But I don't see how this IS a security issue. If the attacker has root access, then you have bigger problems. If the attacker mounts another filesystem over an important immutable filesystem, then he will be working with privileges he ALREADY HAD on files which are NOT the immutable files. Temporarily he may have the appearance of having modified those files, but it is just a thin facade. Unmount or reboot and the original, unaltered files remain.

So by doing this the attacker can do more? If he is root to start with then you are kind of screwed anyway. He could have changed non-immutable files regardless of this so called "vulnerability" and at the same time he is still not be able to change any immutable files.

So where is the risk here?

Theo might seem coarse, but he has a proven track record and I think people should wait to hear his rationale before judging him. I have seen lots of non-issues get brought up against OpenBSD and it must get pretty tiring to hear people complain about them.


[ reply ]

Link to this comment: http://www.securityfocus.com/comments/columns/380/32984#32984