New Study Shows Correlating Network and Endpoint Data is Highly Manual

We recently commissioned Forrester Consulting to survey IT security professionals to find out what their desired end state was when it came to correlating security intelligence from network and endpoint. Bringing together these two disparate threat vectors allows organizations to:

◈ Increase detection and prevention capabilities
◈ Reduce manpower and resources needed for containment (and therefore costs)
◈ Exponentially decrease remediation time

In short, these are perceived benefits as they are not really happening today. Surprisingly, most organizations reported high confidence in their current threat detection and remediation systems.

But do they really have the problem covered?

Turns out – No. Perception and reality differ in this case. Many respondents claim to have integrated systems but in practice, being able to make decisions about endpoint and network security requires considerable time and effort from teams, if the data can be used at all. This shouldn’t really come as much of a shock at all since we asked what security technologies they had implemented and what they were planning to implement. While there is no clear standout winner for what is going to be implemented, what is clear is of the 21 solutions that we inquired about, respondents are spreading their capital expenses all over the place. This is why most organizations are doing the work manually.

Too many tools, little integration, no automation

With so many different security solutions in place, it’s no wonder there is so much time spent doing manual analysis and investigation into security incidents. Earlier this summer I spoke with a lot of security professionals at the Gartner Security Summit and at Cisco Live who talked about how siloed their products were. The data produced by one tool couldn’t even be consumed by another, and the information they could correlate took forever. One conversation in particular that stands out was an incident responder from a large power company who talked about how they had taken more than 6 months to investigate a single incident because they couldn’t track back the path of infection, and identify how it was propagating through their network. This is not an uncommon story that we hear. Over the last decade so many tools have been deployed that it is now making the job harder, not easier. If only they could have a security architecture where the tools talked to each other, and correlated data automatically.

Automating data analysis for improved detection is a reality

The term “architecture” has been used so much it quite possibly is one of the few terms that requires more definition than “cloud”.  Simply put, we view an architecture as something that works together. Not a bunch of API’s that get cobbled together to push data somewhere (and eventually the API gets changed and that’s all broken…), and then the manual analysis happens, but a set of technologies, and specifically security tools, that all work together – automatically – to reduce the manual effort. This means having your endpoint detection and response solution (EDR) correlating files seen by your firewall or intrusion detection system with those analyzed your sandbox, and connect it with telemetry from the web proxy to identify associated traffic as well as command & control (CNC) infrastructure, and additional tools attackers are using – and all without you having to do anything.

While it may sound absurd, we call it Advanced Malware Protection, or AMP Everywhere. When you put the same eyes everywhere, you see everything. More visibility means a better ability to prevent advanced attacks.

For a good technical overview of how AMP works, check out this chalk talk.

Continue reading

How to Use the Plug and Play Template Editor in DNA Center – Part 3

The first and second blog posts in this series gave an overview of network Plug and Play (PnP) and how it has evolved in Cisco DNA Center.   They showed a very simple workflow to provision a device with a configuration template with a variable called “hostname.”   This was done by the UI and programmatically via the API.

This blog post looks at creating PnP configuration templates using template editor in Cisco DNA Center.  Here, we will cover the User Interface and basic concepts, and subsequent blog posts will cover advanced topics, Day-N provisioning and the associated API.

Template Editor

The template editor is a standalone application at the bottom of the Cisco DNA Center home page.  It can be used for Day-0 (PnP) or Day-N configurations.

When the editor is opened for the first time, a project needs to be created along with a template. Projects are like folders to contain and structure the templates you build.  The example below, shows the “base config” template used in the earlier blogs.  “pnp” and “adam” are just project names.

Creating a new Template

Click the “+” at the top of the template page or the gear beside a project to add a new template.  The “+” allows you to create a project or a template, while the gear creates a template with the project.

The “add new template” slide out will appear.  This contains metadata about the template, such as the device types it applies to and the flavor of IOS. The example below applies to routers and switches (all models) which run IOS-XE.  It is possible to restrict the template to a specific version of code or model of device.

NOTE:  It is possible to have a single template or a composite sequence of templates. Currently composite sequences are not supported in PnP.

Click on the template to edit it.  The three boxes on the top right are used to navigate between the following views:

◈ Edit – to edit/commit the template.

◈ Variable – provide metadata about the variables used in the template. “$” is used to signify a variable.

◈ Simulation mode – View the rendered template by providing a set of test values for the variables.

It is important to realize that templates have a 2-phase commit.  A template can be saved, but it needs to be “committed” before it can be used. Templates have version control based on the “commit process”.

First Version

After entering some commands, the template needs to be saved and committed.  Any string that starts with “$” will be treated as a variable. In this example, “$hostname” is a variable.  Multiple variables are supported.

Variable Types

After committing an initial template version, the variables view can be used to change the type of the variable if required.  Variables can also be marked as “not a variable”, which is useful for configuration strings that contain  “$”.  I will discuss this more in the advanced blog post.

Simulation

Simulations can be used to test the template with dummy variables.  This is particularly useful later on when using loops and other control structures in a template.

Select the simulation tab, and then the “New Simulation” action.

You then need to provide a value for the variables, and run the simulation to see the result.  Notice how the hostname variable has been replaced by its value (“fred”).

The simulation feature is particularly relevant with more sophisticated templates.

Continue reading

New XR Programmability Learning Labs and Sandbox Let You Explore

Turning team focus to network automation and programmability

I came from a network service provider background. Then, when I starting working at Cisco, I was working on the Cisco Security network team. The global network we built, owned, and managed was much like a service provider network. We had lots of transit links, and circuits with service providers, and tons of peering links, and sessions all over the world. Managing this was a full-time job, and I am just talking about managing the WAN (wide area network) here. Which is why, like many of you and other network teams out there whose network requires speed, scale, and data analytics, my team and I turned our focus to network automation and programmability.

The majority of our network devices (both core and edge) were running IOS XR. IOS XR has always been one of my favorite platforms, so it was with great excitement that when I began working for the Cisco DevNet team, my specialist area would be working with the IOS XR teams and platform.

What is new to learn here?

A great question, I am pleased you asked! We have built a dedicated sandbox environment for IOS XR programmability and learning labs to go with this.  The IOS XR Programmability sandbox and learning labs provide an environment where developers and network engineers can explore the programmability options available in this routing platform. These include:

◈ Model Driven Programmability with YANG Data Models, NETCONF and gRPC
◈ Streaming Telemetry
◈ Service-Layer APIs
◈ Application Hosting

What gear can you access in the sandbox?

We wanted to build a sandbox that provides the right level of simplicity for users to get started while offering a flexible platform they can build on. The sandbox provides two Cisco IOS XRv 9000 devices (R1 and R2) connected back to back, plus a Linux host that acts as a development box (DevBox). The image version on Sandbox tile is 6.4.1 this is available on both the two IOS-XR nodes.

The new IOS XR programmability sandbox lets you explore programmability options available in this routing platform.

The all-new learning labs and track

You can use the IOS-XR programmability learning track to familiarize yourself with the rich set of programmable interfaces and APIs offered by IOS-XR. The goal of this track is to introduce you to the architectural tenets of the IOS-XR network stack and showcase how APIs at every layer of the stack – from Manageability APIs like YANG models, CLI, ZTP hooks to Service Layer APIs at the network infrastructure layer can be used to completely transform the way you manage and provision your network.

◈ IOS-XR CLI automation Cisco IOS-XR offers a comprehensive portfolio of APIs at every layer of the network stack, allowing users to leverage automated techniques to provision and manage the lifecycle of a network device. In this module, we start with the basics: the Command Line Interface (CLI) has been the interaction point for expect-style scripters (TCL, expect, pexpect etc.) for ages; but these techniques relying on send/receive buffers are prone to errors and inefficient code. This is where the new onbox ZTP libraries come handy. Use them for automated device bring-up or to automate Day1 and Day2 behavior of the device through deterministic APIs and return values in a rich Linux environment on the router.

◈ IOS-XR Model-Driven Automation: YANG models Cisco IOS-XR offers a comprehensive portfolio of APIs at every layer of the network stack, allowing users to leverage automated techniques to provision and manage the lifecycle of a network device. APIs that are derived, documented and versioned using deterministic models are contractually obliged to match the expectations laid out by the model. Following this ethos, in IOS-XR, all the capabilities of the software, traditionally offered through the Command Line Interface (configuration commands, show commands, exec commands) are mapped to equivalent Config and Oper YANG models backed by the internal IOS-XR Database called SYSDB. In this module, we start taking the first steps towards model-driven programmability as we dive deeper into IOS-XR Yang models. We look at the interaction with these models with tools such as ncclient, YDK or gRPC clients and tips to map your CLI configurations to corresponding YANG-Modeled XML/JSON representations.

◈ IOS-XR Streaming Telemetry: SNMP is dead! It is time to move away from slow, polling techniques employed by SNMP for monitoring that are unable to meet the cadence or scale requirements associated with modern networks. Further, Automation is often misunderstood to be a one-way street of imperative (or higher-layer declarative) commands that help bring a network to an intended state. However, a core aspect of automation is the ability to monitor real-time state of a system during and post the automation process to accomplish a feedback loop that helps make your automation framework more robust and accurate across varied circumstances. In this module, we learn how Streaming Telemetry capabilities in IOS-XR are all set to change network monitoring for the better – allowing tools to subscribe to structured data, contractually obliged to the YANG models representing operational state of the IOS-XR internal database (SYSDB) at a cadence and scale that are orders of magnitude higher than SNMP.

◈ IOS-XR Service-Layer APIs: Cisco IOS-XR offers a comprehensive portfolio of APIs at every layer of the network stack. For most automation use cases, the manageability layer that provides the CLI, YANG models and Streaming Telemetry capabilities, is adequate. However, over the last few years, we have seen a growing reliance in web-scale and large-scale Service Provider networks on off-box Controllers or on-box agents that extract away the state machine of a traditional protocol or feature and marry their operation to the requirements of a specific set of applications on the network. These agents/controllers require highly performant access to the lowest layer of the network stack called the Service Layer and the model-driven APIs built at this layer are called the Service-Layer APIs. With the ability to interact with RIB, the Label Switch Database (LSD), BFD events, interface events and more capabilities coming in the future, it is time to take your automation chops to the next level.

The sandbox provides two Cisco IOS XRv 9000 devices (R1 and R2) connected back to back, plus a Linux host that acts as a development box (DevBox).

Getting Started

The development box includes a “hello world” sample app to check the uptime on routers to get you started.

hello-ydk.py

The script illustrates a minimalistic app that prints the uptime of a device running Cisco IOS XR. The script opens a NETCONF session to a device via the devices IP address, reads the system time and prints the formatted uptime.

Sample Output:

Continue reading

Why Organizations With Sensitive Research or Intellectual Property Need a Zero Trust Cybersecurity Framework Approach

The emergence of Zero Trust has shifted the center focus of some security frameworks from securing the perimeter to protecting sensitive data. While both are extremely important, this shift to a sensitive data-centric framework has advantages. To further understand the benefits of Zero Trust, consider a few specific scenarios:

◈ A large university that does over $100M in Federal Research

◈ Any company with intellectual property or in the process of acquiring or selling off organizations

◈ A state, county, or large city that needs to protect their Criminal Justics Information Services (CJIS) data

◈ Industrial Controls Systems (ICS); power, water, roads, or buildings.

◈ Election infrastructure security

Using those above situations, let’s start with the basics that you need to understand:

◈ Who is after your sensitive information:

     ◈ Where does it sit?

     ◈ What are the capabilities of the bad actors?

     ◈ What are the three biggest gaps that you need to address asap?

◈ Do you have an accurate inventory of your hardware?   Can’t protect what you do not know about…

◈ Inventory of your software and their application flows?  Most moves to cloud fail due to lack of insight related to dependencies.

◈ What are your key risks (threats, brand image, fines, and compliance).

◈ Understand what your top 50 pieces of sensitive data are.  Rarely does anyone do full data classification.

◈ Understand where your top 50 pieces of sensitive data presently reside.

◈ What are your organiazations capabilities around Segmentation, Priviledge Escalation Monitoring, and Multi Factor Authentication?

◈ Can you spot priviledge escalaton (user and application processes) ?

◈ How well are your security solutions integrated? Automated? Use the same intelligence?

Then analyze where you are with the necessary people, process, and technology basics. Most organizations should leverage the resources and technologies that they already have and understand where the gaps are, so they can address them over the next one to three years.   Cisco Advanced Security Services can help you with this analysis, strategy, implementation analysis, design, pilot, and implementation work.

Go to a workshop with Cisco Advanced Services, so you understand what the gaps are, how to best address them, and prioritize your work.  This end-to-end approach will help you address your key use cases to get the outcomes you need addressed.

Some of Cisco’s Related Zero Trust Services

◈ Strategy, Risk, & Programs IT Governance
     ◈ Security Strategy & Policy
     ◈ Security Program Maturity Assessment
     ◈ 3rd Party Risk Program
     ◈ Security Program Development
     ◈ Identity & Access Management
◈ Infrastructure Security
     ◈ Network Architecture Assessment
◈ Integration, Automation, and Advance Analytics

Cisco is actively involved with organizations with these types of challenges. We have the product and services experience to help you determine a practical systems approach to Zero Trust.  Reach out to your Cisco Security Services team so we can help guide your through this.

9 Pillars Of The Zero Trust Ecosystem – Jeff’s View

Continue reading

Moving Towards The Zero Trust Cybersecurity Framework?

The first step should be an investigation and analysis of what your sensitive data is, where it lives, and who accesses it. Then analyze the three Foundational Pillars (see below) to see where you are with the necessary people, process, and technology basics. Most organizations should leverage the resources and technologies they already have in place, and understand where the gaps are so they can address them over the next one to three years. Cisco Advanced Security Services can help you with this analysis, strategy, and implementation work.

The three foundational Pillars are:

1. Zero Trust Platform
2. Security Automation and Orchestration
3. Security Visibility and Analytics

These Zero Trust Foundational Pillars work great whether you leverage the CIS 20, NIST 800, or the ISO 27000 family cybersecurity frameworks. A few key things you need for all of them include:

◈ Segmentation, Priviledge Escalation Monitoring, and Multi Factor Authentication
◈ Inventory of your hardware and software plus application flows
◈ What are your key risks (threats, brand image, fines, and compliance)
◈ Understand what your top 50 pieces of sensitive data are
◈ Understand where your top 50 pieces of sensitive data presently resides
◈ Who is after this information?   What are their capabilities

A quick high level overview of the 3 foundataional pillars based on the information from Forrester Research: :

1. Zero Trust Platform
• Data security, which is ultimately a technology solution
• Managing the data, categorizing and developing data classification schemas, and encrypting data both at rest and in transit
2. Security Automation, Orchestration Security, and Risk leadership to leverage and use tools and technologies that enable automation and orchestration across the enterprise.
• The ability to have positive command and control of the many components that are used as part of the Zero Trust strategy.
3. Security Visibility and Analytics
• You can’t combat a threat you can’t see or understand. Tools such as traditional security information management (SIM), more-advanced security analytics platforms, security user behavior analytics (SUBA), and other analytics systems enable security professionals to know and comprehend what’s taking place in the network.
• This focus area of the extended Zero Trust ecosystem helps with the ability of a tool, platform, or system to empower the security analyst to accurately observe threats that are present and orient defenses more intelligently.

Do a workshop with Cisco Advanced Services so you understand what the gaps are, how to best address them, and prioritize your work.  This end to end approach will help you address your key use cases to get the outcomes you need addressed.

Be sure to take into consideration the Core principles that make up Zero Trust:

1. Identify and Catalog your Sensitive Data
2. Map the data flows of your sensitive data
3. Architect your Zero Trust network
4. Create your automated rule base
5. Continuously monitor your trusted ecosystem

We have the product and services experience to help you determine a practical systems approach to Zero Trust  Reach out to your Cisco Security Services team so we can help guide your through this.

Continue reading

How Antifragile Systems of Trust Can Strengthen Blockchain Initiatives

What doesn’t kill us makes us stronger. We resist gravity, and our muscles become stronger. We negotiate conflicts, and our emotional intelligence increases. But what if this also applied to IT systems? The next generation of networked systems could autonomously improve themselves as outside forces threaten them.

I and the rest of the Cisco Innovation team already believe blockchain is going to play an important role in securing data, but blockchain could also help entire networks learn to better protect themselves and intervene more quickly.

“Antifragility is beyond resilience or robustness. The resilient resists shocks and stays the same; the antifragile gets better. … The antifragile loves randomness and uncertainty, which also means — crucially — a love of errors, a certain class of errors. Antifragility has a singular property of allowing us to deal with the unknown, to do things without understanding them — and do them well. Let me be more aggressive: We are largely better at doing than we are at thinking, thanks to antifragility. I’d rather be dumb and antifragile than extremely smart and fragile, any time.”

We could take a lesson from this idea to design decentralized antifragile systems — whether they are business networks, applications or infrastructure — that become stronger the more people or machines try to break them. Blockchain technology offers specific advantages that all enterprise networks require — one single version of the truth, data immutability and automated processes among them. Combining blockchain technology with machine learning algorithms would allow multi-stakeholder systems to continually improve themselves, better react to challenges in the future and perhaps even anticipate potential problems.

I wrote an article last year about FCAPS, blockchain and trust that is relevant here. Anyone who works in telecommunications is at least tangentially familiar with the ISO/OSI and ITU network management models that shape our processes for fault, configuration, account, performance and security management — FCAPS.

My theory is that setting up machine learning algorithms to constantly measure specific parameters and adjust themselves accordingly could allow those systems to learn new solutions to preserve their operation levels. You would break systems of trust down into key components and use tight feedback loops to record and respond to specific parameters. Those tightly defined parameters could evolve over time. In a multi-stakeholder ecosystem, each stakeholder would extend insights to the collective. So in essence, there is no one system of trust but a series of trusted domains, each independently serving a specific, special function.

Let’s say your system is designed to move data from one point to another point with some certain level of efficiency. Then a configuration change happens — perhaps it’s a simple user error or a malicious configuration change. The system could learn from that instance so that the next time such a change happens, the system can automatically correct it or alert an admin to the issue that would cause it to move out of spec. Now imagine these changes as a collection of events across multiple enterprise domains.

Extremely few environments today have only one vendor as the source of their technology. Just one data center will have multiple vendors involved, and you have to be sure that this heterogeneous system runs efficiently across all providers. Let’s say your business application spans multiple clouds — how do you ensure your product delivers the quality your customers expect if something happens to one of your service providers?

Some of these issues can be learned and prevented in the future to create more robust, responsive systems. This article proposing a manifesto for antifragile software from the University of Bologna is also interesting. Inspired by the Agile Manifesto, the researchers have laid out a common framework for making software that is antifragile; the principles could also be applied to making antifragile networks. The Antifragile Software Manifesto is all about building networks of trust and continuous improvement. Here are some highlights:

◈ Our highest priority is to satisfy the customer by building a non-linear, proactive, and self-adaptive system.
◈ All stakeholders, and the broader environment, lead the antifragile organization.
◈ Continuous exposure to faults and automatic fixing is the primary measure.
◈ An antifragile organization promotes a context aware environment. The stakeholders should be able to maintain a system indefinitely.

Sounds like enterprise blockchain, right? We are striving to create networks that are decentralized and self-adapting, with many stakeholders and built-in redundancy, where all participants have skin in the game but no one actor can sabotage it. Adding machine-learning algorithms that constantly improve these enterprise blockchain networks will even better protect the data and stakeholders involved.

For our enterprise clients, confidence and privacy of data are understandably very important matters. Sustaining these states can be accomplished using selective disclosure. You define who gets to see what information as you write data to a blockchain network. Policies should be set up in the network so data being posted is always compliant with the governance models all parties have agreed upon. Your users’ roles and responsibilities within the network will grant them the according degrees of access to information. So an auditor could be granted access to all information, but a data entry specialist would have only access to certain areas.

The next level would be creating a data overlay specifically for the purpose of providing algorithms access to ecosystem-attested data — data that is compliant, shared correctly and of high value. In theory, machine-learning algorithms may produce higher quality insights for antifragility as ecosystem-wide observations are shared. Infrastructure is a perfect example of a complicated system with many stakeholders and massive amounts of high-quality data for a machine-learning algorithm to munch on. Putting the blockchain in between infrastructure and machine learning would potentially give us the ability to increase quality for all users while maintaining control of data access and preventing tampering.

If you’d like to read more about Cisco’s take on blockchain, you will enjoy our whitepaper. And feel free to leave a comment if you have any thoughts to add on how blockchain could be integrated with machine learning to create smarter systems.

Continue reading

The Five Focus Areas for 5G Security Innovation and Thought Leadership

5G brings the promise of new revenue opportunities for service providers. Service providers will be able to offer new differentiated services and capabilities, connecting customers to multi-cloud services and applications with specific KPIs.

To help service providers in the delivery of 5G, Cisco’s Cloud-to-Client approach unifies multi-vendor solutions into a single, standards-based architecture and spans across multi-cloud, IP routing, 5G core, service edge, access networks, Internet of Things (IoT), and security.

New revenue opportunities are infinite, so are security threats. Many IoT services will utilize new 5G air interfaces. Networks will be more distributed leading to a surge in entry points for more destructive threats, and new transient or moving threat boundaries.

To prepare service providers for these looming security challenges, Cisco offers a comprehensive security approach for the optimal deployment and consumption of 5G services, revealed in detail in our 5G security innovation with Cisco white paper.

5G Security Risks

5G will increase in wireless capacity by 1,000 times and connect 7 billion people and 7 trillion “things”, estimates a joint initiative between EU Commission and European ICT.

This massive throughput connectivity and capabilities in 5G require a major network architectural change, from radio access to the core. It bridges wireless and wireline networks through an evolving architecture which can involve network slicing, Control and User Plane Separation (CUPS), Mobile Edge Computing (MEC), just to name a few of the changes to the network. As the network changes at the same time, new challenges and threats will come about.

As we move into the 5G era we are also seeing more sophisticated attacks. Gartner believes half of the malware next year will use some encryption to hide malware and organizations today do not have a solution for this.

In my view, 5G’s evolving architectural nature and an expanding threat surface call for an integrated end-to-end approach to cybersecurity. Our security innovations based on visibility (even in encrypted traffic) and control for the entire 5G network, up to all applications, can provide a secure delivery of new cases with service assurance.

Intrinsic Security is Key

In the world of 5G, traditional siloed security and add-on edge appliances have limitations, are complicated and costly. Security today does not interoperate enough with the network and there will be gaps if we follow the same approach with 5G.

Cisco’s security innovation is holistic and intrinsic to the network. Leveraging the network functions themselves for visibility and fast threat identification, segmentation to reduce the attack surface and the impact of an attack; threat protection to stop a breach across multiple points of the network; and support from our threat intelligence Talos team.

5G Security Innovation with Cisco

5G security needs an integrated approach to deal with close to 20 billion threats per day. With visibility and control from end-to-end, Cisco’s full suite of 5G solutions and end-to-end security architecture can help service providers in the Asia-Pacific region stop threats at the edge, protect users wherever they work, control who gets on the network, simplify network segmentation, and find and contain problems fast.

Continue reading