Hey, Sorry about the email from yesterday...I just totally lost track of time. For setgid files it is the same for setuid. It will depend on the use by the users. The only sgid files I normally run are: man, slocate, and lockfile. Libsafe can be found here: http://www.research.avayalabs.com/project/libsafe/ Libsafe intercepts calls made to libraries known to have buffer overflow problems. If it detects a problem then the executing commands will be dropped and logged to syslog along with the uid of user that ran the program. Thanks to Italo for explaining suidperl and sperl. I appreciate that. BTW, anybody know what utempter is used for and how it is used? On Wed, 9 Jan 2002, BUNTER MATTHEW wrote: > --- Reçu de RVIDOI.BUNTERMA 04 72 96 57 77 09/01/02 09.37 > > All, > > Just joined yesterday so apologies if I am asking something that > has been covered recently. > > Trying to add a setuid/setgid section to a Linux security > standard. I would like some opinions as to which files can be left > with setuid and setgid and which should definitely NOT be left > setuid or setgid. > > I have been having a good crawl around the net for a while and can > find various links on how to identify and edit these types of > files but not which ones should be altered or left alone. I > already have the Solaris recommendations. > > This will have to cover both server and workstation > implementations. > > Thanks in advance, > > Matt > > ---- 09/01/02 09.37 ---- Envoyé à --------------------------- > -> SECURITY-DISCUSS(a)LINUXSECURITY.COM > ------------------------------------------------------------------------ > To unsubscribe email security-discuss-request@linuxsecurity.com > with "unsubscribe" in the subject of the message. > -- duane -- GnuPG Public Key: http://sukkha.homeip.net/pgp.html -- Fun reading: 8-) http://linuxtoday.com/search.php3?author=Duane:Dunston ------------------------------------------------------------------------ To unsubscribe email security-discuss-request@linuxsecurity.com with "unsubscribe" in the subject of the message.
