Date: Sat, 19 Jun 1999 16:19:00 +1000 (EST) From: Darren Reed <avalon@coombs.anu.edu.au> To: brian@CSUA.Berkeley.EDU (Brian W. Buchanan) Cc: ftobin@bigfoot.com, freebsd-security@FreeBSD.ORG Subject: Re: proposed secure-level 4 patch Message-ID: <199906190619.QAA28681@cheops.anu.edu.au> In-Reply-To: <Pine.BSF.4.05.9906182304420.70357-100000@smarter.than.nu> from "Brian W. Buchanan" at Jun 18, 99 11:05:46 pm
next in thread | previous in thread | raw e-mail | index | archive | help
In some mail from Brian W. Buchanan, sie said: > > On Sat, 19 Jun 1999, Frank Tobin wrote: > > > Okay, a good friend of mine Kris Wehner has written a patch to implement > > the proposed securelevel of 4, which would disallow the opening of > > secure ports (<1024) while in the securelevel of 4. The patch is against > > 3.2-STABLE kernel, as of within 12 hours. I'd like to hear more comments > > before I send it as a send-pr. The patch is attached. > > Kris's patch blocks binding ports <= 1024, but 1024 is not a secure port. > The last one is 1023. Sigh, this appears to be a mis-use of "securelevel". As securelevel increases, the system is supposed to be more secure - i.e. more functions are unavailable, even to root. Using a securelevel of -2 for this is `better', but it means your kernel must boot up with a securelevel of -1 (or less), init scripts change it to be >= 0 so that init raises it to (at least) 1 once they're all finished. Really, using this patch without a securelevel > 0 at run-time indicates someone who's really not all that interested in security. Sounds like a sysctl is the knob you're looking for to enable and disable this feature. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-security" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199906190619.QAA28681>