CCSP Questions and Answers

Hardened perimeter devices are more useful at attenuating the risk of external attack.

A cloud customer utilizing proprietary APIs or services from one cloud provider that are unlikely to be available from another cloud provider will most negatively impact portability.

The Uptime Institute publishes the most commonly used and widely known standard on data center tiers and topologies. It is based on a series of four tiers, with each progressive increase in number representing more stringent, reliable, and redundant systems for security, connectivity, fault tolerance, redundancy, and cooling.

Potential clients are not served by SOC Type 1 audits. A Type 2 or Type 3 report would be appropriate for potential clients. SOC Type 1 reports are intended for restricted use, where only the service organization itself, current clients, or auditors would have access to them.

Object storage runs on its own independent systems, which have their own redundancy and distribution. To ensure data consistency, sufficient time is needed for objects to fully replicate to all potential locations before being accessed. Object storage is optimized for high availability and will not be any less reliable than any other virtual machine within a cloud environment. It is hosted on a separate system that does not have dependencies in local host servers for access control, and it is optimized for files of all different sizes and uses.

Local, physical access in a data center is done via KVM (keyboard, video, mouse) switches.

JavaScript Object Notation (JSON) and Extensible Markup Language (XML) are the most commonly used data formats for the Representational State Transfer (REST) API, and are typically implemented with caching for increased scalability and performance.

An intrusion detection system (IDS) is implemented to watch network traffic and operations, using predefined criteria or signatures, and alert administrators if anything suspect is found. An intrusion prevention system (IPS) is similar to an IDS but actually takes action against suspect traffic, whereas an IDS just alerts when it finds anything suspect. A firewall works at the network level and only takes into account IP addresses, ports, and protocols; it does not inspect the traffic for patterns or content. A web application firewall (WAF) works at the application layer and provides additional security via proxying, filtering service requests, or blocking based on additional factors such as the client and requests.

Jurisdiction will dictate much of the compliance and audit requirements for a data center. Although all the aspects listed are very important to security, from a strict compliance perspective, jurisdiction is the most important. Personnel access, natural disasters, and utility access are all important operational considerations for selecting a data center location, but they are not related to compliance issues like jurisdiction is.

Broad network access to cloud services, although it is an integral aspect of cloud computing, would not being a specific benefit to an organization with cyclical business needs. The other options would allow for lower costs during periods of low usage as well as provide the ability to expand services quickly and easily when needed for peak periods. Measured service allows a cloud customer to only use the resources it needs at the time, and resource pooling allows a cloud customer to access resources as needed. On-demand self-service enables the cloud customer to change its provisioned resources on its own, without the need to interact with the staff from the cloud provider.

During the data-in-use state, the information has already been accessed from storage and transmitted to the service, so reliance on a technology such as digital signatures is imperative to ensure security and complement the security methods used during previous states. Data in transit relies on technologies such as TLS to encrypt network transmission of packets for security. Data at rest primarily uses encryption for stored file objects. Archived data would be the same as data at rest.

A cross-site request forgery (CSRF) attack forces a client that a user has used to authenticate to an application to send forged requests under the user's own credentials to execute commands and requests that the application thinks are coming from a trusted client and user. Although this type of attack cannot be used to steal data directly because the attacker has no way of seeing the results of the commands, it does open other ways to compromise an application. Missing function-level access control exists where an application only checks for authorization during the initial login process and does not further validate with each function call. Cross-site scripting occurs when an attacker is able to send untrusted data to a user's browser without going through validation processes. An injection attack is where a malicious actor sends commands or other arbitrary data through input and data fields with the intent of having the application or system execute the code as part of its normal processing and queries.

Within IaaS, where the cloud customer is responsible for everything beyond the physical network, the number of users on a system would not be a factor in billing or service charges. The core cloud services for IaaS are based on the memory, storage, and CPU requirements of the cloud customer. Because the cloud customer with IaaS is responsible for its own images and deployments, these components comprise the basis of its cloud provisioning and measured services billing.

The cloud service integrator role is responsible for connecting and integrating existing services and applications with cloud-based services.A cloud service administrator is responsible for testing, monitoring, and securing cloud services, as well as providing usage reporting and dealing with service problems. The cloud service user is someone who consumes cloud services. The cloud service business manager is responsible for overseeing the billing, auditing, and purchasing of cloud services.

The goal with any cloud-hosting setup is for the cloud customer to be able to perform most or all its functions for service provisioning and configuration without any need for support from or interaction with the cloud provider beyond the automated tools provided. To fulfill the tenants of on-demand self-service, required interaction with the cloud provider--either half time, full time, or a commensurate amount of time based on the contract--would be in opposition to a cloud's intended use. As such, these answers are incorrect.

Incident response is the process through which security or operational issues are handled, including and coordination with and communication to the appropriate stakeholders. None of the other terms provided is the correct response.

Traditional security controls would not be able to restrict a user from printing something that they have the ability to access and read, but IRM solutions would allow for such a restriction. If a user has permissions to read a file, he can also copy the file or print it under traditional controls, and the ability to modify or write will give the user the ability to delete.

Dynamic resource scheduling (DRS) is used within all clustering systems as the method for clusters to provide high availability, scaling, management, and workload distribution and balancing of jobs and processes. From a physical infrastructure perspective, DRS is used to balance compute loads between physical hosts in a cloud to maintain the desired thresholds and limits on the physical hosts.

Vendor recommendations would not be pertinent to the gap analysis after an audit. Although vendor recommendations will typically play a role in the development of corporate policies or contractual requirements, they are not required. Regulations, corporate policy, and contractual requirements all determine the expected or mandated controls in place on a system.

GitHub is a software development platform that serves as a code repository and versioning system. It is solely used for software development and would not be appropriate for applying baselines to systems. Puppet is an open-source configuration management tool that runs on many platforms and can be used to apply and maintain baselines. The Software Center Configuration Manager (SCCM) was developed by Microsoft for managing systems across large groups of servers. Chef is also a system for maintaining large groups of systems throughout an enterprise.

To monitor data in use, the DLP solution's optimal location would be on the user's client or workstation, where the data would be used or processed, and where it would be most vulnerable to access or exposure. The network perimeter is most appropriate for data in transit, and an application server would serve as middle stage between data at rest and data in use, but is a less correct answer than a user's client. A database server would be an example of a location appropriate for monitoring data at rest.

The management plane will have broad administrative access to all host systems throughout an environment; as such, it represents the most pressing security concerns. A compromise of the management plane can directly lead to compromises of any other systems within the environment. Although hypervisors represent a significant security concern to an environment because their compromise would expose any virtual systems hosted within them, the management plane is a better choice in this case because it controls multiple hypervisors. Encryption and object storage both represent lower-level security concerns.

The National Fire Protection Association (NFPA) publishes a broad range of fire safety and design standards for many different types of facilities. Building Industry Consulting Services International (BICSI) issues certifications for data center cabling. The Uptime Institute publishes the most widely known and used standard for data center topologies and tiers. The International Data Center Authority (IDCA) offers the Infinity Paradigm, which takes a macro-level approach to data center design.

A reservation guarantees to a cloud customer that they will have access to a minimal level of resources to run their systems, which will help mitigate against DoS attacks or systems that consume high levels of resources. A limit refers to the enforcement of a maximum level of resources that can be consumed by or allocated to a cloud customer, service, or system. Both guarantee and assurance are terms that sound similar to reservation, but they are not correct choices.

If the CMB is receiving numerous change requests to the point where the amount of requests would drop by modifying the baseline, then that is a good reason to change the baseline. None of the other reasons should involve the baseline at all.

Disaster recovery and business continuity are vital concerns with availability. If data is destroyed or compromised, having regular backup systems in place as well as being able to perform disaster recovery in the event of a major or widespread problem allows operations to continue with an acceptable loss of time and data to management. This also ensures that sensitive data is protected and persisted in the event of the loss or corruption of data systems or physical storage systems.

Interoperability is ability to easily move between cloud providers, by either moving or reusing components and services. This can pertain to any cloud deployment model, and it gives organizations the ability to constantly evaluate costs and services as well as move their business to another cloud provider as needed or desired. Portability relates to the wholesale moving of services from one cloud provider to another, not necessarily the reuse of components or services for other purposes. Although resiliency is not an official concept within cloud computing, it certainly would be found throughout other topics such as elasticity, auto-scaling, and resource pooling. Scalability pertains to changing resource allocations to a service to meet current demand, either upward or downward in scope.

ExplanationCross-site scripting (XSS) is an attack where a malicious actor is able to send untrusted data to a user's browser without going through any validation or sanitization processes, or perhaps the code is not properly escaped from processing by the browser. The code is then executed on the user's browser with their own access and permissions, allowing the attacker to redirect the user's web traffic, steal data from their session, or potentially access information on the user's own computer that their browser has the ability to access. Missing function-level access control exists where an application only checks for authorization during the initial login process and does not further validate with each function call. An injection attack is where a malicious actor sends commands or other arbitrary data through input and data fields with the intent of having the application or system execute the code as part of its normal processing and queries. Cross-site request forgery occurs when an attack forces an authenticated user to send forged requests to an application running under their own access and credentials.

The store phase occurs immediately after the create phase, and as data is committed to storage structures, the first opportunity for security controls to be implemented is realized. During the create phase, the data is not yet part of a system where security controls can be applied, and although the use and share phases also entail the application of security controls, they are not the first phase where the process occurs.

A bastion host is a server that is fully exposed to the public Internet, but is extremely hardened to prevent attacks and is usually dedicated for a specific application or usage; it is not something that will serve multiple purposes. This singular focus allows for much more stringent security hardening and monitoring.

With SaaS providing a fully functioning application that is managed and maintained by the cloud provider, cloud customers incur the least amount of support responsibilities themselves of any service category.

Best practices and industry standards dictate that a BCDR solution should be tested at least once a year, though specific regulatory requirements may dictate more regular testing. The BCDR plan should also be tested whenever a major modification to a system occurs.

With the distributed nature of cloud environments, the foremost challenge for data discovery is awareness of the location of data and keeping track of it during the constant motion of cloud storage systems.

With software-defined networking (SDN), the filtering of network traffic is separated from the forwarding of network traffic so that it can be independently administered.

Reproducibility is the measure of how easy it is to reproduce and successful use an exploit. Scoring within the DREAD model ranges from 0, signifying a nearly impossibly exploit, up to 10, which signifies something that anyone from a simple function call could exploit, such as a URL.

Perhaps the biggest concern for any user is having their personal and sensitive information disclosed by an application. There are many aspects of an application to consider with security and protecting this information, and it is very difficult for any application to fully ensure security from start to finish. The obvious focus is on security within the application itself, as well as protecting and storing the data.

Upon successful user authentication, the identity provider gives information about the user to the relying party that it needs to make authorization decisions for granting access as well as the level of access needed.

In order to obtain and comply with certifications, independent external audits must be performed and satisfied. Although some testing of certification controls can be part of an internal audit, they will not satisfy requirements.

Regulatory requirements are those imposed upon businesses and their operations either by law, regulation, policy, or standards and guidelines. These requirements are specific either to the locality in which the company or application is based or to the specific nature of the data and transactions conducted.

It is imperative that an application perform checks when each function or portion of the application is accessed, to ensure that the user is properly authorized to access it. Without continual checks each time a function is accessed, an attacker could forge requests to access portions of the application where authorization has not been granted.

NIST Special Publications 800-145 is titled "The NIST Definition of Cloud Computing" and contains definitions and explanations of core cloud concepts and components.

Rapid elasticity allows a cloud environment to automatically add or remove resources to or from a system or application based on its current demands. Whereas a traditional data center model would require standby hardware and substantial effort to add resources in response to load increases, a cloud environment can easily and rapidly expand to meet resources demands, so long as the application is properly implemented for it.

DNSSEC relates to the integrity of DNS resolutions and the prevention of spoofing or redirection, and does not pertain to the actual security of transmissions or the protection of data.

The SOAP protocol does not support caching, whereas the REST API does.

The guidelines from ASHRAE establish 40-60 percent relative humidity as optimal for a data center.

Due to the lack of an adequate privacy law or protection at the federal level in the United States, European privacy regulations generally prohibit the exporting or sharing of PII from Europe with the United States. Participation in the Safe Harbor program is voluntary on behalf of an organization, but it does require them to conform to specific requirements and policies that mirror those from the EU. Thus, organizations can fulfill requirements for data sharing and export and possibly serve customers in the EU.

The cloud service operations manager is responsible for preparing systems for the cloud, administering and monitoring services, providing audit data as requested or required, and managing inventory and assets.

The Sarbanes-Oxley (SOX) Act is not an act that pertains to privacy or IT security directly, but rather regulates accounting and financial practices used by organizations. It was passed to protect stakeholders and shareholders from improper practices and errors, and it sets forth rules for compliance, regulated and enforced by the Securities and Exchange Commission (SEC). The main influence on IT systems and operations is the requirements it sets for data retention, specifically in regard to what types of records must be preserved and for how long.

With static masking, a separate and distinct copy of the data set is created with masking in place. This is typically done through a script or other process that takes a standard data set, processes it to mask the appropriate and predefined fields, and then outputs the data set as a new one with the completed masking done.

With cloud providers hosting a number of different customers, it would be impractical for them to pursue additional certifications based on the needs of a specific customer. Cloud environments are built to a common denominator to serve the greatest number of customers, and especially within a public cloud model, it is not possible or practical for a cloud provider to alter their services for specific customer demands.

The cloud service broker is responsible for obtaining new customers, analyzing the marketplace, and securing contracts and agreements.

iSCSI is a protocol that allows for the transmission and use of SCSI commands and features over a TCP-based network. iSCSI allows systems to use block-level storage that looks and behaves as a SAN would with physical servers, but to leverage the TCP network within a virtualized environment and cloud.

An insecure direct object reference occurs when a developer has in their code a reference to something on the application side, such as a database key, the directory structure of the application, configuration information about the hosting system, or any other information that pertains to the workings of the application that should not be exposed to users or the network. Unvalidated redirects and forwards occur when an application has functions to forward users to other sites, and these functions are not properly secured to validate the data and redirect requests, allowing spoofing for malware of phishing attacks. Sensitive data exposure occurs when an application does not use sufficient encryption and other security controls to protect sensitive application data. Security misconfigurations occur when applications and systems are not properly configured or maintained in a secure manner.

The cloud service administrator is responsible for testing cloud services, monitoring services, administering security for services, providing usage reports on cloud services, and addressing problem reports

While the main focus of confidentiality revolves around technological requirements or particular security methods, an important and often overlooked aspect of safeguarding data confidentiality is appropriate and comprehensive training for those with access to it. Training should be focused on the safe handling of sensitive information overall, including best practices for network activities as well as physical security of the devices or workstations used to access the application.

Static application security testing (SAST) is conducted against offline systems with previous knowledge of them, including their source code. Live testing is not part of static testing but rather is associated with dynamic testing. Production system scanning is not appropriate because static testing is done against offline systems. Injection attempts are done with many different types of testing and are not unique to one particular type. It is therefore not the best answer to the question.

This is not a normal configuration and would not likely provide genuine benefit.

The Dynamic Host Configuration Protocol (DHCP) automatically configures network settings for a host so that these settings do not need to be configured on the host statically. Given the rapid and programmatic provisioning of resources within a cloud environment, this capability is crucial to cloud operations. Both DNS and its security-integrity extension DNSSEC provide name resolution to IP addresses, but neither is used for the configuration of network settings on a host. DCOM refers to the Distributed Component Object Model, which was developed by Microsoft as a means to request services across a network, and is not used for network configurations at all.

Structured storage involves organized and categorized data, which most closely resembles and operates like a database system would.

DRM and DLP are used for increased authentication/access control and egress monitoring, respectively, and would actually decrease portability instead of enhancing it.

Whereas IRM is used to protect a broad range of data, DRM is focused specifically on the protection of consumer media, such as publications, music, movies, and so on. IRM is used to protect general institution data, so financial records, personnel data, and security profiles would all fall under the auspices of IRM.

With a typical BCDR solution, an organization would need some number of staff to quickly travel to the location of the BCDR site to configure systems and applications for recovery. With a cloud environment, everything is done over broad network access, with no need (or even possibility) to travel to a remote site at any time.

When a security professional is considering cloud solutions for BCDR, a top concern is the jurisdiction where the cloud systems are hosted. If the jurisdiction is different from where the production systems are hosted, they may be subjected to different regulations and controls, which would make a seamless BCDR solution far more difficult.

The data retention policy covers the duration, format, technologies, protection, and accessibility of archives, but does not address the specific costs of its implementation and maintenance.

The record protocol of TLS performs the authentication and encryption of data packets, and in some cases compression as well. It does not perform any acceleration functions.

Dynamic optimization is the process through which the cloud environment is constantly maintained to ensure resources are available when and where needed, and that physical nodes do not become overloaded or near capacity, while others are underutilized.

The Simple Object Access Protocol (SOAP) does not support caching, whereas the Representational State Transfer (REST) API does. The other options are all capabilities that are either not supported by SOAP or not supported by any API and must be provided by external features.

Multitenancy is the aspect of cloud computing that involves having multiple customers and applications running within the same system and sharing the same resources. Although considerable mechanisms are in place to ensure isolation and separation, the data and applications are ultimately using shared resources. Broad network access refers to the ability to access cloud services from any location or client. Portability refers to the ability to easily move cloud services between different cloud providers, whereas elasticity refers to the capabilities of a cloud environment to add or remove services, as needed, to meet current demand.

The American Society of Heating, Refrigeration, and Air Conditioning Engineers (ASHRAE) recommends 40 to 60 percent relatively humidity for data centers. None of these options is the recommendation from ASHRAE.

Regardless of the particular cloud service used, physical security of hardware and facilities is always the sole responsibility of the cloud provider. The cloud provider may release information about their physical security policies and procedures to ensure any particular requirements of potential customers will meet their regulatory obligations. Personal security of developers and regulatory compliance are always the responsibility of the cloud customer. Responsibility for operating systems, and the auditing of them, will differ based on the cloud service category used.