This one time, at band camp, [EMAIL PROTECTED] wrote: >Someone wrote... > >> >>And totally unimplementable on a machine where the same binaries can have >> >>different MD5 sums across different installations, e.g. the one you all are >> >>(most likely) reading this mail on now. > >Why whould they be different? I guess I'm sorta asking >what do you mean by installation? Distributions? Versions?
I should have elaborated: Assuming you build some of your software from source, then you can't have a vendor-supplied tripwire-like firewall that has a hardcoded list of checksums. Assuming. But real users just suck down packages from their nearest mirror (near being the USA in the case of up2date and Red Hat (you current users quiet down) :-) so I guess a commercial Linux vendor could in fact start distributing a hardcoded checksum database. Of course then you get into the issue of trust... You certainly wouldn't see anything like this implemented on Debian testing or unstable... and most likely no-one could be bothered. tripwire, aide, osiris, and samhain are all packaged. -- [EMAIL PROTECTED] http://spacepants.org/jaq.gpg -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
