🚀 go-pugleaf

filter@firthcom.demon.co.uk (Steve Firth) wrote:

>That's one route. Others include exploiting known weaknesses such as
>program linking, printer sharing, oddball Java weaknesses. Basically if
>they can launch or control an application on your machine (somehow) then
>they have a route in.

It's the 'somehow' bit I'm trying to understand. I appreciate that it
wouldn't be wise to get too specific in a public forum, but how could
someone even begin an attack without knowledge of the IP addresses?

Feel free to use email if you feel that to be safer.

>How about taking it as
>read that connectign a system to the net without a firewall is a
>security risk. How big a risk depends upon the sensitivity of the
>information that you store and the financial risk if someone should
>happen to get in and (say) trash all the info on your system.

Trashing info isn't really a concern since we have automated backup s/w
which does a complete backup of each machine each evening just prior to
shutdown (except for the PowerBooks which aren't permanently connected
anyway and which have SyQuest drives for manual backups). But most of our
work is commercially sensitive in that it would have value to a competitor
of one of our clients. I'm thus trying to make some kind of realistic
assessment of the risk.

Ben

-- 
Ben Lovejoy             | For special addresses
ben@lovejoy.demon.co.uk | (auto-responder, etc):
www.lovejoy.demon.co.uk | autoresponder@lovejoy.demon.co.uk