that the device being
    configured is also quite simple. This is rarely the case. Ever
    notice that the devices with the fewest features have the biggest
    manuals?

    8. Manuals rarely contain numbers such as the IP address of the
    router. This is intentional as the presentation of numbers, forumula,
    and RF theory, will cause the users brain to temporarily hang. It is
    the prime purpose of tech writers to remove all the numbers and useful
    information supplied by the engineers. However, if the manuals were
    written by engineers, nobody could understand them, so I guess this is
    a tolerable compromise.

    9. The addition of foreign language translations has converted the
    English into something resembling a foreign language.

    10. If there is a "help" section, there's often a question as to whom
    it was intended to help. I've suggested that manuals also include a
    "beyond help" section, but that has been consistantly vetoes by the
    tech writers as bad form and tactless.

    11. There seems to be a presumption that if the product were any good,
    it wouldn't need a manual. This is largely true but has been
    misinterpreted by the manufacturers as the size of the manual
    determines the quality of the product, where smaller is better.

    12. Manual writers are usually frustrated authors of fiction, who tend
    to be creative. This is not a good thing when dealing with technical
    terms designed to be specific. Every month, I see a new terms for
    common wireless devices.

    I could go on and on, but it's Monday morning and the computah is
    ringing with yet another dull and inefficient VoIP conference call.
    (Hello? I can't hear you, can you hear me? Talk slower. What? Don't
    talk when I'm interrupting. Garble-garble. Turn off the video. Can
    you hear me now? Ad Nausium)

    Reading manuals causes temporary brain damage. Writing manuals seems
    to make it permanent.
     
    Jeff Liebermann, Apr 2, 2007
    #5
  6. yawnmoth

    seaweedsteve Guest

    I agree! I read the manuals just in case there IS some useful
    information. There are sometimes gems of actual functional info
    buried in the mountains of noise and repetition. But it's usually
    all in the "set-up guide" as you say, and often comes down to a
    sentence or two that tells me something I actually need to know.
    Hey! I always thought it was the um, anyway, now I know. I was a
    tech writer in Silicon valley for a brief stint in the late 80s. I
    liked it, but they weren't consumer products I wrote about. I think
    that I got it right explained and organized all the loose knowledge
    about that Rolm mark 7600 confabulator into a cohesive document.

    But, of course, most people are hacks, don't have enough interest in
    what they do...


    Sounds like VOIP on landlines (vs Satellite) is problematic too. I had
    imagined that it would work for normal broadband...

    Steve
     
    seaweedsteve, Apr 4, 2007
    #6
  7. Yep. I just skim the docs for the numbers, URL's, and startup
    incantations to get me started. The rest is usually obvious.

    I once wrote a manual for a direction finder for the USCG. One full
    page on how to turn it on. Another full page on how to turn it off.
    However, the manual turned out quite well because I was being coached
    by an experienced tech writer from HP. Among the various lessons, I
    learned something that seems to be lacking in most consumer manuals.
    If the manual gives instructions for the user to do something, the
    manual should also show what the expected result should be. For
    example, if one is expected to type in a MAC address, the screen dump
    should show a real MAC address and what it should look like.
    I won't hold that against you. Not the tech writing part, but the
    part where you claimed to have enjoyed it. You can always recognize
    the experts in any field. They're the ones that do the job perfectly,
    and hate every minute of it because they've done it so often and for
    so long. Anyone who is still enjoying themselves, just hasn't done it
    often enough or long enough to become terminally disgusted and
    cynical. Incidentally, that's roughly why I've had 3 professions in
    my life and am considering a 4th.
    Argh. Telco manuals are written in a foreign language. It may
    resemble English, but telco policy is to never use any terms found
    elsewhere in the electronics industry. That dates back to the daze of
    the first AT&T breakup (there will be more breakups as soon as the
    Democrats get into power), where the various Baby Bells did not want
    to be accused of getting into the computah business. It may look,
    act, run, and smell like a computah, but to the telco crowd, it's
    called a switch. At the time, it was certain death to anyone that
    offered to publish a list of translations. Hopefully, you've
    recovered from the experience.
    It varies. I have customers that watch over my shoulder, take notes,
    ask good questions, and learn from the experience. I'm a frustrated
    instructor, so I tend to give a running commentary on what I'm doing.
    I often suspect they are interested because they don't want to pay me
    to fix it after they screw it up again, but I prefer to give my
    customers the benefit of the doubt. However, for most people, you're
    correct. They just want to plug and play without the learning
    experience. "Why can't this stuff be easier?" is the most common
    exhibition of frustration. I have various stock answers, but I never
    shove it in their face by mentioning that they don't spend the time to
    learn anything about their computers. Many of my customers entire
    business operations are based on the functioning of their computers,
    yet they treat them like it was some manner of fashion accessory. I
    don't complain, because they pay me, but an adjustment of priorities
    might be in order.
    You're assuming that this was with a dedicated broadband connection,
    and not an office corporate LAN. The problem is never the computer or
    even the ISP. It's the local constipation and bottleneck caused by
    what the user or his accomplices across the office are doing at the
    time he's trying to make a VoIP phone call. I have QoS setup to give
    priority to G.711 and G.729 traffic on my systems. Many routers do
    not. The result is that someone on the same WLAN connection watching
    YouTube videos at the same time as the VoIP call will trash the call.
    Even the VoIP users are clueless and will be doing something else that
    will slow their computer down. For example, one person was having
    nothing but trouble with the 9AM conference call, but would be fine at
    any other time. When I rescheduled her computahs virus scan for some
    other time, call quality dramatically improved.
     
    Jeff Liebermann, Apr 4, 2007
    #7
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.