mailing list archives
Re: [NSE] Adding mkdir support
From: David Fifield <david () bamsoftware com>
Date: Thu, 5 Apr 2012 09:52:34 -0700
On Wed, Apr 04, 2012 at 08:55:50PM +0100, Djalal Harouni wrote:
On Sat, Mar 31, 2012 at 10:20:09PM +0200, Patrik Karlsson wrote:
Your comment got me thinking and while working on the luadoc documentation
I went through all of the functionality the fs library provides. To be
honest, I don't think it should be a problem including any of the functions.
Ok, I took a quick look, and I think that we should modify the following:
make_dir() function calls mkdir() in this way:
mkdir (path, S_IRUSR | S_IWUSR | S_IXUSR | S_IRGRP |
S_IWGRP | S_IXGRP | S_IROTH | S_IXOTH );
At least we should use 0755 regardless of the current value of umask
and avoid situations where groups are shared...
1) use 0700 mode ? (I like this one).
2) add an extra argument to control the mode ?
I think option (1) is the right one.
BTW while we are at this thing, IMHO we should add checks to only write
to directories owned by the current user and writable only by the current
user, and try to avoid issues that can raise from following crafted urls
and links from servers. Perhaps we should define something like "/" root
directory and write all the stuff there in case...
These kinds of checks can be difficult to get right but I agree that we
can make a best effort here. It's possible to imagine attacks like
creating a directory with a guessable name in /tmp that another program
is using as a lock.
I think we can first adopt the mkdir function and later add these
checks. I like the idea of having a sandbox directory that contains all
file write operations done by NSE. But that's likely to break some
things at this point and we would need a proposal as to how it would
Sent through the nmap-dev mailing list
Archived at http://seclists.org/nmap-dev/