On Mon, Jul 20 2009, Ron Johnson wrote: > On 2009-07-20 21:29, Manoj Srivastava wrote: >> On Sun, Jul 19 2009, Ron Johnson wrote: >> >>> On 2009-07-08 20:23, Miles Bader wrote: >>> [snip] >>>> Hmm, my / is 290MB, though /tmp, /var, /boot, and /usr are all separate >>>> partitions. >>> *Why*? IOW, what benefit do you derive in 2009 (as opposed to 1989, >>> when disks weren't always large enough to hold it all) from splitting >>> these out? >> >> Security? >> >> /dev/sdb2 / ext3 >> noatime,errors=remount-ro 0 1 > > Why device names instead of labels or UUIDs?
*Shrug*. Been a while, and it has been working form me for years. Why change? >> /dev/sda1 /boot ext3 >> noatime,rw,defaults,noauto 0 2 > > noauto????? Who the hell wants the braindead initramfs mucking around with a working boot system? It also ensures that I have to be actively thinking about modifying my boot process before changes happen. >> /dev/mapper/anzu_main-usr_lv /usr ext3 >> noatime,ro,defaults 0 2 > > I understand why this is ro; why then is /boot rw? Cause it is never mounted. >> /dev/mapper/anzu_main-home_lv /home ext3 >> noatime,rw,nosuid,nodev 0 2 > > What does nodev mean? (My google fu must be lacking.) Is "Do not > interpret character or block special devices on the file system." just > extra security so that a rogue app doesn't try to create a device file > anywhere but /dev? So no one can create a device or a block char file elsewhere in the file system, yes. >> /dev/mapper/anzu_main-ulocal_lv /usr/local ext3 >> noatime,rw,nosuid,nodev 0 2 >> /dev/mapper/anzu_main-var_lv /var ext3 >> noatime,rw,nosuid 0 2 >> /dev/mapper/anzu_main-spool_lv /var/spool ext3 >> noatime,rw,nosuid,nodev 0 2 > > Seems to me that this whole exercise is to ensure that /dev is in it's > own partition. Layered security is always better than waiting for the silver bullet all secure mechanism. It is all about increasing the work factor for Mallory. manoj -- All is well that ends well. John Heywood Manoj Srivastava <sriva...@acm.org> <http://www.golden-gryphon.com/> 1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org