Research

We realize it has been a while, but we are insanely excited to open up the next phase of the Securosis Nexus beta test. This is an open beta but we reserve the right to kick out anyone who annoys us. Getting Started Signing into the Nexus is easy. Just go to http://nexus.securosis.com, click “Sign Up” and enter “4111-1111-1111-1111” as your credit card number – sorry if that’s your real credit card number. You will then receive activation information via email.   What to Expect The Nexus is code complete but our content is far from complete. The system is completely functional, so now we need help making sure it scales beyond our internal testing. We have some starter content in there, but it is only representative of where we are headed, and this structure is temporary. We will be adding content weekly basis as we get closer to launch, and will send out occasional updates to beta testers so you know what we are up to. We are fully supporting the “Ask an Analyst” feature, which means you get free advice (well, in exchange for your help testing). But this is a (free) beta test, so we make no promises of timeliness. 🙂 We anticipate staying in test mode for 3-6 months because it will take at least that long to write all the content. Most of the material is brand new, and this isn’t merely a repository for our white papers. If you find bugs or have questions, email us at nexus@securosis.com or use the Support link. Thanks! We are really looking forward to getting more people into the system and taking it for a test drive. Share:

Some of our first customer conversations about big data and SIEM centered on how to integrate the two platforms. Several customers wanted to know how they could pull data from different existing log management and analytics systems into a big data platform. Most were told by their vendors that big data was; and they wanted to know what that integration would look like and how it would affect operations. Likely you won’t be doing the integration, but you will need to live with the design choices of your vendor. The benefit depends on their implementation choices. There are three basic models for integration of big data with SIEM: Log Management Container Some vendors have chosen to integrate big data while keeping their basic architecture: semi-relational or flat file system which supports SIEM functions, and fronts a big data cluster which handles log management tasks. We say ‘semi-relational’ because it is typically a relational platform such as Oracle or SQL Server, but stripped of many relational constructs to improve data insertion rates. SIEM’s event processing and near real-time alerts remains unchanged: event streams are processed as they arrive, and a specific subset of events and profile information stored within a relational – or proprietary flat file – database. Data stored within the big data cluster may be correlated, but normalization and enrichment is only performed at the SIEM layer. Raw events are streamed to the big data cluster for long-term storage, possibly compressed. SIEM functions may be supported by queries to reference specific data points within the big data archive but support is limited. In essence big data is used to scale event storage and accommodate events – regardless of type or format. Peer-to-Peer Like the example above, in this scenario real-time analysis is performed on the incoming event stream, and basic analysis performed in a semi-relational or flat file database. The difference here is functional rather than architectural. The two databases are truly peers – each provides half the analysis capability. The big data cluster periodically re-calculates behavioral profiles and risk scores, and shares these with SIEM’s real-time analysis component. It also processes complex activity chains and multiple events tied to specific locations, users or applications may that indicate malicious behavior. The big data cluster does a lot of heavy lifting to mine events, and shares these updated profiles with SIEM to hone policy enforcement. The big data cluster allows provides a direct view for Security Operations Centers (SOC) to run ad hoc queries on a complete set of events to look for outliers and anomalous activity. Full Integration The next option is to leverage only big data for event analysis and long term log storage. Today most of these SIEM platforms use proprietary file systems – not relational databases or big data. These proprietary systems were born of the same need to scale to accommodate more data with less insertion overhead than big data. These proprietary repositories were designed to provide clustered data management, distributing queries across multiple machines. But they are not big data – they don’t have the essential characteristics we defined earlier, and often don’t have the the 3Vs either. You will notice that both peer-to-peer and log management oriented versions use two databases; one relational and one big data. There is really no good reason to maintain a relational database alongside a big data cluster – other than the time it takes to migrate and test the migration. That aside, it is a simple engineering effort to swap out a relational platform with a big data cluster. Big data clusters can be assembled to perform ultra fast queries, or efficient large scale analysis, or leverage both types of queries on a single data set. Many relational features are irrelevant to security analytics, so they are either stripped out for performance or remain present, reducing performance. Again, there is no reason relational databases must be part of SIEM – the only impediment is the need to re-engineer the platform to swap the new cluster in. This does not exist today but expect it in the months to come. Continuing this line of thought, it is very interesting to think of ways to further optimize a SIEM system. You can run more than one big data cluster, each focused on a specific type of operation. So one cluster would run fully indexed SQL queries for fast retrieval while another might run MapReduce queries to find statistical outliers. In terms of implementation, you might choose Cassandra for its index capabilities and native compression, and Hadoop for MapReduce and large-scale storage. The graphic to the right shows this possibility. It is also possible to have one data cluster with multiple query engines running against the same data set. The choice is up to your SIEM vendor, but the low cost of data storage and processing capacity mean the performance boost even from redundant data stores is still likely to outweigh the costs of added processing. The fit for security analytics is largely conjecture, but we have seen both models scale well for various other data analyses. Standalone Those of you keeping score at home have noticed I am throwing in a fourth option: the standalone or non-integration model. Some of our readers are not actually interested in SIEM at all – they just want to collect security events and run their own reports and analysis without SIEM. It is perfectly feasible to build a standalone big data cluster for security and event analytics. Choose a platform optimized for your queries (fast, or efficient, or both if it is worth building multiple optimized clusters), the types of data to mine, and developer comfort. But understand that you will need to build a lot yourself. A wide variety of excellent tools and logging utilities are available as open source or shareware, but you will responsibility for design, organization, and writing your own analytics. Starting from scratch is not necessarily bad but all development (tools, queries, reports, etc.) will fall to your team. Should you choose to integrate with SIEM or log management, you will

FUD can be your friend when trying to get security projects funded. But it needs to be wisely used and you only have one bullet in the proverbial chamber. The folks at Prolexic just rolled out a new white paper on using FUD to make the case internally about DDoS. The paper requires registration, so I didn’t. I know all about the FUD involved in DDoS – I don’t need these guys educating me about that. So here are some really FUD-elicious reasons why business folks need to be worried about DDoS: The damage from a DDoS attack actually goes far beyond IT and can impact: Stock price and investor confidence Sales revenues and profitability Brand reputation Customer service Employee morale Search engine rankings and more How’s that for some chicken little action? I think a DDoS may clog your toilets as well, so bring the plungers. And make sure you have psychologists on call – employee morale will be in the dumpers with every incremental 10Gbps of DDoS traffic hammering your systems. And your scrubbing center can make it all better. Just ask them. Yes, I’m being a bit facetious. OK, very facetious. I can imagine investors have no faith in the Fortune 10 banks that get hammered by DDoS every day. Man, I could go on all day… Anyhow, this one was just too juicy to let pass. Now I’ll get back to doing something productive… Share:

As we resume our Network-based Malware Detection (NBMD) 2.0 series, we need to dig into the malware detection/analysis lifecycle to provide some context on where network-based malware analysis fits in, and what an NBMD device needs to integrate with to protect against advanced threats. We have already exhaustively researched the malware analysis process. The process diagram below was built as part of Malware Analysis Quant. Looking at the process, NBMD provides the analyze malware activity phase – including building the testbed, static analysis, various dynamic analysis tests, and finally packaging everything up into a malware profile. All these functions occur either on the device or in some cloud-based sandbox for analyzing malware files. That is why scalability is so important, as we discussed last time. You basically need to analyze every file that comes through because you cannot wait for an employee’s device to be compromised before starting the analysis. Some other aspects of this lifecycle bear mentioning: Ingress analysis is not enough: Detecting and blocking malware on the perimeter is a central pillar of the strategy, but no NBMD capability can be 100% accurate and catch everything. You need other controls on endpoints, supplemented with aggressive egress filtering. Intelligence drives accuracy: Malware and tactics evolve so quickly that on-device analysis techniques must evolve as well. This requires a significant and sustained investment in threat research and intelligence sharing. Before we can dig into these two points we need to point out some other relevant research on these topics for additional context. The Securosis Data Breach Triangle shows a number of opportunities to interrupt a data breach. You can either protect the data (very hard), detect and stop the exploit, or catch the data with egress filtering. Success at any one of these will stop a breach. But putting all your eggs in one basket is unwise, so work on all three. For specifics on detecting and stopping exploits, refer to our ongoing CISO’s Guide to Advanced Attackers – particularly Breaking the Kill Chain, which covers stopping an attack. Remember – even if a device is compromised, unless critical data is exfiltrated it’s not a breach. The best case is to detect the malware before it hurts anything – NBMD is very interesting technology for this – but you also need to rely heavily on your incident response process to ensure you can contain the damage. Ingress Accuracy As with most detection activities, accuracy is critical. A false positive – incorrectly flagging a file as malware – disrupts work and wastes resources investigating a malware outbreak that never happened. You need to avoid these, so put a premium on accuracy. False negatives – missing malware and letting it through – are at least as bad. So how can you verify the accuracy of an NBMD device? There is no accepted detection accuracy benchmark so you need to do some homework. Start by asking the vendor tough questions to understand their threat intelligence and threat research capabilities. Read their threat research reports and figure out whether they are on the leading edge of research, or just a fast follower using other companies’ research innovations. Malware research provides the data for malware analysis, whether on the device or in the cloud. So you need to understand the depth and breadth of a vendor’s research capability. Dig deep and understand how many researchers they have focused on malware analysis. Learn how they aggregate the millions of samples in the wild to isolate patterns using fancy terms like big data analytics. Study and understand how they turn that research into detection rules and on-device tests. You will also want to understand how the vendor shares information with the broader security research community. No one company can do it all, so you want leadership and a serious investment in research, but you also need to understand how they collaborate with other groups and what alternative data sources they leverage for analysis. For particularly advanced malware samples, do they have a process to undertake manual analysis? Be sensitive to research diversity. Many NBMD devices use the same handful of threat intelligence services to populate their devices. That makes it very difficult to get intelligence diversity to detect fast-moving advanced attacks. Make sure you check out lab tests of devices to compare accuracy. These tests are all flawed – it is just barely theoretically possible to accurately model a real-world environment using live ammunition (malware), but things would immediately change. But these tests can be helpful for an apples-to-apples device comparison. The Second Derivative As part of a proof of concept, you may also want to route your ingress traffic through 2 or 3 of these devices in monitoring mode, to test relative accuracy and scalability on real traffic. That should give you a good indication of how well the device will perform for you. Finally, leverage “The 2nd Derivative Effect (2DE)” of malware analysis. When new malware is found, profiled, and determined to be bad, there is an opportunity to inoculate all the devices in use. This involves uploading the indicators, behaviors, and rules to identify and block it to a central repository; and then distributing that intelligence back out to all devices. The network effect in action. The more devices in the network, the more likely the malware will show up somewhere to be profiled, and the better your chance of being protected before it reaches you. Not always, but it’s is as good a plan as any. It sucks to be the first company infected – you miss the attack on its way in. But everyone else in the network benefits from your misfortune. This ongoing feedback loop requires extensive automation (with clear checks and balances to reduce bad updates) to accelerate distribution of new indicators to devices in the field. Plan B (When You Are Wrong) Inevitably you will be wrong sometimes, and malware will get through your perimeter. That means you will need to rely on the other security controls in your environment. When they fail you will want to make sure you don’t get popped by the same attack

I am in the airport lounge after attending the WWDC keynote, and here are some quick thoughts on what we saw today: The biggest enhancement is iCloud Keychain. Doesn’t seem like a full replacement for 1Password/etc. (yet), but Apple’s target is people who won’t buy 1Password. Once this is built in, from the way it appears designed, it should materially help common folks with password issues. As long as they buy into the Apple ecosystem, of course. It will be very interesting to see how the activation lock feature works in the real world. Theft is rampant, and making these devices worthless will really put a dent in it, but activation locking is a tricky issue. Per-tab processes in Safari. I am very curious about whether there is more additional sandboxing (Safari already has some). My main concern these days is Flash, and that’s why I use Chrome. If either Adobe or Apple improve Flash sandboxing I will be very happy to switch back. For enterprises Apple’s focus appears to be on iOS and MDM/single sign on. I will research the new changes more. Per-app VPNs also looks quite nice, and might simplify some app wrapping that currently does this through alternate techniques. iWork in the cloud could be interesting, and looks much better than Google apps – but collaboration, secure login, and sharing will be key. Many questions on this one, and I’m sure we will know more before it goes live. I didn’t see much else. Mostly incremental, and I mainly plan to keep an eye on what happens in Safari because it is the biggest point of potential weakness. Nothing so dramatic on the defensive side as Gatekeeper and the Java lockdowns of the past year, but integrating password management is another real-world, casual user problem that hasn’t been cracked well yet. Share:

As I continue working through my reading backlog I find interesting stuff that bears comment. When the folks over at NSS Labs attempted to poke holes in the concept of security layers I got curious. Only 3% of over 606 combinations of firewall, IPS, and Endpoint Protection (EPP) actually successfully blocked their full suite of attacks? There is only limited breach prevention available: NSS looked at 606 unique combinations of security product pairs (IPS + NGFW, IPS + IPS, etc.) and only 19 combinations (3 percent) were able to successfully detect ALL exploits used in testing. This correlation of detection failures shows that attackers can easily bypass several layers of security using only a small set of exploits. Most organizations should assume they are already breached and pair preventative technologies with both breach detection and security information and event management (SIEM) solutions. No kidding. It not novel to say that exploits work in today’s environment. Instead of just guessing at optimal combination of devices (which seems to be a value proposition NSS is pushing in the market now), what about getting a feel for the incremental effectiveness of just using a firewall. And then layering in an IPS, and finally looking at endpoint protection. Does IPS really make an incremental difference? That would be useful information – we already know it is very hard to block all exploits. NSS’s analysis of why layering isn’t as effective as you might think is interesting: groupthink. Many of these products are driven by the same research engines and intelligence sources. So if a source misses all its clients miss. Clearly a recipe for failure, so diversity is still important. Rats! Dan Geer and his monoculture stuff continue to bite us in the backside. But of course diversity adds management complexity. Usually significant complexity, so you need to balance different vendors at different control layers against the administrative overhead of effectively managing everything. And a significant percentage of attacks are successful not due to innovative exploits (of the sorts NSS tests), but because of operational failures implementing the technology, keeping platforms and products patched, and enforcing secure configurations. Photo credit: “groupthink” originally uploaded by khrawlings Share:

From Share and share alike? Not Quite, by Mike Mimoso at Threatpost: “With retail, the challenge is that most of the companies we share with are direct competitors,” Phillips said. “From a security perspective, you have to get over that and share because we’re all facing the same challenges. There’s no way any of us will win the war on our own.” If sharing information on attacks provides your competitors a business advantage, you have serious issues unrelated to security. Share:

After a week of travel I am finally working through my reading list, and got around to RSnake’s awesome “Talk with a Black Hat” series. Check out Part 1, Part 2 and Part 3. He takes us behind the curtain – but instead of discussing impact, which your fraud and loss group can tell you – he documents tactics being used against us all the time. At the beginning of Part 1, RSnake tackles the ethical issues of communicating with and learning from black hats. I never saw this as an issue, but if you did, just read his explanation and get over it: I think it is incredibly important for security experts to have open dialogues with the blackhat community. It’s not at all dissimilar to police officers talking with drug dealers on a regular basis as part of their job: if you don’t know your adversary you are almost certainly doomed to failure. Right. A few interesting tidbits from Part 1, including “The whole blackhat market has moved from manual spreading to fully automated software.” And that this fellow’s motivation was pretty clear: “Money. I found it funny how watching tv and typing on my laptop would earn me a hard worker’s monthly wage in a few hours. [It was] too easy in fact.” And the lowest hanging fruit for an attack. Yup, pr0n sites. Now to discuss my personal favourite: porn sites. One reason why this is so easy: The admins don’t check to see what the adverts redirect to. Upload an ad of a well-endowed girl typing on Facebook, someone clicks, it does a drive by download again. But this is where it’s different: if you want extra details (for extortion if they’re a business man) you can use SET to get the actual Facebook details which, again, can be used in social engineering. There is similarly awesome perspective on monetizing DDoS (which clearly means it is not going away anytime soon), and that was only in Part 1. Part 2 and 3 are also great, but you should read them yourself to learn about your adversaries. And to leave you with some wisdom about mindsets: Q: What kind of people tend to want to buy access to your botnet and/or what do you think they use it for? A: Some people say governments use it, rivals in business. To be honest, I don’t care. If you pay you get a service. Simple. Simple. Yup, very simple. Photo credit: “Charles F Esolda” originally uploaded by angus mcdiarmid Share:

So why are we looking at big data, and what problems can we expect it to solve that we couldn’t before? Most SIEM platforms struggle to keep up with emerging needs for two reasons. The first is that threat data does not come neatly packaged from traditional sources, such as syslog and netflow events. There are many different types of data, data feeds, documents, and communications protocols that contain diverse clues to a data breaches or ongoing attacks. We see clear demand to analyze a broader data set in order hopes of detecting advanced attacks. The second issue is that many types of analysis, correlation, and enrichment are computationally demanding. Much like traditional multi-dimensional data analysis platforms, crunching the data takes horsepower. More data is being generated; add more types of data we want, and multiply that by additional analysess – and you get a giant gap between what you need to do and what you can presently do. Our last post considered what big data is and how NoSQL database architectures inherently address several of the SIEM pain points. In fact, the 3Vs (Volume, Velocity, & Variety) of big data coincide closely with three of the main problems faced by SIEM systems today: scalability, performance, and effectiveness. This is why big data is such an important advancement for SIEM. Volume and velocity problems are addressed by clustering systems to divide load across many commodity servers, and variability through the inherent flexibility of big data / NoSQL. But of course there is more to it. Analysis: Looking at More Two of the most serious problems with current SIEM solutions are that they struggle with the amount of data to be managed, and they cannot deal with the “data velocity” of near-real-time events. Additionally, they need to accept and parse new and diverse data types to support new types of analysis. There are many different types of event data, any of which might contain clues to security threats. Common data types include: Human-readable data: There is a great deal of data which humans can process easily, but which is much more difficult for machines – including blog comments and Twitter feeds. Tweets, discussion fora, Facebook posts, and other types of social media are all valuable for threat intelligence. Some attacks are coordinated in fora, which means companies want to monitor these fora for warnings of possible or imminent attacks, and perhaps even details of the attacks. Some botnet command and control (C&C) communications occur through social media, so there is potential to detect infected machines through this traffic. Telemetry feeds: Cell phone geolocation, lists of sites serving malware, mobile device IDs, HR feeds of employee status, and dozens of other real-time data feeds denote changes in status, behavior, and risk profiles. Some of these feeds are analyzed as the stream of events is captured, while others are collected and analyzed for new behaviors. There are many different use cases but security practitioners, observing how effectively retail organizations are able to predict customer buying behavior, are seeking the same insight into threats. Financial data: We were surprised to learn how many customers use financial data purchased from third parties to help detect fraud. The use cases we heard centered around SIEM for external attacks against web services, but they were also analyzing financial and buying history to predict misuse and account compromise. Contextual data: This is anything that makes other data more meaningful. Contextual data might indicate automated processes rather than human behavior – a too-fast series of web requests, for example, might indicate a bot rather than a human customer. Contextual data also includes risk scores generated by arbitrary analysis of metadata, and detection of odd or inappropriate series of actions. Some is simply collected from a raw event source while other data is derived through analysis. As we improve our understanding of where to look for attack and breach cluse, we will leverage new sources of data and examine them in new ways. SIEM generates some contextual data today, but collection of a broader variety of data enables better analysis and enrichment. Identity and Personas: Today many SIEMs link with directory services to identify users. The goal is to link a human user to their account name. With cloud services, mobile devices, distributed identity stores, identity certificates, and two-factor identity schemes, it has become much harder to link human beings to account activity. As authentication and authorization facilities become more complex, SIEM must connect to and analyze more and different identity stores and logs. Network Data: Some of you are saying “What? I thought all SIEMs looked at network flow data!” Actually, some do but others don’t. Some collect and alert on specific known threats, but only a tiny portion of that passes down the wire. Cheap storage makes it feasible to store more network events and perform behavioral computation on general network trends, service usage, and other pre-computed aggregate views of network traffic. In the future we may be able to include all data. Each of these examples demonstrates what will be possible in the short term. In the long term we may record any and all useful or interesting data. If we can link in data sets that provide a different views or help us make better decisions, we will. We already collect many of these data types, but we have been missing the infrastructure to analyze them meaningfully. Analysis: Doing It Better One limitation of many SIEM platforms is their dependence on relational databases. Even if you strip away relational constructs that limit insertion performance, they still rely on a SQL language with traditional language processors. The fundamental relational database architecture was designed and optimized for relational queries. Flexibility is severely limited by SQL – statements always include FROM and WHERE clauses, and we have a limited number of comparison operators for searching. At a high level we may have Java support, but the actual queries still devolve down to SQL statements. SQL may be a trusty

So why are we talking about this? Because APIs are becoming the de facto service interface – not only for cloud and mobile, but for just about every type of service. The need for security around these APIs is growing, which is why we have seen a rush of acquisitions to fill security product gaps. In what felt like a couple weeks Axway acquired Vordel, CA acquired Layer7, and Intel acquired Mashery. The acquirers all stated these steps were to accommodate security requirements stemming from steady adoption of APIs and associated web services. Our goal for this paper is to help you understand the challenges of securing APIs and to evaluate technology alternatives so you can make informed decisions about current trends in the market. We will start our discussion by mentioning what’s at stake, which should show why certain features are necessary. API gateways have a grand and audacious goal: enablement. Getting developers the tools, data, and functionality they need to realize the mobile, social, cloud and other use cases the enterprise wants to deliver. There is a tremendous amount of innovation in these spaces today, and the business goal is get to market ASAP. At the same time, security is not a nice-to-have – it’s a hard requirement. After all, the value of mobile, social, and cloud applications is in mixing enterprise functionality inside and outside the enterprise. And riding along is an interesting mix of user personas: customers, employees, and corporate identities, all mingling together in the same pool. API gateways must implement real security policies and protocols to protect enterprise services, brands, and identity. This research paper will examine current requirements and technical trends in API security. API gateways are not sexy. They do not generate headlines like cloud, mobile, and big data. But the APIs are the convergence point for all these trends, and the crux of IT innovation today. We all know cloud services scale almost too well to be real, at a price that seems to good to be true. But the APIs are part of what makes them so scalable and cheap. Of course open, API-driven, multi-tenant environments bring new risks along with their new potentials. As Netflix security architect Jason Chan says, securing your app on Amazon Cloud is like rock climbing – Amazon gives you a rope and belays you, but you are on the rock face. You are the one at risk. How do you manage that risk? API gateways play a central role in limiting the cloud’s attack surface and centralizing policy enforcement. Mobile apps pose similar risks in an entirely different technical environment. There is endless amount hype about iOS and Android security. But where are the breaches? On the server side. Why? Because attackers are pragmatic, and that’s where the data is. Mobile apps have vulnerabilities that attackers can go after one by one, but a breach of the server-side APIs exposes the whole enterprise enchilada. Say it with me in your best Taco Bell Chihuahua accent: The whole enchilada! Like cloud applications, API gateways need to reduce the enterprise’s risk by limiting attack surface. And mobile apps use web services differently than other enterprise applications, communications are mostly asynchronous, and the identity tokens are different too – expect to see less SAML or proprietary SSO, and more OAuth and OpenID Connect. API gateways address the challenges raised by these new protocols and interactions. APIs are an enabling technology, linking new and old applications together into a unified service model. But while cloud, mobile, and other innovations drive radical changes in the data center, one thing remains the same: the speed at which business wants to deploy new services. Fast. Faster! Yesterday, if possible. This makes developer enablement supremely important, and is why we need to weave security into the fabric of development – if it is not integrated at a fundamental level, security gets be removed as an impediment to shipping. The royal road is to things that make it easy for developers to understand how to build and deploy an app, grok the interfaces and data, and quickly provision developers and their app users to login – this is how IT shops are organizing teams, projects, and tech stacks. The DMZ has gone the way of the dodo. API gateways are about enabling developers to build cloud, mobile, and social apps on enterprise data, layered over existing IT systems. Third-party cloud services, mobile devices, and work-from-anywhere employees have destroyed (or at least completely circumvented) the corporate IT ‘perimeter’ – the ‘edge’ of the network has so many holes it no longer forms a meaningful boundary. And this trend, fueled by the need to connect in-house and third-party services, is driving the new model. API gateways curate APIs, provision access to users and developers, and facilitate key management. For security this is the place to focus – to centralize policy enforcement, implement enterprise protocols and standards, and manage the attack surface. This paper will explore the following API gateway concepts in detail. The content will be developed and posted to the Securosis blog for vetting by the developer and security community. As always, we welcome your feedback – both positive and negative. Our preliminary outline is: Access Provisioning: We will discuss developer access provisioning, streamlining access to tools and server support, user and administrator provisioning, policy provisioning and management, and audit trails to figure out who did what. Developer Tools: We will discuss how to maintain and manage exposed services, a way to catalogue services, client integration, build processes, and deployment support. Key Management: This post will discuss creating keys, setting up a key management service, key and certificate verification, and finally the key management lifecycle (creation, revocation, rotation/updating). Implementation: Here we get into the meat of this series. We will discuss exposing APIs and parameters, URL whitelisting, proper parameter parsing, and some deployment options that effect security. Buyers Guide: We will wrap this series with a brief buyers guide to help you understand the differences between implementations, as well as key considerations when establishing your evaluation priorities. We will also cover