::::: Welcome to Web Series Today "collaborative environment"
::::: Help get LGPedia back online!!!!! <== ACT TODAY!!!!!
::::: Welcome to new Web Series Today blog "authors". <= PLEASE READ CAREFULLY!!!


Sunday, December 16, 2007

Re: Assignment - Saffryn



Ask and ye shall receive.

---------------------------------

This is the assignment given by WatchingOpAphid here

Posted by Anonymous at 6:37 AM

Labels:

2 comments:

  1. QtheCDecember 16, 2007 at 5:25 PM

    Interesting video. It consists of a secure shell login to a remote IP address, executing a command to compare words in a dictionary to bytes in a file named tcpdump_captured.mpg , encoding the words found using md5 checksums. Then the .mpg files is zipped, some log files are removed (to hide tracks I suppose), the connection is closed, and the result file is grabbed with wget unzipped and mailed. That file does not appear to be accessible at the moment.

    So, the most interesting line seems to be this one:

    :~$ cat /usr/share/dict/words |
    while read r; do if xdc -p $r -f
    tcpdump_captured.mpg -b 10240 :
    then echo $r; exit 1; fi; done | md5

    Looking at the output, it seems to be repeating a pattern of three 128 bit md5 checksums:

    8c3c1a60b5c10f3faacd15ae3b1db6fb
    56bf8e88ec14fe46644db0257107616b
    d73b04b0e696b0945283defa3eee4538

    ... leading me to think that three words from the list in /usr/share/dict/words were found in tcpdump_captured.mpg

    Unfortunately, I don't know exactly what the xdc script does - that does not appear to be standard linux or unix. But it doesn't matter.

    Anyone with a unix shell should be able to recover the three matching words by doing something like this:

    : cat /usr/share/dict/words | while read r; do echo $r; echo $r | md5; done > tempfile

    tempfile now contains a list of all the words interleaved with the md5 checksums for each word

    Then just search through tempfile for the three md5 checksums we are looking for one at a time. The keystrokes to do this are:

    : less tempfile
    then
    /8c3c1a60b5c10f3faacd15ae3b1db6fb
    then hit the up arrow once to find the matching word

    repeat for
    /56bf8e88ec14fe46644db0257107616b
    /d73b04b0e696b0945283defa3eee4538

    Anyone want to try this and report the results? (note this is bourne shell aka sh syntax)


    (side note: it is not 'computationally feasible' to generate the origina words directly from each mp5 checksum value)

    ReplyDelete
  2. QtheCDecember 16, 2007 at 5:30 PM

    Let' see, a couple of minor notes. You may have to look around a little on any given Unix or Linux system to find the words list.

    It might be in /usr/dict/words
    or /usr/dict/linux.words
    for instance.

    once you are in "less" just hit "q" to quit out of it.

    ReplyDelete

Subscribe to: Post Comments (Atom)



If you want to become an "author" on Web Series Today please read: http://tinyurl.com/becomeaWSTauthor

For more detailed information about Web Series Today please read: Web Series Today:

For other info contact: [email protected]


Join the discussion: http://www.tinyurl.com/webseriescommunity