Menu

#2 bad arguments to byte_test

open-fixed
None
5
2005-01-13
2004-03-03
Anonymous
No

Jason, I appreciate you picking up snortcenter. I also
have tried to contact the author, without success.
I ran into the issue on rule 2379 for example where I get
a byte_test error. I had this with the previous version of
snortcenter and had to make some modifications to
bypass this error. I would be willing to supply these
modifications if you are interested. Basically it consists
of adding a sequence number to the content table so
that the content rules are maintained in sequence, then
removing the text "byte_jump" and "byte_test" in the
respective fields. I have reviewed a number of rules with
this modification, and it appears to keep the proper
syntax. This change could also likely be applied to the
uri-content table, or even better, to merge content and
uri-content table as I don't see a lot of difference in
them.

Dave Fennell
dfennell@talisman-energy.com

Discussion

  • Jason Alexander

    Jason Alexander - 2004-03-03
    • assigned_to: nobody --> jaalex
     
  • Jason Alexander

    Jason Alexander - 2004-03-03

    Logged In: YES
    user_id=984574

    I'll take a look at that rule tonight. I thought I had the
    problems with them fixed as I was able to import all the rules
    that were in the 2.1 gzip about a week ago. If you want to
    send the patches that would be great. I'll take a look at them

    I have to say that Byte Jump and Byte Test are the worst
    parts of the rules to deal with

     
  • Jason Alexander

    Jason Alexander - 2004-03-04

    Logged In: YES
    user_id=984574

    Dave,

    Your right. Both the Byte_test and Byte_jump are totally
    horked. I thought I had them working but while the are
    making parsable rules the messing up rules. IF you have
    those patches send them on.

    Jason

     
  • Alex Butcher

    Alex Butcher - 2004-03-05

    Logged In: YES
    user_id=257968

    Does order really matter for uri-content checks? It does for
    content/byte_jump/byte_test, but I'm not sure working to
    maintain the order of uri-content checks is worth the
    effort. Also, I haven't found any rules which use more than
    one uri-content check (and presumably they'd only make sense
    if logically-ORed anyway, so order won't matter).

    Alex B.

     
  • Jason Alexander

    Jason Alexander - 2004-03-05

    Logged In: YES
    user_id=984574

    Alex,

    I didn't think it. I have some time this weekend that I
    plan on working on this. I have to see if I can look at the
    rules and determine if it matters.

     
  • Jason Alexander

    Jason Alexander - 2005-01-13
    • status: open --> open-fixed
     

Log in to post a comment.

MongoDB Logo MongoDB