>As far as I know, there are zero ways to attack an OS 9 Mac.

Appletalk uses plain text passwords by default, that means anyone within 
reach of your appletalk network can sniff the packets, know your 
password, and gain full access to your machine.

That is just one off the top of my head. Classic OS versions are not 100% 
immunue from compromise, it is just in the real world, they are very 
limited and unlikely to be hacked from the internet. There are ways to do 
it, and it does mostly involve running software that is vulnerable. That 
is why I said, you are safer if you are running a Classic OS. OS 9 btw 
can send AppleTalk over TCP/IP, that means, those cleartext passwords 
*could* be getting out to the internet... if you are on a cable modem, 
your neighbor can sniff the packets, if you have DSL it is a bit harder 
depending on how your ISP has you set up. At least with 8 and earlier, 
AppleTalk didn't go over IP (from a host standpoint), so unless you had 
an ISP that would route AppleTalk traffic (highly unlikely), then there 
was no way for your passwords to leave your local network.

>If you want to be "infection proof", get a Mac OS 9 
>system. It cannot be compromised by hackers. Especially if file sharing 
>is not active, it is not possible.

Never ever say something is not possible when it comes to hackers. :-)

>As far as OS X goes, how can a machine get infected/infiltrated from web 
>browsing?

The same way Windows can be infected just from browsing. JavaScript and 
viruses designed to attack the host OS. Currently, none exist for OS X... 
currently. Give it time, there will be. Plus, who is to say it will 
happen because you are web browsing. Just because you are only using the 
web, doesn't mean that is the only service that can pass on the network. 
Once you connect your computer to the internet, it is connected... ALL OF 
IT. The only thing stopping traffic from coming back in is either a 
firewall, or hopes and dreams that your computer will safely ignore all 
inbound traffic (are you sure you didn't accidentally load spyware when 
running that new cool automatic desktop pattern changer... sure, no one 
knows of spyware for OS X yet... YET. Or, depending on what software you 
are running, are you sure one or more applications hasn't opened a port 
to the internet? Maybe they didn't mean to, it was just poor 
programming... my first version of Baton Mail left the relay wide open to 
anyone... if you started Baton Mail while connected directly to the 
internet, anyone else could have used Baton Mail to relay their email 
thru your computer and YOUR email account!!! I didn't do that by design, 
I did it because by mistake, but I released a version before I realized 
the flaw and was able to patch it).

I can't stress this enough, OS X is safe right now... but ONLY right now. 
Is there a reason to spend lots of money on security products for OS X... 
no. I'm very much against bothering with Anti-Virus software. The 
problems they cause outweigh the benifits gained. But when it comes to a 
firewall, when you can pick up a hardware solution for under $30 
consistantly, and it won't need to be upgraded next year or likely 
anytime in the near future, it isn't a bad idea. It also isn't a 
manditory idea unless you plan to run Windows at some point (down the 
road it will likely become manditory for OS X as well, but right now, it 
is still mostly safe to run without it).

Remember, OS X has already had a number of security patches to fix 
vulnerabilities. Each time those were released, depending on how often 
someone ran the Software Update, it was anywhere from a few days, to 
possibly months before they patched their copy of OS X. If they were 
sitting unprotected on the internet, they were running vulnerable 
software that entire time. 

Half of OS X's security comes from obscurity. People aren't trying to 
hack it, so it isn't getting hacked. If Apple does their job right, and 
OS X becomes as popular as we all hope, then that tide will turn, and 
people will start trying to hack OS X, and as soon as they do start 
trying, you will see machines being exploited left and right. Many of 
them will be because people are behind on their updates. That is actually 
a good bit of the Windows exploits... not that Windows has no security, 
but that people don't update often enough, so when an exploit is found, 
it stays available for a long time.

OS 9 and earlier were much harder to hack because there was nothing by 
default on them to hack. That doesn't mean they were 100% immune, it just 
means no one found anything easy and commonplace enough to exploit. Had 
OS 9 and earlier been more popular, more would have been found. Now that 
they are discontinued, there is virtually no one still trying, so since 
they were near impossible to hack, they will likely stay near impossible 
to hack. 

>There are occasional security hole reports, but I did not know 
>of any which worked on end-user machines.

That doesn't mean they don't exist. The fact is, the ONLY difference 
between OS X and OS X Server is the tools the server version comes with 
(ok, and a limiter on the number of people that can connect for file 
sharing). If there is an exploit for OS X Server, then it will work with 
OS X Workstation as well. (go into the terminal, type postfix... WOAH... 
look at that, OS X Workstation has a full version of the Postfix mail 
server sitting on it! Same with a number of other server applications... 
like Apache... and ALL of them are just as vulnerable on OS X Workstation 
as they are on OS X Server, or NetBSD, or Linux... unless Apple releases 
a patch to fix an exploit when it is found, then your machine will be 
vulnerable to said exploit. Fortunatly, Apple stays on top of those 
patches, so it has yet to become an issue with OS X)

There are tons of *nix exploits out there, most of which apply to OS X 
because of its BSD core, and more are found daily. All it takes is to be 
on the wrong side of the patch curve and you have a vulnerable machine.


If you are running OS X connected to the internet via any 24x7 internet 
access, it is a good idea to be running some kind of a firewall. OS X has 
one built in that can be turned on, but hardware solutions have the added 
benefit of seperating the machine from the firewall (security basic, the 
further something is from access, the harder it is to access it... if the 
firewall and the workstation are one in the same box, then any attempts 
to hack the firewall are automatically attempts to hack the workstation, 
you increase your chance of being compromised by running both as the same 
machine). And of course, a hardware solution makes it easier to connect 
multiple other computers to the internet over the same line (still 
possible with just the one OS X machine, but then you have to depend on 
it being up and running all the time that the additional machines want to 
use the internet... plus, unless you put a 2nd ethernet card in the OS X 
machine, then you are running a far less secure version of NAT routing. 
That is, all the "protected" traffic is sitting on the exact same wires 
as the non-protected).

If you are running Windows, you are insane to NOT have a firewall between 
you and the internet.

If you are running any Classic Mac OS, a firewall is just added 
insurance. I won't dispute that it is far from needed for most users of 
classic Mac OS versions... but for the few dollars it will cost you for a 
hardware firewall/route, it is really cheap insurance, and carries the 
extra bonuses of being able to run pretty much anything you want behind 
the firewall without as much concern (wouldn't it be nice to know that 
when your friend stops by with their OS X laptop, or your idiot inlaws 
with their Windows laptop, that they can just plug into your network and 
safely get online).


So to sum up, my stance is the same. If you have a 24x7 internet 
connection, you better have a firewall if you run Windows, you should 
have one if you run OS X, and it couldn't hurt if you run OS 9 and 
earlier, but isn't overly neccessary.

-chris
<http://www.mythtech.net>

___________________________________________________________________________
To unsubscribe send a mail message with a SUBJECT line of "unsubscribe" to
<[EMAIL PROTECTED]>  or  <[EMAIL PROTECTED]>

Reply via email to