lkml.org 
[lkml]   [1998]   [Jun]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: Secure-linux and standard kernel
Date
From
On Thu, June 25 1998, Chris Evans <chris@ferret.lmh.ox.ac.uk> wrote:
|
|On Thu, 25 Jun 1998, MOLNAR Ingo wrote:
|
|> if 'ping' is setuid root and the first thing say ping.c does is:
|>
|> drop_cap(ALL_BUT_RAWSOCKET);
|
|Yes this is good isn't it. Note that traceroute's a better example; ping
|for linux already does:
|
|main()
|fd = socket(SOCK_RAW ...)
|setuid(getuid());
|
|
|
|rlogin, rsh, dump, restore, even lpr, lprm, lpq, etc. all benefit a
|similar way with
|
|drop_cap(ALL_BUT_BIND_TO_LOW_PORT);

I really don't like this idea. It's certainly easier to implement than the
"nice" way, and maybe its good in the mean time, but it would require:

1. maintenance of separate source code for linux binaries (or asking the
"upstream maintainers" to insert this code)

2. reliance on the binary for behaving well - what if someone breaks in and
changes the binary not to call this?

3. the admin can't control things in an administrative way

I think a more "generic" solution would be to attach access control to
sockets, e.g. have /dev/tcp/1-65535, /dev/raw/, /dev/udp, /dev/icmp
(?) as special device files (only creat inodes for sockets with
special access control) and set the owner/group/premissions like any
other file. If such a file exists then use its permissions, if not
then resort back to the standard method (root-only for raw and TCP/UDP
sockets less than 1024, open for everyone else (?) on other sockets)

You can even link the same i-node with all ports with identical access
permissions to save inode consumption.

What do you think?

Cheers,

--Amos

--Amos Shapira | "Of course Australia was marked for
133 Shlomo Ben-Yosef st. | glory, for its people had been chosen
Jerusalem 93 805 | by the finest judges in England."
ISRAEL amos@gezernet.co.il | -- Anonymous
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu

\
 
 \ /
  Last update: 2005-03-22 13:43    [from the cache]
©2003-2011 Jasper Spaans