Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!apple!olivea!tymix!cirrusl!sunstorm!dhesi From: dhesi%cirrusl@oliveb.ATC.olivetti.com (Rahul Dhesi) Newsgroups: comp.unix.internals Subject: Re: non-superuser chown(2)s considered harmful Message-ID: <2789@cirrusl.UUCP> Date: 10 Dec 90 02:05:22 GMT References: <18786@rpp386.cactus.org> <1990Dec7.171501.18028@mp.cs.niu.edu> <18792@rpp386.cactus.org> <1990Dec8.184047.22221@mp.cs.niu.edu> Sender: news@cirrusl.UUCP Organization: Cirrus Logic Inc. Lines: 33 Apart from the many detailed differences between System V (Release 3) and 4.xBSD, there is a fundamental difference in philosophy that some of you may be missing. The design of System V assumes a small, isolated system system with few users and any networking limited to UUCP. The design of BSD assumes a bigger system with many users and connectivity between machines using (usually) TCP/IP on high-speed links. On the sort of small, isolated system for which System V is designed, the ability to chown files away without restriction, and the absence of any disk quota or inode quota mechanism, does not cause a serious problem. However, consider a UNIX system with 1,000 or more users. In the absence of reasonable resource quotas, the administrator would have time to do little else but track down culprits involved in deliberate or accidental denial-of-service incidents. Hence the greater restrictions such as a restricted chown, as well as more powerful and flexible access control mechanisms such as disk quotas, multiple group memberships, and soft and hard limits on memory and CPU usage. Those of you who use only BSD-derived systems shouldn't try too hard to understand the System V user's point of view -- it will be quite strange to you, almost like trying to understand the unusual practices of a foreign country. Similarly, those of you who use only System V- derived systems should be prepared to accept what BSD users say without understanding just why they say it. -- Rahul Dhesi UUCP: oliveb!cirrusl!dhesi