Category Archives: Puzzling discussions

Discussions about IT security issues really puzzling me.

Concerns about using open source libraries from an IT security point of view

18 December 2017

Some days ago I participated in a discussion about the necessity of using open source libraries in industrial software development and the data scientist workbench. IT security is often perceived as spoil sport in such discussions …

To be honest, I like open software. I prefer for example Firefox on Windows 10 because the configuration of Edge is really annoying. However, when it comes to the use of open software libraries in scientific or industrial software development projects or by data scientists I have two major concerns:

1. I have just no clue what the open software libraries do in addition to their intended use.

This sounds a bit paranoid. The question is:

Can we make sure, that no malicious code snippets are hidden in an open software library which send the company’s secrets to a cyber criminal’s command and control server, or which encrypt all data?

In my opinion this is not possible. Reviewing e.g. the 300 thousand lines of code of the OpenSSL-1.0.2 project is a herculean task, which has to be repeated for every patch and release. We can automate the software review process with advanced code analyzers. With such analyzers, we can make sure that open source code has no or few critical errors. But analyzers cannot find malicious code snippets, they just make sure that such snippets cause no critical errors during program execution.

Advanced Persistent Threat (APT) solutions may detect malicious behavior. But when a developer or data scientist includes open software into his code, the threat type changes from external to insider threat, thus APT solutions are no longer effective.

Eventually, we have to trust the developers of open software. Thus, the use of open software depends largely on the risk appetite of an organization.

2. I have no idea how to fix vulnerabilities in software which uses open software libraries.

Firefox gets security patches immediately after vulnerabilities are published. For example, the remote code execution vulnerability CVE-2017-7827, published 11/15/2017, was patched on the morning of 11/17/2017. When I logged in to my Linux box in the evening, even a patch for the Firefox ESR version was installed.

The OpenSSL-1.0.2 library mentioned above can be used potentially in many applications, in the worst case, some of them may be connected directly to the internet. The developers of Firefox take care of security bugs in this library. Who cares in the case of self-developed software? And how fast? Just remember the Equifax data breach some months ago. The reason for this really costly data breach was an unpatched vulnerability in the Apache Struts framework …

The focus of open software developers is innovation. Thus, the use of open software will be a major driver in the digital transformation, and we should foster this use to stay at the cutting edge of digital transformation.

Nevertheless, we must be aware of the risks of this use and take proper precautions for their mitigation.

Have a great week.

Advertisements

Puzzling: Five years old critical vulnerabilities exploited in November 2017

26 November 2017

Section Exploited Vulnerabilities of the Recorded Future Cyber Daily is sometimes really frightening. On November 9th, 2017, 249 successful exploits of CVE-2012-1823, a vulnerability in PHP, were recorded. This is hard to believe because CVE-2012-1823 was published on May 11th, 2012. Although a patch was available at the date of publication, it seems that the operators of this systems were not able to implement them within the past five years.

However, it would have been of urgent need in this case. CVE-2012-1823 is a so-called RCE (Remote Code Execution) vulnerability, which allows remote attackers to execute arbitrary code on a victim’s computer, and, in the worst case, to hijack the victim’s network.

RCE vulnerabilities are included in the critical vulnerabilities. Critical vulnerabilities are

  • exploitable from the network
  • need only low or medium skills to exploit
  • need no authentication
  • cause great damage, have high severity
  • allow remote attackers to execute arbitrary code on the victims computer

If an application system is operated in the DMZ, critical vulnerabilities must be patched directly upon publication to prevent attackers from getting onto your network. Or at least, between the time of publication and an exploit or proof of concept shows up. Since examples of how to exploit this PHP vulnerability were available in early May 2012, immediate action was required.

The big question is: Why were this vulnerable PHP versions not directly patched?

Exploitation of older vulnerabilities is not an isolated case. The HPE 2016 Cyber Risk Report shows, that in 2016

  • 47% of successful exploits use five or more years old vulnerabilities.
  • 68% of successful exploits use three or more years old vulnerabilities, 47% of them were critical vulnerabilities.
  • Stuxnet, CVE-2010-2568, was used in 29% of successful exploits.

An analysis of the critical vulnerabilities by vendors shows, that more critical vulnerabilities were found in non-Microsoft products than in Microsoft products.

Critical vulnerabilities 2010 - 2016

Critical vulnerabilities 2010 – 2016 by vendors. Click to enlarge.

But automated patch management is only available for Microsoft and few of the other vendors’ (e.g. Adobe, Oracle, SAP) products. Thus, we can expect that many critical vulnerabilities remain unpatched, which results in an ever-growing pool of opportunities for cyber criminals.

An ever growing pool of opportunities

An ever-growing pool of opportunities. Click to enlarge.

1) For the chart above I assumed that 50% of critical vulnerabilities remain unpatched. This assumption is based on the analysis of the 2017 NIST NVD data as of August 31st, 2017.

Since no automated patch management exists for PHP we can expect, that CVE-2012-1823 was rarely patched. But the worst is yet to come: From the HPE 2016 Cyber Risk Report we learn, that even six years old Microsoft vulnerabilities (Stuxnet, CVE-2010-2568) are not patched.

How to tackle this issue? From my point of view, the cause is compliance driven security. We often do patching of everything to meet compliance with a certain standard, instead of focusing on the real important issues, e.g., the critical vulnerabilities. Or, in other words, we close a lot of mouse holes while the barn door remains wide open.

WIth this, we must move from patching to vulnerability management, and priority patching for the critical vulnerabilities. Through a differentiated inspection of vulnerabilities we get out of the patch treadmill and can start working on the important cyber security issues.

By the way, if you haven’t subscribed to the Recorded Future Cyber Daily yet, consider to do it this week.

Have a great week.

WannaCry, Rumsfeld and Production Firewalls

21 May 2017

Today, Firewalls are the preferred means to separate a production network from a company’s intranet. Firewall configuration is performed by the Rumsfeld Conundrum: Block everything you don’t know!

Rumsfeld Conundrum for firewall configuration

Rumsfeld Conundrum for firewall configuration

For production management and IT and OT operations, we need some communication between systems in the company intranet and the production network. These required (known) connections are defined in the firewall rule base. The firewall allows communication between these known systems, and blocks any other connection attempts.

As long as the SMB V1.0 protocol is not used for communication across the firewall, the Rumsfeld Conundrum works pretty well.

Unfortunately, the SMB protocol is frequently used to implement required connections between Windows-based computers in the company intranet and the production network, e.g. for the exchange of manufacturing orders. With this, production systems become vulnerable to WannaCry although a firewall is in place because the firewall does not block communication across required connections. In the worst case, if WannaCry spreads across required connections to systems in the production network, this may result in loss of production.

Immediate action is required. The firewall rule base is a good starting point to determine how big the problem is, and to identify the systems that must be immediately patched or otherwise secured, if patching is not possible due to technical or regulatory restrictions.

Firewalls are an indispensable part of a defense in depth concept, but plain packet filtering is no effective means against attacks like WannaCry.

Have a good week, and take care of you production networks.

Vastly improve your IT security in 2 easy steps?

1 April 2017

Keep your software patched and defend against social engineering, and you will win the battle against the bad guys. Let me be clear: From my point of view this is simply not enough. Nevertheless, Roger A. Grimes’ post “Vastly improve your IT security in 2 easy steps” published on March 21, 2017 at InfoWorld is really worth reading, in particular the section about patching.

The key to diminishing this risk is to identify the right software to patch and do it really, really well. The risk reducers I respect know the difference between the largest unpatched program in their environment and the unpatched program mostly likely to be exploited in their environment. A security expert knows there is usually a gulf between the two.

In particular in the production domain, where patching has always to be delayed to the next scheduled maintenance, this is a very important hint.

The big question is: How can we identify the right software on the right and important systems? Without an up-to-date asset directory with the relevant details about cyber security this is a very complex and expensive matter.

But even with an up-to-date asset directory this remains a complex task.

Rockwell/Allen Bradley Systems directly connected to the Internet

Rockwell/Allen Bradley Systems directly connected to the Internet in North America

For example, the likelihood of a cyber-attack on an Industrial Control System (ICS), which is directly connected to the internet, is many times higher than the likelihood of an attack on an ICS which is completely isolated in a security zone within the production network. The first ICS is definitely one of those systems Roger Grimes has in mind, the latter can be ignored.

But the likelihood of a cyber-attack is only half the story. For example, in functional safety the risk is the combination of the probability that a hazard will lead to an accident and the likely severity of the accident if it occurs. Thus, from this point of view, even the first ICS may be uncritical unless it is not used for controlling a critical infrastructure.

To identify the right and important systems is the hard task. It requires an up-to-date asset inventory and a smart risk management process. The plain patching process is just a piece of cake.

Have a good weekend.

Cybersecurity is just too much trouble for the general public, claims study

8 October 2016

In report ‘Cybersecurity is just too much trouble for the general public, claims study’ published on 6 October at the Tripwire state-of-security blog, Graham Cluley cites from the NIST study Security Fatigue:

“Participants expressed a sense of resignation, loss of control, fatalism, risk minimization, and decision avoidance, all characteristics of security fatigue. The authors found that the security fatigue users experience contributes to their cost-benefit analyses in how to incorporate security practices and reinforces their ideas of lack of benefit for following security advice.”

We should not be surprised ‘that the public is suffering from “security fatigue” and a feeling of helplessness when it comes to their online security’. Most of the advice for end users in the information security domain is just puzzling. Let me make this clear with an example.

Renowned German Stiftung Warentest assessed 15 e-mail providers in the October 2016 edition of the Test magazine. Focus of the assessment was data privacy, ‘the protection of customers and emails against unwanted looks’. And, of course, usability. Table 1 below shows the Stiftung Warentest quality ranking.

Provider Quality Ranking (1)
Mailbox.org Tarif Mail 1.4
Posteo 1.4
Mail.de Plusmail 2.2
GMX Topmail 2.3
Web.de Club 2.3
Web.de Freemail 2.5
GMX Freemail 2.6
Telekom Freemail 2.6
Freenetmail Basic 2.7
Telekom Mail / Cloud M 2.7
1&1 Mail Basic 3.1
AOL Mail 3.1
Yahoo Mail 3.2
Microsoft Outlook.com 3.3
Google Gmail 3.4

Table 1: Stiftung Warentest rankings

(1)    Quality Ranking: 0.5 .. 1.5: Very good, 1.6 .. 2.5: Good, 2.6 .. 3.5: Average

At a first glance, the table suggests that it is sufficient to use one of these providers (all were rated from very good to average) and security is guaranteed.

Unfortunately, this assessment is very misleading. Email encryption is just one aspect of information security. It protects against cyber criminals, state-sponsored attackers or insider attacks because the information is not readable unless the attacker has access to the encryption key.

If an attacker is able to compromise a user’s account, e.g. through a password phishing attack, he might have full access to all emails, although they are encrypted.

To secure an account against phishing with frequent password changes and the use of individual passwords for different services, is not sufficient. And usability is bad, even if password managers are used. Two-Factor Authentication (TFA) or one-time passwords are the tools of choice to enhance security against phishing attacks.

Table 2 shows the Stiftung Warentest results updated with details about TFA availability.

Provider Quality Ranking (1) TFA available With soft token With SMS With hard token
Mailbox.org Tarif Mail 1.4 (2) Yes Yes Yes
Posteo 1.4 Yes Yes  
Mail.de Plusmail 2.2 Yes Yes Yes
GMX Topmail 2.3 No
Web.de Club 2.3 No
Web.de Freemail 2.5 No
GMX Freemail 2.6 No
Telekom Freemail 2.6 No
Freenetmail Basic 2.7 No
Telekom Mail / Cloud M 2.7 No
1&1 Mail Basic 3.1 Undef. (2)
AOL Mail 3.1 Yes Yes
Yahoo Mail 3.2 Yes   Yes  
Microsoft Outlook.com 3.3 Yes Yes Yes
Google Gmail 3.4 Yes Yes Yes Yes

Table 2: Rankings updated with details about TFA

(1)    Quality Ranking: 0.5 .. 1.5: Very good, 1.6 .. 2.5: Good, 2.6 .. 3.5: Average

(2)    It was not possible to determine whether TFA is available from the provider’s homepage

Only 7 of the 15 email providers allow the use of a second factor. The limitation to one aspect of information security creates puzzling results and a false sense of security. It is therefore no wonder that consumers show the ‘characteristics of security fatigue’.

TFA with soft tokens is under normal conditions activated within seconds, and very easy to use. From my point of view, service providers should create the needed attention and force the use of TFA. It is not sufficient to notify the users of new waves of phishing attacks.

Have a good weekend.

Apple delivered patches to mitigate state-sponsored Trident attack – Millions of Android devices potentially vulnerable?

10 September 2016

During my bicycle trip to the springs of the White Main in the Fichtel mountains news about the state-sponsored Trident attack on IOS devices went around the world. The topic was front page news even of local newspapers, very often with a certain malicious joy, because Apple’s IOS is well-known for its superb security.

Within some days Apple developed patches for the vulnerabilities and delivered them to IOS devices in the field. This was taken for granted from the public, but it is very remarkable, because only Apple and Microsoft are able to deliver ad hoc patches for their mobile device operating systems.

In report ‘A Hacking Group Is Selling iPhone Spyware to Governments’, published on 25 August on WIRED, one could read:

“NSO Group won’t be able to use this particular attack anymore on iPhones running the latest version of iOS—and one of the operating system’s strongest selling points is its high adoption rates for new versions. In the meantime, the Citizen Lab and Lookout researchers say that there is evidence that the group has ways to get Pegasus spyware onto other mobile operating systems, notably Android.

With this, all devices running Android, and this is the majority of devices, are potentially vulnerable for the Trident attack, and will remain vulnerable for their entire lifetime.

Or have you ever heard from a smart phone vendor who delivers patches for Android devices in a timely manner, and for older devices?

Have a good weekend.

France says fight against messaging encryption needs worldwide initiative

13 August 2016

The report “France says fight against messaging encryption needs worldwide initiative“, published on Reuters technology news last Thursday, is truly worrying.

“Messaging encryption, widely used by Islamist extremists to plan attacks, needs to be fought at international level, French Interior Minister Bernard Cazeneuve said on Thursday, and he wants Germany to help him promote a global initiative.”

I can, of course, understand the motivation of the French Interior Minister. He must do his utmost to protect France from further terrorist attacks.

“French intelligence services are struggling to intercept messages from Islamist extremists who increasingly switch from mainstream social media to encrypted messaging services, with Islamic State being a big user of such apps, including Telegram.”

Although the French Interior Minister has not requested decryption options from service providers yet, the direction of a Franco-German initiative is from my point of view clear: Service providers shall make decryption options available to national police and intelligence and security services.

With this, some attacks can certainly be prevented, but on the other hand, it puts many innocent people, which care of civil rights in authoritarian regimes, at risk.

In “Exclusive: Hackers accessed Telegram messaging accounts in Iran – researchers“, published in Reuters CYBERSECURITY at 2 August 2016, Joseph Menn and Yeganeh Torbati reported, that Iranian hackers compromised accounts on Telegram.

The security researchers who researched the attack said that “… the Telegram victims included political activists involved in reformist movements and opposition organizations. They declined to name the targets, citing concerns for their safety.”

“We see instances in which people … are targeted prior to their arrest,” Anderson said. “We see a continuous alignment across these actions.”

That is precisely the problem when national security services demand decryption options from service providers: The information can be used to prevent terrorist attacks, as well as for violent actions against dissidents among the citizens. Hopefully the German Interior Minister will remember the recent German history (Stasi) and reject those demands once and for all.

By the way, end-to-end encryption is the just the comfortable way of secure communications. Terrorist can turn to less comfortable, but high secure encryption options like PGP. With this the French initiative makes no longer sense because the messages are encrypted before the transport to the service provider. Even end-to-end encryption is not required.

Even though it is apparent from the context, Benjamin Franklin’s quote about liberty and safety fits very well here:

Those who would give up essential liberty, to purchase a little temporary safety, deserve neither liberty nor safety.

Have a good weekend.