Possible (temp) solution for Symlink/Core Security Hole

From: William Flett <will_at_dcs.rhbnc.ac.uk>
Date: Tue, 21 Apr 1998 11:34:02 +0100 (BST)

 Dear All,

 Richard L Jackson Jr <rjackson_at_osf1.gmu.edu> sent a nice summary for the
 symlink attack patch status on DU4.0x but another BUGTRAQ suggestion
 (sorry I can't find the original message) appears to give a good temporary
 fix against this hole.

 Simply remove the world read permission from the offending SUID programs.
 This seems to have the effect of stopping anyone forcing a core dump of
 those nasty priveleged programs without any obvious bad side effects.

 PLEASE DON'T rush off and do this until we have a concensus as to whether
 this is really a harmless way of blocking this problem. It certainly
 looks better than completly disabling core dump via the kernel.

 I'll send back a prompt summary when the votes are in.

 All the best,

 -Will

Computer Science Department | mail: W.Flett_at_dcs.rhbnc.ac.uk
Royal Holloway, University of London | talk: +44 1784 443428 (direct)
Egham, Surrey TW20 0EX, England | fax: +44 1784 439786
Received on Tue Apr 21 1998 - 12:35:56 NZST

This archive was generated by hypermail 2.4.0 : Wed Nov 08 2023 - 11:53:37 NZDT