|
Donation/Premium |
|
 |
|
|
|
|
|
|
|
 |
 |
| View previous topic :: View next topic |
| Author |
Message |
Scott_Hollingsworth
Sergeant
 Premium Member
 Joined: May 09, 2006 Posts: 116 Location: USA
|
Posted: Wed Jan 23, 2008 6:00 pm Post subject: You mis-understand pharming |
|
|
moike,
Pharming does not redirect a browser to another URL. Your browser will have the correct URL displayed in the address bar. But, the IP address that the DNS lookup provided is not that of the URL you see. It is the IP address chosen by the attacker so your computer connects to the attacker's server but the URL name is correct.
So the bookmark advice, while still good to follow for other reasons, will not protect against a pharming attack.
In order for the attacker to also spoof an SSL connection, then they would have to obtain a copy of the SSL certificate, which is unfortunately an impossible task but not easy. Then your browser will not complain in the least about a bad SSL certificate. Unfortunately, many users are conditioned to accept bad certificates due to poor admin practices. I am still hounding a colleage to splurge for a cert. from an established CA rather than a self-signed cert.
|
|
| Back to top |
|
 |
moike
PIRT Handler Premium Member
 Joined: May 26, 2006 Posts: 1873
|
Posted: Wed Jan 23, 2008 9:13 pm Post subject: |
|
|
| Quote: | | Pharming does not redirect a browser to another URL. |
I glossed over a step in the process: pharming would indeed start with the real URL. A quick and dirty way to obtain a valid SSL certificate would be for the fake site to redirect SSL communications to a 'look-alike-URL' SSL site and hope the user does not notice. Bookmarking the SSL page would protect in this circumstance since the first contact would fail due to wrong / no certificate.
You raise a good point in that with the prevalence of infected hosts, that it is not difficult to imagine a compromised Certificate Authority issuing a valid SSL certificate of a legitimate site to the wrong person. And bookmarking would not protect against that scenario.
|
|
| Back to top |
|
 |
Scott_Hollingsworth
Sergeant
 Premium Member
 Joined: May 09, 2006 Posts: 116 Location: USA
|
Posted: Thu Jan 24, 2008 5:58 pm Post subject: |
|
|
Now I see your train of thought. Yes, that would make sense as a possible scenario.
I also was thinking a little differently on the certificate issue as well. I was thinking along the lines of a well constructed, targeted blended threat to obtain a copy of the pharmed site's actual certificate. I realize the effort this would take and the chances are it would be detected. Also the issuing CA could then add the pilfered certificate to their CRL. But could the attacker also spoof the CA's CRL?
With the way things are evolving these days, small windows of opportunity for attackers do not seem to be providing protection to potential victims anymore.
Another approach pharming can lead to is a man-in-the-middle attack. This is not as difficult to pull off with SSL as it was thought to be.
BTW: I had intended to reply to another thread, but I inadvertently started a new topic.
|
|
| Back to top |
|
 |
moike
PIRT Handler Premium Member
 Joined: May 26, 2006 Posts: 1873
|
Posted: Thu Jan 24, 2008 7:44 pm Post subject: |
|
|
| Scott_Hollingsworth wrote: | | I was thinking along the lines of a well constructed, targeted blended threat to obtain a copy of the pharmed site's actual certificate. Also the issuing CA could then add the pilfered certificate to their CRL. |
Unfortunately, most browsers have disabled realtime CRL checking for performance reasons - in general it would require a Browser update / Windows Update cycle to block a leaked certificate.
| Scott_Hollingsworth wrote: | | But could the attacker also spoof the CA's CRL? |
If the browser uses OCSP, I assume that it would be nearly as difficult as spoofing SSL itself.
| Scott_Hollingsworth wrote: | | Another approach pharming can lead to is a man-in-the-middle attack. This is not as difficult to pull off with SSL as it was thought to be. |
Is it to the point where the SSL protocol can be defeated using MITM on a high percentage of attempts, or is it just the alarms over the weakening of SHA-1 and quantum computing on the near horizon?
|
|
| Back to top |
|
 |
|
|
|
You can post new topics in this forum You can reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You can attach files in this forum You can download files in this forum
|
Powered by phpBB © 2001 phpBB Group
|