Messages in this thread | | | Date | Mon, 29 Dec 1997 15:31:43 -0500 (EST) | From | Jon Lewis <> | Subject | Re: hardlinks.... sucks... ;-( |
| |
On Mon, 29 Dec 1997, Rob Hagopian wrote:
> has claimed the prize money for cracking it... the symlink patch also > should not effect 99.44% of programs (I have yet to hear of a program that > needs to make a link in /tmp [hard or soft] to a file it doesn't own).
I finally did run into a situation where the symlink patch caused some trouble, though the situation was resolvable. Here's a hypothetical situation similar to the one I ran into.
Say we have a directory, /home/html/testing, mode 1775, owned by root.wwwadm. The dir is sticky because several "admins" have write access to it, but root doesn't want them stepping on each others creations. Several people in group wwwadm make symlinks from this directory to other directories. Kernel is upgraded to one with the stack+symlink patches, suddenly the symlinks above do not work unless chown'd to root.
Maybe that sort of situation was fairly unique, but I think it demonstrates, as others have suggested, that it would be nice if there were some way via /proc to enable|disable the symlink security features on a directory by directory basis.
> These should dramaticly decrease the security holes due to bad user-space > programming and so I really don't see why they should be excluded from the
In my case, I decided the advantages outweighed the disadvantages, and have not considered removing the patches. I doubt I'll ever willingly compile kernels (at least for my own use) without these patches again.
------------------------------------------------------------------ Jon Lewis <jlewis@fdt.net> | Unsolicited commercial e-mail will Network Administrator | be proof-read for $199/message. Florida Digital Turnpike | ______http://inorganic5.fdt.net/~jlewis/pgp for PGP public key____
| |