High Availability and Fault Tolerance: © 2011 Vmware Inc. All Rights Reserved

High Availability and Fault Tolerance
Module 11
© 2011 VMware Inc. All rights reserved

You Are Here
Course Introduction Data Protection
Introduction to Virtualization Access & Authentication Control
Virtual Machines Resource Management and Monitoring
VMware vCenter Server High Availability
Configure and Manage Virtual Networks Scalability
Configure and Manage Virtual Storage Patch Management
Managing Virtual Machines Installing vSphere Components
VMware vSphere: Install, Configure, Manage – Revision A 11-2

Importance
Most organizations rely on computer-based services like email,

databases, and Web-based applications. The failure of any of these
services can mean lost productivity and revenue. Configuring highly
available, computer-based services is extremely important for an
organization to remain competitive in contemporary business
environments.
With VMware vSphere® 5, a new high availability architecture has
been released.

Module Lessons
Lesson 1: Introduction to vSphere High Availability

Lesson 2: Configuring vSphere High Availability
Lesson 3: vSphere High Availability Architecture
Lesson 4: Introduction to Fault Tolerance

Lesson 1:
Introduction to
vSphere High Availability

Learner Objectives
After this lesson, you should be able to do the following:

Describe the various options that you can configure to ensure high
availability in a vSphere 5 environment.
Discuss the response of vSphere High Availability when a
VMware® ESXi™ host, a virtual machine, or an application fails.

VMware Offers Protection at Every Level
Protection against hardware failures

Planned maintenance with zero downtime
Protection against unplanned downtime
and disasters
High Availability & Fault Tolerance

vSphere Site
Storage Recovery
VMware vSphere®
VMotion Manager
vMotion®, DRS
NIC Teaming,
Storage
Multipathing
3rd-Party
Backup Solutions,
VMware Data Recovery
Component Server Storage Data Site

vCenter Server Availability - Recommendations
Make VMware vCenter Server™ and the components it relies on

highly available.
vCenter Server relies on:
vCenter Server database:
• Cluster the database. Refer to the specific database documentation.
Active Directory structure:
• Set up with multiple redundant servers.
Methods for making vCenter Server available:
Use vSphere High Availability to protect the vCenter Server virtual
machine.
Use VMware vCenter Server Heartbeat™.

High Availability
A highly available system is one that is continuously operational for

a desirably long length of time.
Level of availability Downtime per year
What level of virtual 99% 87 hours (3.5 days)

machine availability is
99.9% 8.76 hours
important to you?
99.99% 52 minutes
99.999% 5 minutes

vSphere High Availability
vSphere HA
Level of availability High availability
Amount of downtime Minimal
Works with all supported guest operating

Guest operating systems supported
systems
VMware ESXi hardware supported Works with all supported ESXi hardware
Use to provide high availability for the

Uses virtual machines that require that level of
protection.

vSphere HA Failure Scenarios
ESXi host failure

Guest OS failure
Application failure

High Availability Failure Scenario - Host
LUN 1 LUN 2 LUN 3
virtual machine A virtual machine B

When a host fails,
virtual machine A virtual machine C virtual machine E
vSphere HA restarts
virtual machine B virtual machine D virtual machine F the affected virtual
machines on other
hosts
ESXi host ESXi host ESXi host
vCenter Server = vSphere HA cluster

High Availability Failure Scenario – Guest Operating System
LUN 1 LUN 2 LUN 3
virtual machine A virtual machine C virtual machine E When a virtual

VMware tools VMware tools VMware tools machine stops
sending heartbeats
virtual machine B virtual machine D virtual machine F
VMware tools
or the virtual
VMware tools VMware tools
machine process
crashes (vmx),
ESXi host ESXi host ESXi host vSphere HA resets
the virtual machine

HA Failure Scenario - Application
LUN 1 LUN 2 LUN 3
application application application When an application

virtual machine A virtual machine C virtual machine E fails, vSphere HA
restarts the affected
application application application
virtual machine on
the same host.
ESXi host ESXi host ESXi host Requires VMware

Tools to be installed

Review of Learner Objectives
You should be able to do the following:

Describe the various options that you can configure to ensure high
availability in a vSphere 5 environment.
Discuss the response of vSphere High Availability when an ESXi
host, a virtual machine, or an application fails.

Lesson 2:
Configuring vSphere High Availability

Learner Objectives

Configure a vSphere HA cluster.

Enabling vSphere HA
Enable vSphere HA by creating a cluster or modifying a vSphere

Distributed Resource Scheduler (DRS) cluster.

Configuring vSphere HA Settings
Disable Host
Monitoring when
performing
maintenance
on any cluster/host.
Enabled is the
default setting.
Admission Control
refers to the amount
of available resources Admission control
that can be used to
helps ensure sufficient
start virtual machines
resources to provide
on a specific ESXi
host. high availability.
The default setting Default setting is
is to disallow power Host failures the
and other operations cluster tolerates.
that will violate the VMware
set Admission recommended
Control Policy. setting

Admission Control Policy Choices
Policy Description Recommended use
Percentage of cluster Reserves specified When virtual machines

resources reserved as percentage of total capacity have highly variable CPU
failover spare capacity and memory reservations
Host failures cluster Reserves enough resources When virtual machines
tolerates to tolerate specified number have similar CPU/memory
of host failures reservations and similar
memory overheads
Specify a failover host Dedicates a host To accommodate
exclusively for failover organizational policies that
service dictate the use of a passive
failover host

Configuring Virtual Machine Options
Configure options at the cluster level or per virtual machine.

VM restart priority determines relative order in
which virtual machines are restarted after a host
failure.
Host Isolation response determines what

happens to virtual machines when a host loses
the management network but continues running.

Configuring Virtual Machine Monitoring
Reset a virtual machine

if its VMware Tools
heartbeat or VMware
Tools application
heartbeats are not
received.
Determine how
quickly failures are
detected.
Set monitoring sensitivity for

individual virtual machines.

Importance of Redundant Heartbeat Networks
In a vSphere HA cluster, heartbeats are:

Sent between the master and the slave hosts
Used to determine if a master or slave host has failed
Sent over a heartbeat network
The heartbeat network is:
Implemented using a VMkernel port marked for management
Redundant heartbeat networks:
Allow for the reliable detection of failures

Redundancy Using NIC Teaming
You can use NIC teaming to create a redundant heartbeat network

on ESXi hosts.
Both port groups must be VMkernel ports.
NIC teaming on an ESXi host

Redundancy Using Additional Networks
You can also create

redundancy by configuring
more heartbeat networks:
On ESXi hosts, add one or
more VMkernel networks
marked for management
traffic.
Configure port group with
these settings:
Set Load Balancing to
originating port ID.
Do not enable Failback.
Configure port group with
active/standby failover.

Network Configuration and Maintenance
Before changing the networking configuration on the ESXi hosts

(adding port groups, removing vSwitches):
Deselect
Enable Host
Monitoring.
Place the
host in
maintenance
mode.
These steps prevent unwanted attempts to fail over virtual
machines.

Cluster Resource Allocation Tab
How much CPU and memory resources is the cluster using now?
How much reserved capacity remains?

Monitoring Cluster Status
cluster’s Summary tab
The vSphere HA Cluster Status

window displays details about host
operational status, virtual machine
protection, and heartbeat datastores
The Configuration Issues window
displays the current vSphere HA
operational status, including the
specific status and errors for each
master and slave host in the cluster.

Lab 18
In this lab, you will modify slot sizes and admission control.
1. Create a cluster enabled for vSphere HA.
2. Add your ESXi host to a cluster.
3. Test vSphere HA functionality.
4. Prepare for the next lab.


Configure a vSphere HA cluster.

Lesson 3:
vSphere High Availability Architecture

Learner Objectives

Describe heartbeat mechanisms used by vSphere HA.
Identify and discuss additional failure scenarios.

vSphere HA Architecture: Agent Communication
datastore datastore datastore
FDM FDM FDM
vpxa hostd vpxa hostd vpxa hostd
ESXi host (slave) ESXi host (slave) ESXi host (master)
vpxd
= Management network
vCenter Server

vSphere HA Architecture: Network Heartbeats
NAS/NFS VMFS Local

(slave) (slave) (master)
Management network 1
vCenter Server Management network 2

vSphere HA Architecture: Datastore Heartbeats
NAS/NFS VMFS Local

(slave) (master) (slave)
Cluster Edit Settings Window
vCenter Server

Additional HA Failure Scenarios
Slave host failure

Master host failure
Host isolation
Management network failures
• Network partition
• Network isolation

Failed Slave Host
NAS/NFS VMFS
(lock file) (heartbeat region)
file locks file locks

(slave) ? (master) (slave)
vCenter Server primary heartbeat network

alternate heartbeat network

Failed Master Host
NAS/NFS VMFS
(lock file) (heartbeat region)
file locks file locks
virtual machine A virtual machine C virtual machine E default gateway

(isolation address)

Role: slave Role: master Role: slave
MOID: 98 MOID: 99 ? MOID: 100
primary heartbeat network

vCenter Server alternate heartbeat network
MOID = managed object ID

Isolated Host

The host is not observing
virtual machine D virtual machine F any election traffic on the
virtual machine B
management and cannot
ping its isolation
ESXi host ESXi host ESXi host address(es), the host is
isolated.
default gateway
(isolation address)

Design Considerations
Host isolation events can be minimized through good design

Implement redundant heartbeat networks
Implement redundant isolation addresses
If host isolation events do occur, good design enables vSphere HA
to determine whether the isolated host is still alive
Implement datastores so that they are separated from the
management network using one or both of the following approaches:
• Fibre Channel over fibre optic
• Physically separating your IP storage network from the management
network

Network Partition
virtual machine A virtual machine C virtual machine E virtual machine G
virtual machine B virtual machine D virtual machine F virtual machine H
ESXi host ESXi host ESXi host ESXi host

MASTER SLAVE SLAVE SLAVE
MASTER
vCenter Server default gateway

(isolation address)


Describe heartbeat mechanisms used by vSphere HA
Identify and discuss additional failure scenarios

Lesson 4:
Introduction to Fault Tolerance

Learner Objectives

List Fault Tolerance requirements and limitations.
Describe Fault Tolerance operation.

What Is Fault Tolerance (FT)?
FT:
A fault-tolerant system is designed so that, in the event of an
unplanned outage, a backup virtual machine can immediately take
over with no loss of service. (The backup virtual machine is called a
secondary virtual machine.)
• Provides a higher level of business continuity than vSphere HA
• Provides zero downtime and zero data loss for applications
FT can be used for any application that needs to be available at all
times.
FT can be used with DRS:
Fault-tolerant virtual machines benefit from better initial placement
and are included in the cluster’s load-balancing calculations.

VMware Fault Tolerance
Fault Tolerance
Level of availability Fault tolerance
Amount of downtime Zero
Works with all supported guest

Guest operating systems supported
operating systems
ESXi hardware supported Widely compatible
Use to provide fault tolerance to

Uses
your critical virtual machines.

Fault Tolerance in Action
vLockstep technology vLockstep technology
primary new
secondary new
VM primary
VM secondary
VM VM
FT provides zero-downtime, zero-data-loss protection to

virtual machines in a vSphere HA cluster.

Fault Tolerance Guidelines
Check the requirements and limitations of FT.

Ensure enough ESXi hosts for fault-tolerant virtual machines:
No more than four fault-tolerant virtual machines (primaries or
secondaries) on any single host
Store ISOs on shared storage for continuous access:
Especially if used for important operations
Disable BIOS-based power management:
Prevents the secondary virtual machine from having insufficient CPU
resources

Enabling Fault Tolerance on a Virtual Machine


List Fault Tolerance requirements and limitations.
Describe Fault Tolerance operation.

Key Points
vSphere HA restarts virtual machines on the remaining hosts in the

cluster.
Hosts in vSphere HA clusters have a master/slave relationship.
Implement redundant heartbeat networks either with NIC teaming or
by creating additional heartbeat networks.
FT provides zero downtime for applications that need to be available
at all times.

High Availability and Fault Tolerance: © 2011 Vmware Inc. All Rights Reserved

Uploaded by

Copyright:

Available Formats

High Availability and Fault Tolerance: © 2011 Vmware Inc. All Rights Reserved

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

High Availability and Fault Tolerance: © 2011 Vmware Inc. All Rights Reserved

Uploaded by

Copyright:

Available Formats

High Availability and Fault Tolerance

© 2011 VMware Inc. All rights reserved

Course Introduction Data Protection

Introduction to Virtualization Access & Authentication Control

Virtual Machines Resource Management and Monitoring

VMware vCenter Server High Availability

Configure and Manage Virtual Networks Scalability

Configure and Manage Virtual Storage Patch Management

Managing Virtual Machines Installing vSphere Components

VMware vSphere: Install, Configure, Manage – Revision A 11-2

© 2011 VMware Inc. All rights reserved

Most organizations rely on computer-based services like email,

VMware vSphere: Install, Configure, Manage – Revision A 11-3

© 2011 VMware Inc. All rights reserved

Lesson 1: Introduction to vSphere High Availability

VMware vSphere: Install, Configure, Manage – Revision A 11-4

© 2011 VMware Inc. All rights reserved

VMware vSphere: Install, Configure, Manage – Revision A 11-5

© 2011 VMware Inc. All rights reserved

After this lesson, you should be able to do the following:

VMware vSphere: Install, Configure, Manage – Revision A 11-6

© 2011 VMware Inc. All rights reserved

Protection against hardware failures

High Availability & Fault Tolerance

Component Server Storage Data Site

VMware vSphere: Install, Configure, Manage – Revision A 11-7

© 2011 VMware Inc. All rights reserved

Make VMware vCenter Server™ and the components it relies on

VMware vSphere: Install, Configure, Manage – Revision A 11-8

© 2011 VMware Inc. All rights reserved

A highly available system is one that is continuously operational for

Level of availability Downtime per year

What level of virtual 99% 87 hours (3.5 days)

VMware vSphere: Install, Configure, Manage – Revision A 11-9

© 2011 VMware Inc. All rights reserved

Level of availability High availability

Amount of downtime Minimal

Works with all supported guest operating

Use to provide high availability for the

VMware vSphere: Install, Configure, Manage – Revision A 11-10

© 2011 VMware Inc. All rights reserved

ESXi host failure

VMware vSphere: Install, Configure, Manage – Revision A 11-11

© 2011 VMware Inc. All rights reserved

LUN 1 LUN 2 LUN 3

virtual machine A virtual machine B

vCenter Server = vSphere HA cluster

VMware vSphere: Install, Configure, Manage – Revision A 11-12

© 2011 VMware Inc. All rights reserved

LUN 1 LUN 2 LUN 3

virtual machine A virtual machine C virtual machine E When a virtual

vCenter Server = vSphere HA cluster

VMware vSphere: Install, Configure, Manage – Revision A 11-13

© 2011 VMware Inc. All rights reserved

LUN 1 LUN 2 LUN 3

application application application When an application

ESXi host ESXi host ESXi host Requires VMware

vCenter Server = vSphere HA cluster

VMware vSphere: Install, Configure, Manage – Revision A 11-14