lkml.org 
[lkml]   [1998]   [Apr]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
Subjectforeign fs security features (was Re: NCPFS overhaul)

>> Then when they attempt to access something on the ncpfs filesystem,
>> the NCP call is made using their IPX connection. The local
>> filesystem still keeps a cache for files and directories,
>> but keeps a record of the Netware userid with the cached data,
>> and refetches the data if the Netware userid is different.
>
> It is done in UnixWare... But I'm not sure that my knowledge is
> good enough - at this time data are cached in system (AFAIK)
> without owning user, only path is relevant. That implies that
> directory contents must be same for all users on that mountpoint.
> And that implies that this is not way to go. If someone (not me :-))
> changes it, we can talk about this again. There is major difference
> between NCP and NFS - on NFS you see same thing if you are root or
> nobody, but you see different things over NCP.

The "per-user hidden file" concept is very important.
The UFS filesystem is going to need it for Digital Unix
filesystems with multi-level secure directories enabled.
Novell Netware support needs it. I'd guess most "Trusted *"
systems have the concept.

Digital Unix puts security information inside the directory.
From what I can tell, older versions of the OS should be able
to mount such filesystems read-only. Security info goes after
the filename. I think it was 16 bytes.

> And if we are talking about per-user mounts, there is NUC protocol
> (undocumented) which allows full access to Netware volumes from Unix
> (with user IDs, devices, links, symlinks). If someone is interested
> in reverse engineering, it is NCP 0x5F :-) (Keep UnixWare handy).

If you can fit into one of these catagories:
(1) NONCOMMERCIAL EDUCATIONAL PURPOSES
(2) NONCOMMERCIAL PERSONAL HOME USE (SUCH AS PERSONAL WEBSITE AND BBS)
you can get UnixWare for the cost of media.
http://www.sco.com/offers/license_products.html

Another interesting thing about Netware: directory traversal can
grant special rights. For example, the Supervisory right on /usr
would give you root-like power over /usr/local/bin even if you were
excluded from /usr/local.

The SMB protocol and NTFS are interesting too. I think security in the
kernel itself was done rather well -- too bad the default NT install
leaves the system wide open! At least the core is not rotten.

Access is specified with a 32-bit mask containing the following bits:

16 DELETE delete
17 READ_CONTROL read the owner, group, and ACL
18 WRITE_DAC write the ACL
19 WRITE_OWNER write the owner (take ownership?)
29 GENERIC_EXECUTE gets mapped to specific rights
30 GENERIC_WRITE gets mapped to specific rights
31 GENERIC_READ gets mapped to specific rights

Those last 3 are special. The OS maps them to object-specific
rights (bits 0 .. 15) as needed. For example, bit 2 can mean:

create sub-key (registry)
append data (normal file)
add sub-dir (directory)
create pipe instance (named pipe)

Normal apps don't worry about object-specific rights. This kind of
system seems to make it easy for the OS to emulate foreign security
systems. I think NT has this system so that it can simultaneously
meet POSIX requirements and act (almost?) like Netware. Since Linux
supports so many foreign filesystems, it might be a good idea to
operate like this.

-
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 13:42    [W:0.060 / U:0.796 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site