@rm@d1ll0 debug bl0cker bye bye

Hi all,
I'm happy to publish the following result reached by
one of the biggest reverser in the current scene:
faina.
Every version of the packer available up to know
simply checks for the OpenMutex API before starting as
debugger process or debugged process respectively.
So, if you want to avoid the debug blocker feature
simply check the OpenMutex calls.
Obviously it works on EVERY windows system.
And obviously it works good for targets that can run directly
without other protections (copymem etc.).
Hope this will help and will be improved (olly scripts etc.).
Cheers Z..

[taos]

uhmmmm? very interesting.... can you give us more info?...
how? where? perhaps a tut...?

It is very simple and it is all written in my previous:
the debug blckr feature simply works as folllows:
1)the program creates a mutex with CreateMutex.
2) then it checks for the mutex with OpenMutex to start as
debugger or debugged
So what you have to do is to change the return value of the OpenMutex.
....or the check after that call (tipically two calls are present).
Now suppose you simply have a program protected with the DB.
After changing the two conditional jumps, you will start with the
debugged program directly. So it will unpack itself and bla bla bla.
(I repeat it avoids only the DB protection, it there are others you have
to manipulate the code).
For the tutorial you have to ask to the author,
there is one but only written in italian and i haven't the time to
translate it.

Thanks to faina_mdc.

gimme that tutorial, and i'll translate it for ya

[ricnar456]

i don't try and seems to be good, but all (99%) of armadillos with copymem2 have nanomites and nanomites cannot run without the father process and crash.
But is very good for work with the iat, magic call, etc, is a good info.

Thanks
Ricardo Narvaja

hi rich,
nice to hear from you.
Is all as you say, i think
a good use is when you want to emulate a debugger
directly for the child.
the author used it in that way.
cheers.

z.