Re: http://klik.berlios.de/

Marcin Pawlik <[email protected]>
Newsgroups gmane.comp.kde.debian
Message-ID <20040202155354.GA3344@mpmain>
On Mon, Feb 02 at 02:21, Kurt Pfeifle wrote:
> Hi, guys,
> 
> any of you running Knoppix should have a look at this:
> 
>     ˙http://klik.berlios.de/

[...]

> Really, really cool stuff...

Yes, looks useful. And dangerous. If I understand it correctly this is
just a client side command execution protocol, without any control,
sandboxing etc. Unfortunately I don't have time to dig deeper. I'm sure
the author is just preparing a nice application and I'm sorry to be so
suspicious but I'm really paranoid about such tools. It'd be nice if
someone could check it. AFAIK there are no sources, protocol
specification and:

#v+
[...]

munmap(0x40001000, 54083)               = 0
vfork(Process 3309 attached (waiting for parent)
Process 3309 resumed (parent 3308 ready)
)                                 = 3309
[pid  3309] --- SIGSTOP (Stopped (signal)) @ 0 (0) ---
[pid  3308] waitpid(3309, Process 3308 suspended
 <unfinished ...>
[pid  3309] getppid()                   = 3308
[pid  3309] close(0)                    = 0
[pid  3309] open("/proc/3308/as", O_RDWR|O_EXCL) = -1 ENOENT (No such file or directory)
[pid  3309] ptrace(PTRACE_ATTACH, 3308, 0, 0) = -1 EPERM (Operation not permitted)
[pid  3309] write(2, ".klik is being traced!\n", 23.klik is being traced!
) = 23
[pid  3309] kill(3308, SIGKILL)         = 0
[pid  3309] exit_group(0)               = ?
Process 3308 resumed
Process 3309 detached
+++ killed by SIGKILL +++
#v-

I'm not a security specialist but it looks like klik is trying to
protect itself from being traced by forking/parent tracing trick. 
Am I right?

Regards,

-- 
Marcin Pawlik
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.