How to tell a fake SSL certificate from a real one

How to tell a fake SSL certificate from a real one

Secure Home | Search | About
 General Computer Security    Post an article   get this group's latest topics as an RSS feed add this group's latest topics to your My MSN content add this group's latest topics to your My Yahoo content add this group's latest topics to your Google content
Subject Author Date
How to tell a fake SSL certificate from a real one Joan Battaglia 10-27-2007
Posted by Nomen Nescio on October 31, 2007, 3:00 pm
If you were  Registered and logged in, you could reply and use other advanced thread options
DevilsPGD wrote:

>
> >The SSL certificate used by your router *must* be registered to the
> >router manufacturer, not some arbitrary IP address. This is a perfect
> >case of a valid certificate generating the warnings it's suppose to
> >generate because it's used in an implementation where it's impossible
> >to do it any other way and still maintain a secure connection to your
> >router. Linksys has no way of knowing that what your router's
> >internal IP address will be.
>
> Not at all. A self-signed certificate generated by the router with a
> valid hostname would be best, since the user can simply install the
> certificate once and never have warnings again.

That's the problem. Routers don't have valid host names from where
you need to access them from. Internally they're straight IP addresses.
Your suggestion would mean that every router would have to have a
public facing administrative interface (and no private administrative
interface).

Noooooooo thankyou. ;)

> Linksys simply cannot get a certificate registered for "192.168.0.1"
> even if they wanted one.
>
> Another option though, and this would be fairly trivial to implement
> if Linksys (and others) were actually serious about security.
> Purchase a certificate for "router.linksys.com", on the internet
> point that IP to 192.168.0.1 (or whatever is the standard for routers
> that use that default) and then have the router's DNS server override
> that mapping and supply the valid IP.

First of all, there is no standard. There's defaults, they vary from
router to router, and they can easily be changed (in many cases they
have to be).

Second of all this is flatly impossible. the 192.168.x.x IP block is
what's known as "non-routable". Those addresses can't be resolved from
the outside under any circumstances. If they could, every router still
set at defaults would conflict with every other similar router.

> This would work for all browsers and all networks where the network
> uses the router as a DHCP server, and the router passes DNS up to the
> ISP (as is the case with most consumer grade routers)
>
> As it is, they distribute a certificate which isn't valid, thereby

That's not true at all. The certificate is perfectly valid. And it's
trivial to decide whether or not to accept it in spite of the "doesn't
match" error *because* it's coming from a non-routable address. You're
manually entering the address, and that address can't exist anywhere
outside your local network (from your perspective). It's a sure bet the
certificate is kosher.

> adding virtually no security, and worse, training users to ignore
> certificate errors.

No security??

On the contrary, it's very hard security. Local networks are easier to
sniff than the Internet proper. For a home user with a single machine it
may not matter, but anything more complex than that and the SSL
encrypted connection is in some ways more critical than SSL connections
to outside servers. You're talking about the boundary equipment that
controls how traffic flows into, out of, and through the local net. Own
that, and you own every machine on that local net.

> I consider that a negative in the grand scheme of things.

I don't think you really understand what a router looks like on a
network. And you obviously don't understand non-routable IP
addresses. ;) There's no negative or positive about it really, it's
just the way things must be if you want a secure connection to your
router. It's silly to think that Linksys or anyone else should have to
have a different certificate for every piece of equipment they produce,
and it can't be done anyway because by default every model/series/class
of equipment is the same.

It simply just can't work that way.... :(


Posted by Anonymous Sender on October 27, 2007, 8:51 pm
If you were  Registered and logged in, you could reply and use other advanced thread options
Joan Battaglia wrote:

> What does a forged SSL situation look like to the user logging into email?
> Do you have an example?
>
> I read with interest all the help kindly provided by the likes of helpful
> folks like VanguardLH & mark carter & others - which basically concluded
> Tor compromised mail login passwords under both circumstances

The problem is that their conclusions are not only unwarranted
technically, the results of the incident they're all talking about
whether they even realize it or not specifically state that all these
passwords were sniffed from connections that *were not* SSL/TLS
encrypted.

http://www.lightbluetouchpaper.org/2007/09/10/embassy-email-accounts-breached-by-unencrypted-passwords/

The only mystery here is why government agencies are not mandating
secure email access. It's astounding that any IT people in charge of
embassy communications could be so utterly clueless.

> - http (the Tor gets your mail password in the clear)
> - https (the Tor _could_ impersonate the "certificate")

Not unless your browser is horribly broken, or you blindly click the
'OK' buttons on dialogs without reading the clear and concise warnings
contained therein.

>
> So, I ask ... how can I tell if a certificate is impersonated by a rogue
> Tor? I routinely say yes to all certificate requests because I never
> understood them. Now I will take the time to read them.

You read the error message.

>
> But, what does a fake SSL situation look like?

A big pop up with a suitable warning symbol, and likely some sort of
"broken lock" notification in your status bar, all telling you that the
certificate doesn't match the site you're visiting, isn't signed, or
has changed.

> For example, I just initiated a connection to my legimate router:
> https://192.168.1.1
> And it said (as it always does):
> "Security Error: Domain Name Mismatch"
> It went on:
> "You have attempted to establish a connection with 192.168.1.1.

Exactly. The certificate was issued under the name of your router
manufacturer and you've entered a URL that doesn't match that name.
This is just one of the warning signs. Since you manually typed in a
non-routable IP address and know what it is you're trying to connect
to, you can ignore this particular warning. If you had been trying to
reach www.mybank.com ignoring it would just as obviously be a
BadThing(tm).

> Obviously this whole situation is a false alarm.
>
> Does anyone have an example of a situation we can go to in order to see
> what a "real" SSL forgery looks like to the user as they try to log into
> their email web site?

It will look exactly the same if this is how the forgers are trying to
attack you. Except the names will have changed of course... "You have
attempetd to establish a connection with www.mybank.com, how ever the
certificate belongs to XXXX". Or it will be unsigned, or won't match
the cert you've received on previous visits to your bank site. Or more
likely a combination of all three.


Posted by Aaron on October 28, 2007, 10:58 am
If you were  Registered and logged in, you could reply and use other advanced thread options


>> Does anyone have an example of a situation we can go to in order to
>> see what a "real" SSL forgery looks like to the user as they try to
>> log into their email web site?
>
> It will look exactly the same if this is how the forgers are trying to
> attack you. Except the names will have changed of course... "You have
> attempetd to establish a connection with www.mybank.com, how ever the
> certificate belongs to XXXX". Or it will be unsigned, or won't match
> the cert you've received on previous visits to your bank site. Or more
> likely a combination of all three.

Seems to be the bottom line here.

I thought I basically understood how SSL works, but i guess it can be
really confusing.

I don't know about all this ssl intercepting thingies, but i used to have a
setup involving a local proxy, proxomitron handling https as well. I had to
accept a local (self-signed???) cert from proximitron (that i downloaded)
before it could work.

I presume anyone in the TOR chain that tried to do so, would cause the same
thing?

Posted by Anonymous on October 28, 2007, 10:58 pm
If you were  Registered and logged in, you could reply and use other advanced thread options
Aaron wrote:

>
>
> >> Does anyone have an example of a situation we can go to in order to
> >> see what a "real" SSL forgery looks like to the user as they try to
> >> log into their email web site?
> >
> > It will look exactly the same if this is how the forgers are trying to
> > attack you. Except the names will have changed of course... "You have
> > attempetd to establish a connection with www.mybank.com, how ever the
> > certificate belongs to XXXX". Or it will be unsigned, or won't match
> > the cert you've received on previous visits to your bank site. Or more
> > likely a combination of all three.
>
> Seems to be the bottom line here.
>
> I thought I basically understood how SSL works, but i guess it can be
> really confusing.

It can be made confusing anyway. ;-)

The underlying principals and actions you should take are fairly
straightforward. If you get an error *read it*. If you don't understand
it, stop. Only when you've figured it out should you continue.

> I don't know about all this ssl intercepting thingies, but i used to have a
> setup involving a local proxy, proxomitron handling https as well. I had to
> accept a local (self-signed???) cert from proximitron (that i downloaded)
> before it could work.
>
> I presume anyone in the TOR chain that tried to do so, would cause the same
> thing?

Yes, that's essentially what an evil Tor node attempts and the same
sort of error you'll get. The wording may be different because there's
different errors, different browsers will represent them in their own
"language", and I don't remember what the specific problem with the
Proximitron cert was, but the principals are the same. Something or
things won't "jive". For evil Tor nodes and other MITM attackers, even
ones with certs signed by trusted authorities, it will most likely be
something akin to "The cert doesn't match the site you're connecting
to". It's not the only scenario that can generate that error, but MITM
attacks will almost always generate them.







Posted by bealoid on October 27, 2007, 9:18 pm
If you were  Registered and logged in, you could reply and use other advanced thread options

[snip]

> I routinely say yes to all certificate requests because I
> never understood them.

Unfortunatly you're not the only person to do so. :-( Other people will
also click [YES] when some malwareis asking to be installed.

> Now I will take the time to read them.

Good Luck.

Similar ThreadsPosted
Howto setup a certificate authority and create a signed certificate using openssl on Debian sarge March 16, 2005, 10:39 am
What are the real dangers of shared hosting ? May 8, 2004, 7:25 am
Have real exploits of arithmetic overflows happened? February 13, 2007, 12:45 pm
Re: It's a fake terrorist scare, folks August 24, 2006, 6:05 pm
Re: It's a fake terrorist scare, folks August 15, 2006, 2:24 am
Re: It's a fake terrorist scare, folks August 16, 2006, 2:14 am
Re: It's a fake terrorist scare, folks August 18, 2006, 2:22 am
Re: It's a fake terrorist scare, folks August 18, 2006, 2:25 am
Re: It's a fake terrorist scare, folks August 20, 2006, 7:25 am
Re: It's a fake terrorist scare, folks August 24, 2006, 4:20 pm

The site map in XML format XML site map

Contact Us | Privacy Policy