Dark Reading is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Cloud

1/26/2017
03:00 PM
John Strand
John Strand
Commentary
Connect Directly
Twitter
RSS
E-Mail vvv
100%
0%

How I Would Hack Your Network (If I Woke Up Evil)

How would an attacker target your company? Here's a first-person account of what might happen.

There's been a lot of talk about the recent hacks against the Democratic National Committee and many, many questions and arguments about who was responsible. 

There are some interesting things about this somewhat painful national conversation. First, it's widely believed that the attacks were launched by Russia. For most people, this resonates because they assume big attacks with big impacts must have been launched by big players. Attribution aside, this is just wrong. These attacks could have been successfully launched by anyone who spent an hour or two learning how to use the Social-Engineer Toolkit, available online.

Second, it shouldn't matter — at all. We must assume that advanced attackers are going to attack us. Further, we can't look at every successful attack as something that must have been mounted by an advanced nation-state actor. A few years ago, everyone was blaming China for attacks. Now, it's Russia. When we do this, it allows us to build a convenient straw man, and it becomes easy for us to brush off the attacks as though they were inevitable. Because surely, if China or Russia were behind the attacks, there is nothing anyone could have done to stop them. The attacks become a force of nature, an act of God.

But here's the thing: many of these attacks aren't advanced. Not at all. And, moreover, we should be able to defend against them.

Let's be very clear: your antivirus (AV) software won't protect you. Every year, we at Black Hills Information Security do a webcast called Sacred Cash Cow Tipping in which we bypass most of the major AV products and explain exactly how we did it. We do this because it's important for companies to understand that these points of defense, in and of themselves, aren't enough to stop a determined attacker. (The most recent video can be found here.)

So, I'm going to break down how, if I were evil, I would attack a network — possibly your network.

First, I will target your user population through phishing. This approach has been in the news quite a bit lately, because of the DNC attacks. It's interesting that many people are surprised by phishing. However, this is the same attack strategy we've been seeing for years. For most of our assessments, we find that roughly 20% to 30% of the user population will click on almost anything. Further, if we can couple our phishing attack with the information we learn from reconnaissance efforts, our probability of success goes way up. For example, if through recon we discover that one of your users is really into politics and often declares his political alliances on Twitter, Facebook, and LinkedIn, then we will use a ruse involving politics. 

That brings us to another point. The more a target posts on social media, the more we will focus on that user. People who are very into social media are more susceptible to targeted attacks. It could be that attackers have more information to work with when attacking. Or it could also be that these people feel the need for some level of affirmation. We feed that. That need makes them a greater risk to your organization.

I will also focus on external interfaces. I will password-spray your Web interfaces, your Outlook Web Access portals, your Secure Shell servers. (For more on password spraying, check out these blog posts by Beau Bullock.) This is where we use a single password (for example, Winter2017) and try that password on any user accounts we can enumerate online. Basically, I will attack things that shouldn't be exposed externally.

Next, I'll pivot as much as possible. Please check out Bloodhound and PowerShell Empire — these tools are fantastic for post exploitation, and could be the topic of a full series of articles. These tools allow an attacker to quickly identify other Windows systems and access their files and folders. This is the core goal of pivoting, using access on one system to access the resources on others.

So, How Can You Stop Me?
There has been a shift in security, and the old security fundamentals aren't effective any longer. The new security fundamentals include implementing application whitelisting, firewalls enabled down to the host level, and user behavioral analytics (UBA). UBA is exceptionally interesting because it is looking at user access patterns for indicators of compromise rather than just looking at program signatures. 

These are just some of the new things that security-minded organizations need to start implementing straight away. I understand that for many organizations, there are massive political and technical complexity challenges in play. But you must start looking into these methods right now. In fact, it's already too late — you should have started years ago. If you did, good for you. If you haven't started, get to it.

Let's summarize. First, your AV won't be a problem for me and will easily be bypassed. Second, I will phish your employees by using as much social media and reconnaissance as I can. Third, I will exploit all externally facing interfaces, portals, and servers. Finally, I will pivot as much as possible. How do you defend against me? Stop using your AV as a crutch, keep a smarter social media image (and encourage employees to do the same), implement whitelisting and firewalls, even at the host level, and UBA. Good luck.

(Note: John Strand will be giving a talk on this topic at upcoming SANS events in Scottsdale, Ariz., and Tysons Corner, Va.)

Related Content:

John Strand is a senior instructor with the SANS Institute. He teaches SEC504: Hacker Techniques, Exploits, and Incident Handling; SEC560: Network Penetration Testing and Ethical Hacking; SEC580: Metasploit Kung Fu for Enterprise Pen Testing; and SEC464: Hacker Guard: ... View Full Bio
Comment  | 
Print  | 
More Insights
Comments
Threaded  |  Newest First  |  Oldest First
nosmo_king
50%
50%
nosmo_king,
User Rank: Strategist
1/27/2017 | 9:31:53 AM
Faster and easier
I find that tossing a few carefully crafted USB sticks into the executive parking lot is easier, cheaper and more effective.
Joe Stanganelli
50%
50%
Joe Stanganelli,
User Rank: Ninja
1/28/2017 | 10:58:46 AM
Re: Faster and easier
@nosmo: Well, sure, that's phishing too -- just a more "physical" type of phishing, compared to email phishing.
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
1/30/2017 | 12:45:09 PM
Re: Faster and easier
"Well, sure, that's phishing too ..."

True, just tricking users to do something that they are not normally do. Clever.
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
1/30/2017 | 12:43:53 PM
Re: Faster and easier
Not only executive but for everyine, we will all wonder what we have in a lost and found flash drive.

 
Joe Stanganelli
50%
50%
Joe Stanganelli,
User Rank: Ninja
1/28/2017 | 11:02:25 AM
%
The 20-30% figure isn't surprising.  At an MIT event I went to not too long ago, one presenter talked about an email sent organization-wide that said something to the effect of: "This is a phishing email.  It is fake.  Do not click on this link" -- and found that 10% of the recipients STILL clicked the link.

One C-suite executive who clicked on the link's response when asked why he clicked it: "I wanted to see what would happen."
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
1/30/2017 | 12:47:05 PM
Re: %
"Do not click on this link" -- and found that 10% of the recipients STILL clicked the link. "

I wonder, the reason they would click because of the question in their mind: "why would I get a link not to click?"
Joe Stanganelli
100%
0%
Joe Stanganelli,
User Rank: Ninja
1/31/2017 | 1:00:42 PM
Re: %
@Dr.T: Perhaps the thinking was "Well, the link says not to click, but it's clearly from my own organization, so it can't be *truly* bad."

The better way to do this is to send fake phishing emails (without letting the users know what they are), and then those who click are brought to a page where they are alerted that they fell for a phishing scam -- and their computer is locked up until they complete a 5-minute InfoSec training so they don't fall for it again.  This technique has been shown to reduce successful email phishing attacks by up to 75%.
Kumzy
50%
50%
Kumzy,
User Rank: Apprentice
2/19/2017 | 8:04:34 AM
Re: %
That is a very great idea. My organization did the same thing about last week IT security sent out an e-mail that webcams had been installed in tbe breakroom refrigerators to forestal people stealing other peoples lunches a link was then provided to view the cam when you click on this link it takes you to a log in page which asks for your log in credentials, needless to say rumor started going round among staff about the webcams and people were curious about the issue so some clicked on the link and supplied their credentials and a message popped up that they have just willingly given up their credentials for a phising attack.
Petar Zivovic
50%
50%
Petar Zivovic,
User Rank: Apprentice
2/28/2017 | 2:00:30 PM
Re: Phishing emails (responding to Joe's comment)
Re: sending fake phishing emails to train: Exactly right. It's effective when done on an ongoing basis (at least once a month.) I have seen the susceptibility rate fall to below 1% and stay there, which I consider phenomenal - until I forgot to do it one month. Then next month, it went up to 16%.

As Jefferson said, the Price of Freedom is Eternal Vigilance.
Joe Stanganelli
50%
50%
Joe Stanganelli,
User Rank: Ninja
3/16/2017 | 11:25:40 AM
Re: Phishing emails (responding to Joe's comment)
@Petar: Does the 16% refer to overall phishing susceptibility, or susceptibility to the fake phishing email itself?

How much of this do you think is potentially attributed to people "learning" what's a "fake" phishing email?

For instance, I wonder to what extent people are thinking, "Oh, that's that wily IT department again.  I don't want to have to do another 5-minute training" (or whatever), vs. "Oh, I shouldn't click on that because it might be bad"?

It's the difference between not slowing down on the highway as a matter of safety vs. slowing down on the highway because you think there's a speed trap ahead.
Petar Zivovic
50%
50%
Petar Zivovic,
User Rank: Apprentice
3/16/2017 | 12:29:55 PM
Re: Phishing emails (responding to Joe's comment)
The service we use allows for a number of templates, sender domains, subjects, etc. Given the wide variety of testing, I would lean towards the 16% being an accurate measure of susceptibility within the company as of that point in time. Note: I am seeing actual phishing emails being submitted for analysis using our phish tool on a daily basis, along with the usual spam and unwanted email, so at least some people are in fact paying attention.

"It's the difference between not slowing down on the highway as a matter of safety vs. slowing down on the highway because you think there's a speed trap ahead." This is a very good point. The number did go up when I missed a month of testing, so people apparently let down their guard. Seems to me that is just part of human nature for some people - they won't discipline themselves unless someone else is watching them. That being said, it's still way better than the initial baseline test.
Joe Stanganelli
50%
50%
Joe Stanganelli,
User Rank: Ninja
3/17/2017 | 11:53:36 AM
Re: Phishing emails (responding to Joe's comment)
@Petar: Fascinating.  I wonder what would happen if you upped the frequency (say, once a week) for a month or so, and then dropped back down to "normal" levels.
Petar Zivovic
50%
50%
Petar Zivovic,
User Rank: Apprentice
7/14/2017 | 5:30:27 PM
Re: Phishing emails (responding to Joe's comment)
I did exactly one a week for May, then did only one for June.

Result: Frequency seems to have little to no effect. Content has a big effect.

Examples: testing "Outlook password change required" and "Mint Financial Access" got few or no clicks.

But I got over two dozen clicks on "HR Notification re:Vacation policy change".

"Mint" was the one sent a month after the weekly tests were done.
Joe Stanganelli
50%
50%
Joe Stanganelli,
User Rank: Ninja
7/18/2017 | 1:33:31 PM
Re: Phishing emails (responding to Joe's comment)
@Peter: Wow. Awesome insights. Thanks for trying this and sharing!

I guess you'd really need an A/B grouping while testing frequency. Group A gets subject line A, Group B gets subject line B. Then that would give you a better idea on if it was more the frequency or the subject line.

But this is cool info to go on. I think people are on to the "pretending to be your bank" scam...but the HR stuff has been pretty hot among scammers lately because it's new (and, also, I suspect, because it's about something interesting enough to be worth an email open but not so jarringly enticing that it's automatically suspicious). 
RetiredUser
50%
50%
RetiredUser,
User Rank: Ninja
7/31/2017 | 2:28:06 PM
Re: %
I love that idea, Joe, of forcing completion of a 5-minute InfoSec training :-)  I've worked at orgs before where you essentially worked within a locked down piece of software with no access to anything else, but freeing up the desktop with the illusion of freedom presents lots of "teachable moment" opportunities for getting training compliance up-to-snuff quickly!
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
1/30/2017 | 12:48:24 PM
Re: %
"I wanted to see what would happen."

I see their reasoning. There should be second level protection. I should be able to click the link and still be protected.
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
1/30/2017 | 12:42:46 PM
Nature of attacks
The nature of attacks are two folds in my view. One you trick the user so you can get a privilege access to the system and another one you know a back door that most others do not. Government sponsored ones are more likely they have a back door to the systems. System vulnerabilities are not the main paths for the attacks.
Dr.T
0%
100%
Dr.T,
User Rank: Ninja
1/30/2017 | 12:47:23 PM
Source of attacks
Now that security attacks created a new industry I suspect that lots of security firms are behind of lots of those attacks to sell their products. I do not have a proof for it, it is just my guess. 
ClarenceR927
50%
50%
ClarenceR927,
User Rank: Strategist
1/31/2017 | 11:24:00 AM
Re: Source of attacks
In the case of the DNC attack the evidence is very clear that the work was done by Russian speaking agents using Russian systems for C&C. This is not in dispute dispite what the author implies.

 

The theory that AV companies write all the best viruses is as old as AV software and has been demonstrated false any time it has been investigated.
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
1/30/2017 | 12:48:51 PM
AV vs. DDOD or social engineering
Agree with the article, AV is an outdated strategy, nobody spends time to write a virus, there is more exciting ways of doing impact such as DDOD and social engineering.

 
Joe Stanganelli
50%
50%
Joe Stanganelli,
User Rank: Ninja
2/2/2017 | 4:07:24 AM
Re: AV vs. DDOD or social engineering
I wouldn't go so far as to say that antivirus = outdated -- at least as a sole lines of defense.

Maintaining up-to-date antivirus is like locking your door.  Sure, there are more exciting and terrible things that can happen to bypass antivirus, but if you're not even doing that, then that's pretty lax.
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
1/30/2017 | 12:51:58 PM
passwords
"I will password-spray your Web interfaces,"

I see the points in the article, however I think making password compels is not a solution, they are not really cracking the passwords, that is too much unnecessary work, they are getting it from the users.
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
1/30/2017 | 12:54:28 PM
security fundamentals
" old security fundamentals aren't effective"

This is a good point, industry has change, there is no more firewall to sell because everybody has it, they needs to sell something new.
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
1/30/2017 | 12:56:34 PM
UBA
"user behavioral analytics (UBA)"

I agree, this may be a good starting point, at the end of the day everything starts with the user behavior.
JulietteRizkallah
50%
50%
JulietteRizkallah,
User Rank: Ninja
1/30/2017 | 3:06:55 PM
Reconnaissance, infiltration, exploitation, exfiltration: the 4 phases of a data breach
While the reconnaissance phase can take months even years, the exfiltration will take days.  So i agree strongly UBA is critical to flag anything abnormal.  Identity governance is another critical tool to mitigate data breach, especially from insiders , of whom government agencies have seen their share.
kasstri
50%
50%
kasstri,
User Rank: Strategist
3/1/2017 | 6:52:04 AM
keyboard
Sure, there are more exciting and terrible things that can happen to bypass antivirus, but if you're not even doing that, then that's pretty lax.
Sodinokibi Ransomware: Where Attackers' Money Goes
Kelly Sheridan, Staff Editor, Dark Reading,  10/15/2019
Data Privacy Protections for the Most Vulnerable -- Children
Dimitri Sirota, Founder & CEO of BigID,  10/17/2019
7 SMB Security Tips That Will Keep Your Company Safe
Steve Zurier, Contributing Writer,  10/11/2019
Register for Dark Reading Newsletters
White Papers
Video
Cartoon Contest
Write a Caption, Win a Starbucks Card! Click Here
Latest Comment: The old using of sock puppets for Shoulder Surfing technique. 
Current Issue
7 Threats & Disruptive Forces Changing the Face of Cybersecurity
This Dark Reading Tech Digest gives an in-depth look at the biggest emerging threats and disruptive forces that are changing the face of cybersecurity today.
Flash Poll
2019 Online Malware and Threats
2019 Online Malware and Threats
As cyberattacks become more frequent and more sophisticated, enterprise security teams are under unprecedented pressure to respond. Is your organization ready?
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2019-17513
PUBLISHED: 2019-10-18
An issue was discovered in Ratpack before 1.7.5. Due to a misuse of the Netty library class DefaultHttpHeaders, there is no validation that headers lack HTTP control characters. Thus, if untrusted data is used to construct HTTP headers with Ratpack, HTTP Response Splitting can occur.
CVE-2019-8216
PUBLISHED: 2019-10-17
Adobe Acrobat and Reader versions , 2019.012.20040 and earlier, 2017.011.30148 and earlier, 2017.011.30148 and earlier, 2015.006.30503 and earlier, and 2015.006.30503 and earlier have an out-of-bounds read vulnerability. Successful exploitation could lead to information disclosure .
CVE-2019-8217
PUBLISHED: 2019-10-17
Adobe Acrobat and Reader versions , 2019.012.20040 and earlier, 2017.011.30148 and earlier, 2017.011.30148 and earlier, 2015.006.30503 and earlier, and 2015.006.30503 and earlier have an use after free vulnerability. Successful exploitation could lead to arbitrary code execution .
CVE-2019-8218
PUBLISHED: 2019-10-17
Adobe Acrobat and Reader versions , 2019.012.20040 and earlier, 2017.011.30148 and earlier, 2017.011.30148 and earlier, 2015.006.30503 and earlier, and 2015.006.30503 and earlier have an out-of-bounds read vulnerability. Successful exploitation could lead to information disclosure .
CVE-2019-8219
PUBLISHED: 2019-10-17
Adobe Acrobat and Reader versions , 2019.012.20040 and earlier, 2017.011.30148 and earlier, 2017.011.30148 and earlier, 2015.006.30503 and earlier, and 2015.006.30503 and earlier have an use after free vulnerability. Successful exploitation could lead to arbitrary code execution .