OpenStack for Architects电子书

OpenStack for Architects

OpenStack for Architects

Credits

About the Authors

www.PacktPub.com

Why subscribe?

Customer Feedback

Preface

What this book covers

What you need for this book

Who this book is for

Conventions

Reader feedback

Customer support

Downloading the example code

Downloading the color images of this book

Errata

Piracy

Questions

1. Introducing OpenStack

What is OpenStack?

OpenStack is an API

OpenStack - an open source software project

OpenStack - a private cloud platform

OpenStack components

Compute

Object Storage

Block storage

Network

Common OpenStack use cases

Public hosting

High-performance compute

Rapid application development

Network Function Virtualization

Drafting an initial deployment plan

The role of the Architect

The design document

The deployment plan

Your first OpenStack deployment

Writing the initial deployment plan

Hardware

Network addressing

Configuration notes

Requirements

Installing OpenStack

Installation instructions

Verifying the installation

Next steps

Summary

References

2. Architecting the Cloud

Picking an OpenStack distribution

Running from the trunk

Community distributions

Commercially supported distributions

Compute hardware considerations

Hypervisor selection

Sizing the hardware to match the workload

Considerations for performance-intensive workloads

Network design

Providing network segmentation

SDN

Physical network design

Storage design

Ephemeral storage

Block storage

Object storage

Expanding the initial deployment

Updating the design document

Cloud controller

Compute node

Management network

Provider network

Tenant network

Updating the deployment plan

Installing OpenStack with the new configuration

Summary

References

3. Planning for Failure (and Success)

Building a highly available control plane

About failure and success

High availability patterns for the control plane

Active/Passive service configuration

Active/Active service configuration

OpenStack service specifics

OpenStack web services

Database services

The message bus

Compute, storage, and network agents

Regions, cells, and availability zones

Regions

Cells

Availability zones

Updating the design document

Planning the physical architecture

Updating the physical architecture design

Implementing H/A in the lab deployment

Provisioning a second controller

Installing the Pacemaker resource manager

Installing and configuring HAProxy

Additional API service configuration

Summary

References

4. Building the Deployment Pipeline

Dealing with Infrastructure as a Software

Eating the elephant

Writing the tests first

Always be deploying

Using configuration management for deployment

Using the community modules

Assigning roles

Choosing a starting point

Test infrastructure

Types of testing

Writing the tests

Running the tests

Putting the pipeline together

Setting up the CI server

Installing Git

Installing a Puppet master

Installing Jenkins

Creating the composition layer

Starting our Puppet modules

Defining the first role and profile

Running the first build

Writing the tests

Assigning the first role to a system

Installing Keystone

Fully automating the pipeline

Summary

References

5. Building to Operate

Expected outcomes of this chapter

Logging, monitoring, and alerting

Logging

Monitoring

What to monitor

Monitoring practices

Monitoring availability

Monitoring performance

Monitoring resource usage

Alerting

Active monitoring

Services

Processes

HA control cluster

Capacity planning

Planning your city

Tracking usage and analyzing growth

Flavor sizing and compute server hardware selection

Backups and recovery

Infrastructure backup architecture

Backup strategies – what to backup

Workload backup architecture

Planning for disaster recovery

Summary

References

6. Integrating the Platform

IdM integration

Authentication and authorization in OpenStack

Configuring Keystone with split assignment and identity

Provisioning workflows

The Horizon user interface

Using the REST APIs

Provisioning with templates

Metering and billing

Listening to OpenStack

Using the notification subsystem

Consuming events from Ceilometer

Reading meters in Ceilometer

Updating the design document

Writing requirements

Testing requirements

Summary

References

7. Securing the Cloud

Security zones within OpenStack

Software vulnerabilities

Instance software security and patching

Infrastructure host security and patching

Patching OpenStack code

Patching the operating system

Red Hat Enterprise Linux and CentOS

Canonical Ubuntu based operating systems

Software repository management

Hardening hypervisors

Standard Linux hardening practices and hypervisors

SELinux and AppArmor

sVirt

SELinux and sVirt in action

SSL and certificate management

Assessing risk

Best practices for endpoint security

Examples

Auditing OpenStack

CADF details

Using CADF with OpenStack

Log aggregation and analysis

Summary

References

8. Conclusion

Emerging trends in OpenStack

Moving up the stack

The impact of containers

Building the roadmap

Introducing new features

Releasing new versions

Summary

References