devops

The ITIL service life cycle consists of service strategy, service design, service _, service operation, and continual service improvement

transition

There is an open standard introduced in Chapter 7 which defines a standard method for reporting and comparing the performance of software applications in computing. The purpose of this standard is to convert measurements into insights about user satisfaction, by specifying a uniform way to analyze and report on the degree to which measured performance meets user expectations. This open standard is called _

apdex

_ persist for months or years, have legal standing, and are designed for security purposes. _ persist for times measured in days (or less) and are designed to support operational and development needs. We should not get confused by these two terms

audit trails; logs

_ has two elements that answer the following questions: Who is trying to access or modify information and do they have the right to perform the operation they requested?

authorization

With cloud elasticity, the pay-as-you-go model, and the ease of provisioning new virtual hardware, capacity planning is becoming more about runtime monitoring and _ rather than planning for purchasing hardware.

autoscaling

One method of viewing the relationship between DevOps and ITIL is that DevOps provides continuous _ of the various ITIL services rather than requiring those services to be packaged into a major release.

delivery

Which is not a weakness of agentless monitoring? - security - network traffic - intrusive monitoring - deployment and maintenance effort

deployment and maintenance effort

A _ consists of the steps that are taken between a developer committing code and the code actually being promoted into normal production, while ensuring high quality.

deployment pipeline

Which could be a source for dependability problems of a system? - Mixed versions of third party software and libraries used by a service - Mapping among architectural elements - Possible failure of an instance of a service - Large amount of cross-team coordination

- Possible failure of an instance or a service

If a recovery solution takes 8 minutes to access the backup in a separate datacenter and another 6 minutes to instantiate new servers using the backed-up data, the RTO is _ minutes

14

Computational resources also need to be protected. This is the _ in CIA acronym.

A

DevOps can have a huge impact on - Software structure - Team structure - Traditional ways of conducting operations - All of the above

All of the above

Which is an example of post-deployment error? - version dependencies existing among the components - a developer connected test code to a production database - parameters for dependent systems were set incorrectly - all of the above

All of the above

Which of the following methods can detect software failures? - use a monitoring software to perform health checks on the system - use a special agent inside the system to perform the monitoring - let the system itself detect problems and report them - all of the above can be used

All of the above

Netflix Eureka is an example of a cloud registry services that acts as a _ server

DNS

A feature toggle can be implemented by a "try-catch" statement with a global Boolean variable.

False

Amazon's rules for teams require all teams to expose their data and functionality by reading each other team's datastore directly to avoid time consuming inter-team communication.

False

CloudWatch and HDFS are two examples of tools that manage the items inside a virtual machine. They can replace a version of a piece of software inside a VM and ensure that configuration settings conform to a specification.

False

Deployment scripts is different from application code. Therefore, errors in deployment scripts such as misconfigurations will only affect the deployment process but not the application itself.

False

DevOps and continuous deployment require the implementation part of service transition to be highly automated so it can deal with high-frequency transition and provide better quality control.

False

Encapsulating either the affected portions of a likely change or the interactions that might cause ripple effects of a change is a good method for making a service run faster.

False

For no reason we should deploy some services of a system to one environment such as VMware and other services to a different environment such as Amazon EC2.

False

Git is a distributed version control system. Using its push command, changes from the server are synchronized with the local repository.

False

If a web page is implemented by a collection of microservices, then its scalability should be managed by the infrastructure of the entire web system, not each individual service, and should be independent to its SLA.

False

If the system is actively contributing to the data being monitored then the monitoring is non-intrusive and does not affect the system design. If the system is not actively contributing to the data being monitored then the monitoring is intrusive and affects the system design.

False

In order to remove the inter-team coordination, a design decision of system architecture should be made to allow each component to be allocated to a single (virtual) machine or container but avoid multiple components to be allocated to a single (virtual) machine.

False

One advantage of adopting the microservice architecture is it reduces monitoring effort dramatically by less logging of inter-service communication.

False

One difference between ITIL's service transition and the DevOps approach is that DevOps assumes fairly large release packages where careful planning, change management, and so on are feasible—in contrast to the high-frequency small releases encountered in typical ITIL scenarios.

False

Releasing a new system or version of an existing system to customers is one of the most sensitive steps in the software development cycle. This is only true the system or version is for external distribution or is used directly by consumers. It does not cause any concern if the system or version is strictly for internal use.

False

Tear down an environment when it is not used is a way to improve the recoverability of deployment pipeline.

False

Throughput has a single-user or client focus, whereas latency provides a system-wide measure involving all of the users. Therefore, high throughput may or may not be related to low latency.

False

Using WMI to monitor and manage Microsoft Windows and other Microsoft software is agent-based monitoring.

False

We should prevent likely changes from rippling through a service by reducing its modules that implement the interfaces.

False

When a user issues a command to create a VM, this user must also decide on which physical machine to create the instance

False

What are the three possible types or sources of logical inconsistency resulted in deployment? - inconsistency between a service and data kept in a database - inconsistency due to OS differences of VMs - inconsistency in functionality between a service and its clients without synchronous coordination. - multiple versions of the same service that are simultaneously active

Inconsistency in functionality between a service and its clients without synchronous coordination.

Which is not a reason for DNS to return two or more IP addresses for a single URL? - Testing - Performance - Realibility - None of the above

None of the above

Although it has limitations in dealing with a cloud environment, Nagios is probably the most widely used monitoring tool due to its large number of plug-ins. Which of the following public or private IaaS providers uses this tool as its monitoring solution? - AWS - OrionVM - Microsoft Azure - TelKomsigma - Digital Ocean

OrionVM; TelKomsigma; Digital Ocean

The overall goal of deployment is to - support team members to work on different versions of the system concurrently - easily integrate code produced by one team with code produced by other teams - place an upgraded version of the service into production - determine exactly how a system came to be in production

Place an upgraded version of the service into production

Threats that are associated with users who deny performing an action without other parties having a way to prove otherwise are represented by the letter of _ of the acronym STRIDE for a threat model.

R

The CRM Salesforce is a good example of - IaaS - SaaS - PaaS - All of the above

SaaS

Which is not a characteristic of the cloud defined by NIST? - Resource pooling - Self diagnosis - Rapid elasticity - Broad network access

Self diagnosis

An architect involved in a DevOps project should ensure feature toggles are removed when the code they toggle has been placed into production and been judged to be successfully deployed.

True

An organization can introduce continuous deployment without major architectural modifications.

True

CA Technologies provides IT management solutions to their customers. They commissioned a survey in 2013. More than 50% of respondents of this survey said they were already seeing an increased frequency of deployment of their software or services.

True

Chef is a configuration management tool that can be used to maintain traceability of errors because of its capability of keeping track of all changes it applies to a machine.

True

Continuous deployment is the DevOps practice that leads to the most far-reaching architectural modifications.

True

Dev and DevOps are becoming more responsible for application SLAs and external SLAs in the DevOps movement.

True

In cloud computing, a VM is an emulation of a physical machine

True

Infrastructure-as-code is about using code (rather than manual commands) for setting up (virtual) machines and networks, installing packages, and configuring the environment for the application of interest.

True

The DevOps pipeline has to deal with different types of software ranging from desktop IDEs to code repositories, build/testing servers, cloud infrastructure APIs, and potentially complex environments for testing/staging/production.

True

The language you wish to expose to the user is an example of configuration setting (or parameter).

True

The main reason of making multiple instances of a service be available is to support a load too heavy for a single instance and to guard against failure.

True

The agent-based and agentless monitoring approaches both have their strengths and weaknesses. The _ approach is better in terms of deployment and maintenance effort. However, it is less secure if the collection repository is outside of your network.

agentless

If we compare blue/green deployment and rolling upgrade, _ does not introduce any logical problems but requires allocating twice the number of VMs required to provide a service, while _ is more efficient in how it uses resources but introduces a number of logical consistency problems.

blue/green deployment; rolling upgrade

A variation of big flip deployment model is to do the traffic switching gradually. A small percentage of requests are first routed to the new version, effectively conducting a _ test

canary

Which is not a goal or purpose of using a monitoring system? - configuration management - capacity planning - intruder detection - performance problems identification

configuration management

The CDC testing practice requires the test cases for testing a microservice to be decided and even co-owned by all the _ of that microservice

consumers

The seven-step process suggested by ITIL for continual service improvement is a _-driven process.

data

One of the design principles for security that needs to be applied to both the application design and the deployment pipeline itself is to utilize _ defaults. Argue why a particular process or client needs to have access, not why that process or client should not have access.

fail safe

DualShock controller was first introduced (in 1997) as a secondary peripheral for the original PlayStation (launched in 1994). Three years later, its successor DualShock 2 was released but the original PlayStation is still compatible with the DualShock 2. Several years later when PlayStation 3 became available in 2006, DualShock 2 can still be used with this newer PlayStation 3. Based on the above video gaming development scenario, we can say the original PlayStation is _ compatible with the DualShock 2, and PlayStation 3 is _ compatible with the DualShock 2.

forward; backward

In cloud computing, a VM _ is a file that contains a bootable operating system and some kind of software installed on it

image

VM _ is the term used to describe the proliferation of VM images to the point where the management systems lose track of them.

image sprawl

ITIL stands for _

information technology infrastructure library

An important trend in DevOps is to manage all the code and parameters for setting up an environment just as you manage your application code, with proper version control and testing. This is an example of _ -as-code

infrastructure

Controlling tools by scripts is far better than controlling tools by commands. The scripts and associated configuration parameters should also be kept under version control, just as the application code. This new movement is called _

infrastructure-as-code

Before writing the actual code for a piece of functionality, you develop an automated test for it. Then the functionality is developed, with the goal of fulfilling the test. This philosophy is called _

test-driven development

_ is the step in which the built executable artifact is tested.

integration testing

_ means no unauthorized people are able to modify information.

integrity

Selecting tools with stable interfaces and flexible scripting facilities could achieve the _ quality goal of a DevOps deployment pipeline.

interoperability

The popular source control system Git can support the _ with static analysis, testing, build, and notification/messaging systems through event-based hooks: The invocation of other systems using appropriate Git outputs can be carried out before or after a Git event such as commit or push.

interoperation

_ testing is a mechanism to continue to test even after placing a system in production or promoting it to normal production.

live

In a discussion of VM in Chapter 2, simple requests such as computation, reading a file, or receiving a local message will have a distribution closer to normal. Complicated requests such as extensive map-reduce jobs, searches across a large database, or launching virtual instances will have a skewed distribution such as a _

long tail

DevOps processes should be monitored so that they can be improved and problems can be detected. For this purpose, this chapter lists five things that are important to monitor, which include 1. A business _ 2. _ time 3. Mean time to _ errors 4. Mean time to report errors 5. Amount of scrap (rework)

metric; cycle; detect

The DevOps goal of minimizing coordination among various teams can be achieved by using a _ architectural style where the coordination mechanism, the resource management decisions, and the mapping of architectural elements are all specified by the _ and hence require minimal inter-team coordination.

microservice; architecture

The quality concern about the amount of efforts required to change the DevOps software, processes, or the operation environment of an application is called _

modifiability

The best way to determine if the system behaves in a defined way when assumptions about the environment hold or the user performs actions in the right order with the right inputs are not met is to run - A/B tests - canary tests - unit tests - negative tests

negative tests

The DevOps movement is taking an approach different from traditional IT management. This approach is to reduce the need for dedicated _ personnel through _ many of the tasks formerly done by these people and having _ assume a portion of the remainder.

operations; automating; developers

Both Canary Testing and A/B testing are a type of - rolling upgrade deployments - red/black deployments - partial deployments - all-or-nothing deployments

partial deployments

Two methods are discussed in Chapter 4 for managing demand on the service from client requests. One is to monitor _ and the other possible technique is to use _ to control the number of instances.

performance; SLAs

Security in DevOps is not only about application and operation security, it is also about the security of the _ itself, such as build/test server security, microservice component security, environment security, and security during dynamic provisioning

pipeline

When operators keep receiving false _ alarms or a flood of alerts from different channels about the same event, operators will quickly get "alert fatigue".

positive

DevOps is a set of _ intended to reduce the _ between committing a change to a system and the change being placed into normal production, while ensuring high _ This is how DevOps is defined in textbook, which focuses on the _, rather than the means.

practies; time; quality; goals

Using role-based access control (RBAC) requires the identity management system to maintain a mapping between users and roles and a mapping between roles and _

privileges

A software architect usually uses the word "ility" to describe _ concerns other than those that focus on the basic functionalities and their correctness

quality

'Version Control Everything' could be an approach to improve the quality of _ of a DevOps deployment pipeline

repeatability

Deployment pipeline does not contain the step of - build and integration tests - performance tests - pre-commit tests - requirement analysis

requirement analysis

One category of DevOps practices identified in the textbook is to treat Ops as first-class citizens from the point of view of _

requirements

One common deployment strategy results in multiple versions of a service being simultaneously active and this, in turn, raises questions of consistency. Which deployment strategy is this? _

rolling upgrade

Which event(s) could change the monitoring configuration? - rolling upgrade - changes to any devops tools - if the alert does not turn into an alarm - canary deployment

rolling upgrade; changes to any devops tools; canary deployment

Logs are used during operations to detect and diagnose problems. They should include context information and provide severity levels, alert levels, and other data. Severity and alert levels are an example of _ information

screening

ITIL v3 emphasizes the concept that IT is a _ that supports business goals.

service

Among several DevOps roles, _ is responsible to maintan and communicate the vision for the service assigned to develop by the team

service owner

_ are a subset of the automated acceptance tests that are used to quickly analyze if a new commit breaks some of the core functions of the application. - smoke tests - unit tests - user acceptance tests - none of the above

smoke tests

Events of interest during service operation include - Software license usage - Status information from systems - Performance metrics from applications - Security information

software licensing usage; status information from systems; performance metrics from applications; security information

Illegally accessing and then using another user's authentication information, such as username and password is an example of identity _

spoofing

Netflix Edda is a tool that was developed to capture _ changes of Amazon cloud resources and enable queries, including correlation with recent operations.

state

The people involved in the Ops side of DevOps typically come from - Technology experts - Service desk operations - Day-to-day provisioning of IT services - Consultants

technology experts; day-to-day provisioning of IT services

Development at Google is _-based rather than _-based. Instead of reintegrating a big branch, their developers deal with integration issues at each _, which is a simpler solution, but requires more frequent action than using branches.

trunk, branch, commit

ChefSpec is a tool for running _ of Chef cookbooks.

unit tests

Rolling back in a rolling upgrade model is more complicated than in blue/green deployment model. The new version B can be in one of four states during its lifetime: uninstalled, partially installed, fully installed but on probation, or committed into production. Two of these states have no rollback possibilities, including (must choose two correct answers for full credit) - version B has only been partially installed - version B has not yet been installed - version B has been committed into production - version B has been fully installed but on probation

version B has not yet been installed; version B has been committed into production