F5 BIG-IP APM - SSO via Kerberos

I’m working lately, and again, a lot with F5 APM. I think, this product has lots of interesting features. For instance, we can use F5 APM for SSL VPN remote access. We can configure a Custom Login Page, we can configure Host Checking, we can configure OTP Authentication. We can configure SSL VPN tunnel with Edge Client or SSL VPN with Portal Access Webtop. In addition, we can use F5 APM for Identity Federation and SSO. For example, we can enable SSO via SAML to applications such as SAP, AWS, Salesforce, etc or even third-party applications. We can also enable SSO via forms based authentication, HTTP authentication, NTLM, Kerberos and OAuth. F5 APM can also work as Citrix ICA Proxy allowing F5 APM to publish Citrix apps. Actually, F5 APM is a full proxy appliance which can be used as a secure access proxy.

These weeks, I’m working on a project to migrate an Apache server to F5 LTM+APM. The aim is to migrate all the configuration from Apache to F5. There are virtual hosts easy to migrate with iRules, Pools and Virtual Servers. However, there are also authentication configurations a little bit more complex to configure and migrate. For instance, SSO is configured via Kerberos in the Apache server. Therefore, users authenticated in the Active Directory can use apps without sign in again.

Apache configuration for Kerberos Authentication

On one hand, I have to configure Basic Authentication for users who haven’t been authenticated yet by the Active Directory. Therefore, F5 APM has to retrieve user credentials (username and password) from a browser. It’s a form-based authentication with an standard login screen. Thereafter, F5 APM populates the username and password session variables in Active Directory. Once users have been authenticated successfully, they can access to apps.

On the other hand, I have to configure Kerberos Authentication for users who have already been authenticated by the Active Directory. With the Kerberos method, the client system must first join a domain and a Kerberos action must follow. Therefore, the client firstly becomes a member and connects to the domain. Secondly, the clients connects to a virtual server on the BIG-IP system. Thirdly, the access policy runs and issues a 401 HTTP request action. Fourthly, if Kerberos is present, the browser forwards the Kerberos ticket along with the request when it receives the 401 HTTP request. Finally, F5 APM validates the Kerberos ticket after the request is received and determines whether or not to permit the request.

How Kerberos end-user logon works

The access policy for Kerberos Authentication with End-User Logons is really easy to configure. We have to add three boxes in the Visual Policy Editor. The first one is an HTTP 401 Response box to request clients credentials. The second one is an AD Auth box for basic authentication. The third one is a Kerberos Auth box for clients who have already been authenticated by Active Directory. Therefore, this last box enables SSO via Kerberos.

Example access policy for end-user login

Thanks my friends!! Are you ready to configure SSO via Kerberos?

Lazarus Group

I was reading about the Lazarus Group last week because two new alarms were received from the Ariolo SIEM: The Disclosure of Chilean Redbanc Intrusion Leads to Lazarus Ties alarm and the Lazarus targeting banks in Russia alarm. However, I wanted to read more and more about the Lazarus Group to know the attacks they are responsible for. I’m going to write about the mainly cyberattacks which have been attributed to them over the last decade.

One of the first cyberattack attributed to Lazarus Group is the Operation Troy in 2009. There were actually three waves of attacks on July against US and South Korean websites. The hackers utilized the Mydoom and Dozer malware to launch a DDoS attack against government websites. The attacks continued, later on, against South Korea. The Operation DarkSeoul in 2013 targeted three South Korean broadcast companies, financial institutes, and an ISP. These attacks damaged 32.000 computers and servers by malicious code.

Operation Troy and Dark Seoul

I think one of the most known attack is the Sony breach in 2014. They took more than 100 terabytes of data from Sony such as personal information of employees and their families, e-mails, information about executive salaries, copies of unreleased Sony films and other information. The attackers used a SMB worm tool as well as listening implant, backdoor, proxy tool, destructive hard drive tool, and destructive target cleaning tool. Little by little, they had been siphoning Sony’s data for over months or years.

Sony breach

If you want to know more about the Lazarus Group is highly recommended to read the Operation Blockbuster. This is a research led by Novetta in 2016 where they have analised malware samples found in different cyber-security incidents. They were able to link the Lazarus Group to a number of attacks through a pattern of code re-usage. For instance, they found six user-agents reused over and over that included the same misspelling of “Mozillar”.

Operation Blockbuster

Another really known attack is the WannaCry attack which was released in 2017. Initially, this attack was attributed to China, Hong Kong, Taiwan or Singapore, finally, this attack was attributed to the Lazarus Group from North Korea. The WannaCry attack used an exploit of Windows SMB protocol and a backdoor tool. The exploit, named EternalBlue, was used to spread out laterally to random computers and the backdoor, named DoublePulsar, was used to grant cybercriminals a high level of control over the computer system. Tihs ransomware malware demanded a payment of around US$300 in bitcoin within three days, or US$600 within seven days.

Information about the file encryption

To sum up, the Lazarus Group is a really dangerous cybercrime group who has attacked many companies over the last decade. Lately, the target is the banking sector, where they get money. For instance, Lazarus Group has stolen $49 million from an institution in Kuwait last year and United Nations investigators estimate they have already stolen $2 billion.

Thanks my friends!! Did you know this hacking group?

Lazarus targeting banks

I wanted to read about the Lazarus Group since I received two alarms from the Ariolo SIEM. These two alarms are “Disclosure of Chilean Redbanc Intrusion Leads to Lazarus Ties” and “Lazarus targeting banks in Russia”, which have been developed by the same actor group, Lazarus group, against the same industry, the finance sector. One customer has received these alarms. Therefore, it’s the best way to take the plunge to read deeply about these two alarms.

The target of the first alarm is the Chilean interbank network Redbanc which was attacked with a malware toolkit that was installed on the company’s corporate network without triggering antivirus detection. It’s amazing how the victim was deceived. The lure begins with a job offer posted to a social media network (LinkedIn) where the victim applies to this offer. The attackers contact the victim and an interview conversation occurs. However, the interviewer eventually asks applicants to download and execute a tool (ApplicationPDF.exe) on their computer in order to generate their application form in PDF format. The execution of this tool kicks off the infection process.

The dropper malware is disguised as a legitimate software for job applications

Once the victim’s computer executes the attacker’s tool, the sample communicates the C&C URL (hxxps://ecombox[.]store/tbl_add[.]php?action=agetpsb), and after connecting, it drops a script file (REG_TIME.ps1) used to invoke the PowerShell process. This malware checks if the victim’s user has administrator privileges. If it has admin privileges, the PowerShell attempts to download the next stage and register it as a service. If it has no admin privileges, the malware is on memory till next reboot. However, the malware is useful as reconnaissance tool to deploy more malware if it’s needed.

ThreadProc decodes the Base64-encoded values and executes the PowerShell script

The target of the second alarm is the Russian banking sector. This campaign uses malicious Office documents delivered as ZIP files, along with a benign PDF document called NDA_USA.pdf that contains a StarForce Technologies agreement, which is a Russian software company that provides copy protection software. 

ThreadProc decodes the Base64-encoded values

Actually, there are three steps. Firstly, the ZIP file has two documents: the lure file, which is a benign PDF file, and the malicious file, which is a Word file with macros. Secondly, the malicious macro download a VBS file from Dropbox and this malicious script is executed. Finally, the script VBS download a CAB file from a malicious server, which extract an EXE file, and execute the RAT malware. Once the RAT (Remote Administration Tool) is executed, the attackers send commands from the C&C server to the victim’s computer.

Infection flow of KEymarble malware

I think these are two really interesting alarms to the banking sector. I work for the spanish banking sector from Spain, and these two alarms are against Russian and Chilean banking sector. However, I’ve also seen these alarms in Spain. Therefore, cyber threat intelligence tools along with SIEM appliances are increasingly useful to detect and block this kind of APT attacks. I think these kind of tools should already be mandatory for most banking companies.

Thanks my friends!! Have you ever read about these alarms?

Cyber Threat Intelligence

I wrote about Collective Intelligence Framework years ago. I think it’s a great idea because it’s an easy way to share malicious activities such as IP addresses, hashes, certificates and domains which are used to attack users and companies. In addition, we can use this information along with SIEM, firewall and IDS to detect and block attacks quickly. Therefore, it’s highly recommended to install collective intelligence tools to have visibility of malicious attacks.

Maybe, you are wondering what is collective intelligence? How can I install collective intelligence tools? Firstly, you have to know about Cyber Threat Intelligence. You have to know there are three types of threat intelligence. Tactical or technical intelligence which can be used to detect attacks. Operational which can be used to know the motivation and capabilities of threat actors. Strategic which can be used to drive high-level organizational strategy. Lately, once you know about threat intelligence, you can deploy tools.

The first type of Cyber Threat Intelligence is the easier to deploy. Technical intelligence is mainly Indicators of Compromise (IoC). There are lots of tools and services which are really useful to know about malicious IP addresses, malicious file types, malicious domains, malicious SSL certificates, etc. All of these IoC are useful to detect attacks. For instance, OTX from Alienvault, Threat Campaign from F5 Networks or FortiGuard from Fortinet are services which help us to detect and mitigate attacks.

There are lots of security tools such as network firewall, WAF or SIEM which use IoC to detect attacks. If there were no IoC, these security tools won’t know, for instance, which IP address is malicious and which is benign? Therefore, sharing this information and these IoC with security tools is mandatory today to identify attacks. Actually, I think, correlation engines, which are installed in SIEM appliances, are mandatory to know all the bad things that are inside companies networks.

There is an interesting tool, which I’m going to install soon, that it’s really useful. MISP or Malware Information Sharing Platform is an open source threat intelligence platform where there are lots of IoC. This is a project funded by the European Union which started around June 2011 and it’s still alive. We can use MISP for sharing, storing and correlating IoC of targeted attacks, threat intelligence, financial fraud information, vulnerability information or even counter-terrorism information.

To sum up, cyber threat intelligence is increasingly used. Technical intelligence and IoC are widely used by most companies. However, I think, there will be a Next Generation of Cyber Threat Intelligence where Operational and Strategic intelligence will become more important than Tactical intelligence and most companies will want to have a threat sharing platform to know who, when, why and how they are attacked.

Thanks my friends!! do you know about Cyber Threat Intelligence?