Messages in this thread | | | Date | Tue, 18 May 1999 10:04:50 +0200 (CEST) | From | Ernst Jan Plugge <> | Subject | Re: Capabilities done right [diff against 2.3.1] |
| |
On Mon, 17 May 1999, Pavel Machek wrote:
> Yes, it is. Think about: (let ping be shell script)
[classic interpreter exploit snipped]
Wasn't this fixed for Linux a _long_ time ago?
Perl for example refuses to run a suid script unless it has been told at compilation time that the OS has closed this hole. Red Hat's perl sure doesn't complain when I run a suid perl script (not that I use them, of course).
The usual fix, I believe, is to not pass the original filename as $1, but a pipe that is connected to the original fd that was created when the kernel opened the script to find out the interpreter's name.
Y.T.,
Ernst Jan Plugge - rmc@dds.nl --
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |