WEP Key

Archived from groups: alt.internet.wireless (More info?)

Every Internet article I read on WEP assails me with the
necessity of using "secure" WEP keys from random hex number
generators, etc. Since the probability of a "hacker" living
within 100 yards of me is practically zero, I only want to keep
out the average Joe Doe. My impression (but not certainty) is
that ANY group of numbers and letters is a hexadecimal number if
a user or program wishes to treat it as such. Therefore, my
question is: When asked to enter a 10 character hex WEP Key, is
something like "SamJones11" workable?

Thanks,
John
18 answers Last reply
More about tomshardware
  1. Archived from groups: alt.internet.wireless (More info?)

    On Thu, 15 Jul 2004 15:10:45 -0700, John <john@nospam.net> wrote:

    >Every Internet article I read on WEP assails me with the
    >necessity of using "secure" WEP keys from random hex number
    >generators, etc. Since the probability of a "hacker" living
    >within 100 yards of me is practically zero, I only want to keep
    >out the average Joe Doe. My impression (but not certainty) is
    >that ANY group of numbers and letters is a hexadecimal number if
    >a user or program wishes to treat it as such. Therefore, my
    >question is: When asked to enter a 10 character hex WEP Key, is
    >something like "SamJones11" workable?


    Thanks, Jim, Barry, Jeff. I knew that A-F restriction, once.
    Really. :-)

    John
  2. Archived from groups: alt.internet.wireless (More info?)

    ugh...i hate wep and their dang hex keys...

    no the "samjones11" won't work because it isn't hexadecimal.

    hexadecimal is represented by numerals 0-9 and letters A-F only.

    good luck

    jtm


    "John" <john@nospam.net> wrote in message
    news:0uudf0981h5ja5lukr9vum4rhvj0eb985r@4ax.com...
    [...]

    When asked to enter a 10 character hex WEP Key, is
    something like "SamJones11" workable?

    Thanks,
    John
  3. Archived from groups: alt.internet.wireless (More info?)

    On Thu, 15 Jul 2004 15:10:45 -0700, John wrote:

    > Every Internet article I read on WEP assails me with the
    > necessity of using "secure" WEP keys from random hex number
    > generators, etc. Since the probability of a "hacker" living
    > within 100 yards of me is practically zero, I only want to keep
    > out the average Joe Doe. My impression (but not certainty) is
    > that ANY group of numbers and letters is a hexadecimal number if
    > a user or program wishes to treat it as such. Therefore, my
    > question is: When asked to enter a 10 character hex WEP Key, is
    > something like "SamJones11" workable?
    >
    > Thanks,
    > John

    No, that's my son's name. You'll have to ask him first. It's also not a hex
    string, which consists of the digits 0 through 9 and the letters a through
    f. On the other hand, if you are referring to a pass phrase, which your
    router's config utility may ask for, then you can certainly use a name. The
    utility will encode the name into a series of hex digits, and the result
    will be the WEP key. You'll have to copy down the hex key, and enter it
    into the clients' config utilities.

    Hope that helps.

    Barry Jones
  4. Archived from groups: alt.internet.wireless (More info?)

    On Thu, 15 Jul 2004 15:10:45 -0700, John <john@nospam.net> wrote:

    >Every Internet article I read on WEP assails me with the
    >necessity of using "secure" WEP keys from random hex number
    >generators, etc. Since the probability of a "hacker" living
    >within 100 yards of me is practically zero, I only want to keep
    >out the average Joe Doe.

    Now days, hacking is done in a vehicle with Netstumbler and a laptop.
    No need to live nearby. Once you are discovered, the rest is a matter
    of capturing enough traffic to recover the WEP key.

    >My impression (but not certainty) is
    >that ANY group of numbers and letters is a hexadecimal number if
    >a user or program wishes to treat it as such. Therefore, my
    >question is: When asked to enter a 10 character hex WEP Key, is
    >something like "SamJones11" workable?

    Nope. It's not in hexidismal. You need to use number 0->9 and the
    letters A->F. Try not to use something repeative as it's easily
    cracked.


    --
    # Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
    # 831.336.2558 voice http://www.LearnByDestroying.com
    # jeffl@comix.santa-cruz.ca.us
    # 831.421.6491 digital_pager jeffl@cruzio.com AE6KS
  5. Archived from groups: alt.internet.wireless (More info?)

    Looking over some of the answers you've gotten, I think some miss the point.
    You are saying that you have read some articles indicating that you should
    use keys generated by a random hex number generator. This issue is separate
    from the issue of how you configure the key into the client or the AP.

    The reason the articles urge you to use a random key generator is that if
    you select a meaningful text string, it is very likely to be a dictionary
    word, or a concatenation of dictionary words. You might be surprised at how
    much this simplifies the task of cracking the key, especially if you use a
    64-bit (sometimes called 40-bit) key. It reduces the total keyspace to a
    much smaller number of potential keys, which can be searched by brute force.
    A random key generator will generate an arbitrary sequence of hexadecimal
    digits, and there are far more of these than there are simple ASCII text
    strings. Even so, as someone else pointed out, if you use a random 64-bit
    hex key, it's still looks quite feasible to crack it by brute force in under
    an hour.

    Use the longest allowed length, 128 bits at least, and use a randomly
    selected hex key. Some clients and APs provide a passphrase generator
    (Netgear, Linksys, for example). These take an ASCII string and generate
    pseudorandom hex keys. I understand that many of these generators are
    derived from the Linksys algorithm. In any case, if you use simple
    dictionary words to seed such a passphrase generator, your key can still be
    brute-forced, because hackers also have the generators. Use a seed that is
    as random as possible.

    If you enter keys by hand, your config menu may allow you to select between
    ASCII and hex data entry. D-Link supports this.

    "John" <john@nospam.net> wrote in message
    news:0uudf0981h5ja5lukr9vum4rhvj0eb985r@4ax.com...
    > Every Internet article I read on WEP assails me with the
    > necessity of using "secure" WEP keys from random hex number
    > generators, etc. Since the probability of a "hacker" living
    > within 100 yards of me is practically zero, I only want to keep
    > out the average Joe Doe. My impression (but not certainty) is
    > that ANY group of numbers and letters is a hexadecimal number if
    > a user or program wishes to treat it as such. Therefore, my
    > question is: When asked to enter a 10 character hex WEP Key, is
    > something like "SamJones11" workable?
    >
    > Thanks,
    > John
    >
  6. Archived from groups: alt.internet.wireless (More info?)

    gary wrote:
    > Looking over some of the answers you've gotten, I think some miss the point.
    > You are saying that you have read some articles indicating that you should
    > use keys generated by a random hex number generator. This issue is separate
    > from the issue of how you configure the key into the client or the AP.
    >
    > The reason the articles urge you to use a random key generator is that if
    > you select a meaningful text string, it is very likely to be a dictionary
    > word, or a concatenation of dictionary words. You might be surprised at how
    > much this simplifies the task of cracking the key, especially if you use a
    > 64-bit (sometimes called 40-bit) key. It reduces the total keyspace to a
    > much smaller number of potential keys, which can be searched by brute force.
    > A random key generator will generate an arbitrary sequence of hexadecimal
    > digits, and there are far more of these than there are simple ASCII text
    > strings. Even so, as someone else pointed out, if you use a random 64-bit
    > hex key, it's still looks quite feasible to crack it by brute force in under
    > an hour.
    >
    > Use the longest allowed length, 128 bits at least, and use a randomly
    > selected hex key. Some clients and APs provide a passphrase generator
    > (Netgear, Linksys, for example). These take an ASCII string and generate
    > pseudorandom hex keys. I understand that many of these generators are
    > derived from the Linksys algorithm. In any case, if you use simple
    > dictionary words to seed such a passphrase generator, your key can still be
    > brute-forced, because hackers also have the generators. Use a seed that is
    > as random as possible.
    >
    > If you enter keys by hand, your config menu may allow you to select between
    > ASCII and hex data entry. D-Link supports this.
    >
    > "John" <john@nospam.net> wrote in message
    > news:0uudf0981h5ja5lukr9vum4rhvj0eb985r@4ax.com...
    >
    >>Every Internet article I read on WEP assails me with the
    >>necessity of using "secure" WEP keys from random hex number
    >>generators, etc. Since the probability of a "hacker" living
    >>within 100 yards of me is practically zero, I only want to keep
    >>out the average Joe Doe. My impression (but not certainty) is
    >>that ANY group of numbers and letters is a hexadecimal number if
    >>a user or program wishes to treat it as such. Therefore, my
    >>question is: When asked to enter a 10 character hex WEP Key, is
    >>something like "SamJones11" workable?
    >>
    >>Thanks,
    >>John
    >>
    >
    >
    >
    D-Link tech support told me not to use ASCII WEP key entry if you have
    Windows XP - there is a problem. Use HEX keys instead.
  7. Archived from groups: alt.internet.wireless (More info?)

    "Lance Courtland" <oscarman4@yahoo.com> wrote in message
    news:j_EJc.88539$IQ4.79530@attbi_s02...
    > gary wrote:
    > > Looking over some of the answers you've gotten, I think some miss the
    point.
    > > You are saying that you have read some articles indicating that you
    should
    > > use keys generated by a random hex number generator. This issue is
    separate
    > > from the issue of how you configure the key into the client or the AP.
    > >
    > > The reason the articles urge you to use a random key generator is that
    if
    > > you select a meaningful text string, it is very likely to be a
    dictionary
    > > word, or a concatenation of dictionary words. You might be surprised at
    how
    > > much this simplifies the task of cracking the key, especially if you use
    a
    > > 64-bit (sometimes called 40-bit) key. It reduces the total keyspace to a
    > > much smaller number of potential keys, which can be searched by brute
    force.
    > > A random key generator will generate an arbitrary sequence of
    hexadecimal
    > > digits, and there are far more of these than there are simple ASCII text
    > > strings. Even so, as someone else pointed out, if you use a random
    64-bit
    > > hex key, it's still looks quite feasible to crack it by brute force in
    under
    > > an hour.
    > >
    > > Use the longest allowed length, 128 bits at least, and use a randomly
    > > selected hex key. Some clients and APs provide a passphrase generator
    > > (Netgear, Linksys, for example). These take an ASCII string and generate
    > > pseudorandom hex keys. I understand that many of these generators are
    > > derived from the Linksys algorithm. In any case, if you use simple
    > > dictionary words to seed such a passphrase generator, your key can still
    be
    > > brute-forced, because hackers also have the generators. Use a seed that
    is
    > > as random as possible.
    > >
    > > If you enter keys by hand, your config menu may allow you to select
    between
    > > ASCII and hex data entry. D-Link supports this.
    > >
    > > "John" <john@nospam.net> wrote in message
    > > news:0uudf0981h5ja5lukr9vum4rhvj0eb985r@4ax.com...
    > >
    > >>Every Internet article I read on WEP assails me with the
    > >>necessity of using "secure" WEP keys from random hex number
    > >>generators, etc. Since the probability of a "hacker" living
    > >>within 100 yards of me is practically zero, I only want to keep
    > >>out the average Joe Doe. My impression (but not certainty) is
    > >>that ANY group of numbers and letters is a hexadecimal number if
    > >>a user or program wishes to treat it as such. Therefore, my
    > >>question is: When asked to enter a 10 character hex WEP Key, is
    > >>something like "SamJones11" workable?
    > >>
    > >>Thanks,
    > >>John
    > >>
    > >
    > >
    > >
    > D-Link tech support told me not to use ASCII WEP key entry if you have
    > Windows XP - there is a problem. Use HEX keys instead.

    This may depend on the adapter. I've used random ASCII strings on my AG650
    without problem. I finally wrote a few lines of C to generate random hex
    keys.
  8. Archived from groups: alt.internet.wireless (More info?)

    On Fri, 16 Jul 2004 02:15:26 GMT, "gary" <pleasenospam@sbcglobal.net>
    wrote:

    >> D-Link tech support told me not to use ASCII WEP key entry if you have
    >> Windows XP - there is a problem. Use HEX keys instead.

    I'd be interested in knowing what model D-Link's are affected and what
    manner of problem they're having.

    >This may depend on the adapter. I've used random ASCII strings on my AG650
    >without problem. I finally wrote a few lines of C to generate random hex
    >keys.

    Well, there are a bunch of online WEP key generators.
    http://www.warewolflabs.com/portfolio/programming/wepskg/wepskg.html

    http://www.warewolflabs.com/portfolio/programming/wepskg/wepskg_v21.zip
    https://www.wireless.org.au/~jhecker/wepgen/index.php
    http://www.clariondeveloper.com/wepgen/
    http://www.wincatalog.com/pasgen.html?source=pasgen124
    http://www.powerdog.com/wepkey.cgi
    http://homepage.mac.com/chally/toolsandutils.html (for Mac)

    Stolen from the PracticallyNetworked.com web pile:

    One ASCII Character is Eight (8) Bits
    One HEX Character is Four (4) Bits
    40 or 64 bit ASCII WEP code has 5 characters
    40 or 64 bit HEX WEP code has 10 characters
    128 bit ASCII WEP code has 13 characters
    128 bit HEX WEP code has 26 characters


    --
    # Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
    # 831.336.2558 voice http://www.LearnByDestroying.com
    # jeffl@comix.santa-cruz.ca.us
    # 831.421.6491 digital_pager jeffl@cruzio.com AE6KS
  9. Archived from groups: alt.internet.wireless (More info?)

    "Jeff Liebermann" <jeffl@comix.santa-cruz.ca.us> wrote in message
    news:g4fef09jenhsi2lgopjsa63nlma1s2quhr@4ax.com...
    > On Fri, 16 Jul 2004 02:15:26 GMT, "gary" <pleasenospam@sbcglobal.net>
    > wrote:
    >
    > >> D-Link tech support told me not to use ASCII WEP key entry if you have
    > >> Windows XP - there is a problem. Use HEX keys instead.
    >
    > I'd be interested in knowing what model D-Link's are affected and what
    > manner of problem they're having.
    >
    > >This may depend on the adapter. I've used random ASCII strings on my
    AG650
    > >without problem. I finally wrote a few lines of C to generate random hex
    > >keys.
    >
    > Well, there are a bunch of online WEP key generators.

    True. But it only requires a few lines of C using rand() or srand(). That's
    probably why there are so many - it's a trivial wheel to reinvent. I needed
    an excuse to test the gnu compiler I loaded with Cygwin.

    > http://www.warewolflabs.com/portfolio/programming/wepskg/wepskg.html
    >
    > http://www.warewolflabs.com/portfolio/programming/wepskg/wepskg_v21.zip
    > https://www.wireless.org.au/~jhecker/wepgen/index.php
    > http://www.clariondeveloper.com/wepgen/
    > http://www.wincatalog.com/pasgen.html?source=pasgen124
    > http://www.powerdog.com/wepkey.cgi
    > http://homepage.mac.com/chally/toolsandutils.html (for Mac)
    >
    > Stolen from the PracticallyNetworked.com web pile:
    >
    > One ASCII Character is Eight (8) Bits
    > One HEX Character is Four (4) Bits
    > 40 or 64 bit ASCII WEP code has 5 characters
    > 40 or 64 bit HEX WEP code has 10 characters
    > 128 bit ASCII WEP code has 13 characters
    > 128 bit HEX WEP code has 26 characters
    >
    >
    >
    >
    > --
    > # Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
    > # 831.336.2558 voice http://www.LearnByDestroying.com
    > # jeffl@comix.santa-cruz.ca.us
    > # 831.421.6491 digital_pager jeffl@cruzio.com AE6KS
  10. Archived from groups: alt.internet.wireless (More info?)

    Jeff Liebermann wrote:
    > On Fri, 16 Jul 2004 02:15:26 GMT, "gary" <pleasenospam@sbcglobal.net>
    > wrote:
    >
    >
    >>>D-Link tech support told me not to use ASCII WEP key entry if you have
    >>>Windows XP - there is a problem. Use HEX keys instead.
    >
    >
    > I'd be interested in knowing what model D-Link's are affected and what
    > manner of problem they're having.

    Jeff:

    I'm using a D-Link DWL-2100AP Access Point connected to a d-Link
    DI-704P Router on a Win98SE machine. The Dell laptop is running Win XP
    Pro with a a DWL-G650 cardbus adapter. The only thing they said was
    "use hex keys - ASCII doesn't work very well".

    Lance Courtland
  11. Archived from groups: alt.internet.wireless (More info?)

    On Fri, 16 Jul 2004 04:02:05 GMT, Lance Courtland
    <oscarman4@yahoo.com> wrote:

    >I'm using a D-Link DWL-2100AP Access Point connected to a d-Link
    >DI-704P Router on a Win98SE machine. The Dell laptop is running Win XP
    >Pro with a a DWL-G650 cardbus adapter. The only thing they said was
    >"use hex keys - ASCII doesn't work very well".

    Now all I have to do is figure out what they mean by "doesn't work
    very well". It either works or it doesn't, with no grades of
    malfunction in between. I know some of the drivers do not include an
    ASCII key generator but I think DLink has them. My guess(tm) is that
    there was some creativity or difference in the ASCII to hexadecimal
    WEP key conversion on either MS XP or D-Link. It's fairly difficult
    to screw up something that simple, but it's possible.

    I have a DWL-G650 and a DWL-900AP+ in stock. Close enough. Time
    permitting, I'll give an ASCII key a try and see what breaks.

    Thanks much.


    --
    # Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
    # 831.336.2558 voice http://www.LearnByDestroying.com
    # jeffl@comix.santa-cruz.ca.us
    # 831.421.6491 digital_pager jeffl@cruzio.com AE6KS
  12. Archived from groups: alt.internet.wireless (More info?)

    "John" <john@nospam.net> wrote in message
    news:0uudf0981h5ja5lukr9vum4rhvj0eb985r@4ax.com...
    > Every Internet article I read on WEP assails me with the
    > necessity of using "secure" WEP keys from random hex number
    > generators, etc. Since the probability of a "hacker" living
    > within 100 yards of me is practically zero, I only want to keep
    > out the average Joe Doe.

    I dont broadcast my SSID, let 'them' figure it out...
  13. Archived from groups: alt.internet.wireless (More info?)

    "Jeff Liebermann" <jeffl@comix.santa-cruz.ca.us> wrote in message
    news:54oef01bksgi2sj9q1kvuk87j9vp1p91id@4ax.com...
    > On Fri, 16 Jul 2004 04:02:05 GMT, Lance Courtland
    > <oscarman4@yahoo.com> wrote:
    >
    > >I'm using a D-Link DWL-2100AP Access Point connected to a d-Link
    > >DI-704P Router on a Win98SE machine. The Dell laptop is running Win XP
    > >Pro with a a DWL-G650 cardbus adapter. The only thing they said was
    > >"use hex keys - ASCII doesn't work very well".
    >
    > Now all I have to do is figure out what they mean by "doesn't work
    > very well". It either works or it doesn't, with no grades of
    > malfunction in between. I know some of the drivers do not include an
    > ASCII key generator but I think DLink has them. My guess(tm) is that
    > there was some creativity or difference in the ASCII to hexadecimal
    > WEP key conversion on either MS XP or D-Link. It's fairly difficult
    > to screw up something that simple, but it's possible.

    D-Link support has told me fairy tales in the past. The quality of the
    information you get from them depends on the luck of the draw. I've had an
    AG650 for a year and a half. I used to use ASCII key entry exclusively,
    never had a single problem. I find it hard to believe that there is any
    significant difference between the G650 and the AG650 config software. I've
    used it both with original version and latest upgrades.

    I'm not sure what you mean by "ASCII to hex WEP key conversion". If you
    enter a 5-character ASCII string in D-Link config, the bytes of this string
    are used directly as the 40-bit fixed key portion. It's not a key generator,
    it's simple key entry, no conversion. How can that get screwed up? I suppose
    if you embed blanks you might get unexpected results, but that would be an
    unlikely exceptional case, hardly justifying the claim that it's broken.
    >
    > I have a DWL-G650 and a DWL-900AP+ in stock. Close enough. Time
    > permitting, I'll give an ASCII key a try and see what breaks.
    >
    > Thanks much.
    >
    >
    > --
    > # Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
    > # 831.336.2558 voice http://www.LearnByDestroying.com
    > # jeffl@comix.santa-cruz.ca.us
    > # 831.421.6491 digital_pager jeffl@cruzio.com AE6KS
  14. Archived from groups: alt.internet.wireless (More info?)

    On Fri, 16 Jul 2004 14:42:30 GMT, "gary" <pleasenospam@sbcglobal.net>
    wrote:

    >D-Link support has told me fairy tales in the past.

    I rarely call support, so I miss this part of the troubleshooting
    adventure. I prefer to Google for recycled user experiences and
    over-drawn conclusions, for clues. In general, methinks tech support
    has hit an all time low since almost all of it is now outsourced.

    >I'm not sure what you mean by "ASCII to hex WEP key conversion".
    (...)
    >How can that get screwed up?

    Sorry. That was my failed attempt at techy humor. There is no
    conversion. Yet, I suspect some errant programmist is capable of bit
    shifting, truncating, or otherwise screwing up what may be the
    simplest exercise in programming possible. I saw this in some
    unrelease code, where the 128bit ASCII WEP key magically went from the
    normal 26 characters, to 13 characters thanks to the author using
    Unicode libraries (2 bytes per character). Every other byte was 00.

    If you wanna look for bugs, you'll find that most radios correctly
    accept 32 characters for the SSID. However, there are a few losers
    (i.e. SMC7004AWR) that will only take 31 because they included the
    null string terminator in the count. Not a big deal unless you use
    disgustingly long SSID's.

    Moral: There's always a way to do it wrong.


    --
    Jeff Liebermann jeffl@comix.santa-cruz.ca.us
    150 Felker St #D http://www.LearnByDestroying.com
    Santa Cruz CA 95060 AE6KS 831-336-2558
  15. Archived from groups: alt.internet.wireless (More info?)

    "Jeff Liebermann" <jeffl@comix.santa-cruz.ca.us> wrote in message
    news:d7uff05tt7etlfpqjdr0naba4dearafla5@4ax.com...
    > On Fri, 16 Jul 2004 14:42:30 GMT, "gary" <pleasenospam@sbcglobal.net>
    > wrote:
    >
    > >D-Link support has told me fairy tales in the past.
    >
    > I rarely call support, so I miss this part of the troubleshooting
    > adventure. I prefer to Google for recycled user experiences and
    > over-drawn conclusions, for clues. In general, methinks tech support
    > has hit an all time low since almost all of it is now outsourced.
    >
    > >I'm not sure what you mean by "ASCII to hex WEP key conversion".
    > (...)
    > >How can that get screwed up?
    >
    > Sorry. That was my failed attempt at techy humor. There is no
    > conversion. Yet, I suspect some errant programmist is capable of bit
    > shifting, truncating, or otherwise screwing up what may be the
    > simplest exercise in programming possible. I saw this in some
    > unrelease code, where the 128bit ASCII WEP key magically went from the
    > normal 26 characters, to 13 characters thanks to the author using
    > Unicode libraries (2 bytes per character). Every other byte was 00.
    >
    > If you wanna look for bugs, you'll find that most radios correctly
    > accept 32 characters for the SSID. However, there are a few losers
    > (i.e. SMC7004AWR) that will only take 31 because they included the
    > null string terminator in the count. Not a big deal unless you use
    > disgustingly long SSID's.
    >
    > Moral: There's always a way to do it wrong.


    Ain't it the truth. But if D-Link does any WEP testing at all, they'd have
    to break things the same way on routers and clients in order for WEP to
    appear to work at all. Since my AG650 also works with a friend's Netgear
    router, using a variety of direct-entered ASCII keys, I'm guessing that
    ASCII key entry is probably okay, and someone in Mumbai over-interpreted a
    troubleshooting database item.

    Their coders are at least as likely to get display-hex/binary conversion
    wrong.

    >
    >
    > --
    > Jeff Liebermann jeffl@comix.santa-cruz.ca.us
    > 150 Felker St #D http://www.LearnByDestroying.com
    > Santa Cruz CA 95060 AE6KS 831-336-2558
  16. Archived from groups: alt.internet.wireless (More info?)

    On Fri, 16 Jul 2004 18:45:29 GMT, "gary" <pleasenospam@sbcglobal.net>
    wrote:

    >Ain't it the truth. But if D-Link does any WEP testing at all, they'd have
    >to break things the same way on routers and clients in order for WEP to
    >appear to work at all.

    Wrong. DLink gets their hardware and software from multiple vendors.
    Zcom, Bromax, Nihon Dengwa, ad nausium. Just go to the FCC ID web
    pile and stuff in a few prefixes used by DLink. They even have
    multiple vendors for various mutations of the same product. (Don't
    ask me how the FCC allows this manner of creativity without
    recertification). Anyway, there's plenty of potential for creativity,
    divergence, and the infamous regression testing.

    >Since my AG650 also works with a friend's Netgear
    >router, using a variety of direct-entered ASCII keys, I'm guessing that
    >ASCII key entry is probably okay, and someone in Mumbai over-interpreted a
    >troubleshooting database item.

    They're not outsourced. They're employees of DLink in India:
    http://www.dlink.co.in/dlink/Contacts/branches.htm
    http://www.dlink.co.in/dlink/Corporate/employment.htm
    http://www.dlink.co.in/dlink/software/job.htm

    Support may actually be correct. I vaguely recall walking a customer
    though a WEP setup using a Linksys something router and an HP laptop
    with a Centrino incantation. WEP would work, but only if I used a
    hexadecimal key. This was before the various wireless updates from
    Microsoft and before several Linksys firmware updates, so I'm fairly
    sure it was fixed long ago. However, it just proves that it can
    happen.

    >Their coders are at least as likely to get display-hex/binary conversion
    >wrong.


    --
    # Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
    # 831.336.2558 voice http://www.LearnByDestroying.com
    # jeffl@comix.santa-cruz.ca.us
    # 831.421.6491 digital_pager jeffl@cruzio.com AE6KS
  17. Archived from groups: alt.internet.wireless (More info?)

    "Jeff Liebermann" <jeffl@comix.santa-cruz.ca.us> wrote in message
    news:esrgf099cf2h7ci3rorbbd046asoohn631@4ax.com...
    > On Fri, 16 Jul 2004 18:45:29 GMT, "gary" <pleasenospam@sbcglobal.net>
    > wrote:
    >
    > >Ain't it the truth. But if D-Link does any WEP testing at all, they'd
    have
    > >to break things the same way on routers and clients in order for WEP to
    > >appear to work at all.
    >
    > Wrong. DLink gets their hardware and software from multiple vendors.
    > Zcom, Bromax, Nihon Dengwa, ad nausium. Just go to the FCC ID web
    > pile and stuff in a few prefixes used by DLink. They even have
    > multiple vendors for various mutations of the same product. (Don't
    > ask me how the FCC allows this manner of creativity without
    > recertification). Anyway, there's plenty of potential for creativity,
    > divergence, and the infamous regression testing.

    Yes, I know, didn't say different. My DI-774 router is such a beast. What I
    meant was that, *if* ASCII key entry is seriously broken on the G650 client
    card, then *if* they do any interoperability testing with routers they
    market (whoever manufactures them), I would expect the G650 to fail unless
    the routers are using the same data entry interface, broken in the same way.
    How else would both sides end up with the same WEP key, if one of them uses
    a seriously broken input method?

    Wi-fi Alliance certification also involves a certain level of inter-vendor
    interoperability testing, but I doubt if ASCII key entry is involved (basic
    WEP functionality certainly is).

    >
    > >Since my AG650 also works with a friend's Netgear
    > >router, using a variety of direct-entered ASCII keys, I'm guessing that
    > >ASCII key entry is probably okay, and someone in Mumbai over-interpreted
    a
    > >troubleshooting database item.
    >
    > They're not outsourced. They're employees of DLink in India:
    > http://www.dlink.co.in/dlink/Contacts/branches.htm
    > http://www.dlink.co.in/dlink/Corporate/employment.htm
    > http://www.dlink.co.in/dlink/software/job.htm

    But they probably work in Mumbai. In any case, it's a Hong Kong outfit, so
    in a sense, the whole company is outsourced. Along with several other
    low-end commercial vendors. Nothing wrong with that, so long as everyone in
    the company pulls their oars in the same direction (and at roughly the same
    time).

    >
    > Support may actually be correct. I vaguely recall walking a customer
    > though a WEP setup using a Linksys something router and an HP laptop
    > with a Centrino incantation. WEP would work, but only if I used a
    > hexadecimal key. This was before the various wireless updates from
    > Microsoft and before several Linksys firmware updates, so I'm fairly
    > sure it was fixed long ago. However, it just proves that it can
    > happen.

    Yup, it can. But if there were major problems, I would think I would have
    run into them. I've used my AG650 against a bunch of different routers using
    WEP ASCII key entry. I'll bet the config utilities for G650 and AG650 are
    nearly identical, except for the 802.11a-related stuff.

    Of course, Netgear released a USB router that wouldn't remember WEP
    configurations across reboot. Which proves that these vendors don't invest a
    lot of money in testing.

    >
    > >Their coders are at least as likely to get display-hex/binary conversion
    > >wrong.
    >
    >
    > --
    > # Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
    > # 831.336.2558 voice http://www.LearnByDestroying.com
    > # jeffl@comix.santa-cruz.ca.us
    > # 831.421.6491 digital_pager jeffl@cruzio.com AE6KS
  18. Archived from groups: alt.internet.wireless (More info?)

    In article <g4fef09jenhsi2lgopjsa63nlma1s2quhr@4ax.com>,
    Jeff Liebermann <jeffl@comix.santa-cruz.ca.us> wrote:
    >
    > Stolen from the PracticallyNetworked.com web pile:
    >
    > One ASCII Character is Eight (8) Bits
    > One HEX Character is Four (4) Bits
    > 40 or 64 bit ASCII WEP code has 5 characters
    > 40 or 64 bit HEX WEP code has 10 characters
    > 128 bit ASCII WEP code has 13 characters
    > 128 bit HEX WEP code has 26 characters

    Well yeah, but if you're limited to only the ASCII characters you can type,
    then that's around a hundred characters -- about 6.6 bits' worth (instead of
    8). Hex doesn't limit you at all.

    --
    -eben ebQenW1@EtaRmpTabYayU.rIr.OcoPm home.tampabay.rr.com/hactar

    "You're one of those condescending Unix computer users!"
    "Here's a nickel, kid. Get yourself a better computer" - Dilbert.
Ask a new question

Read More

WEP Internet Wireless Networking