13-2
User
Guide

for

the

Cisco

Application

Networking

Manager

5.2
OL-26572-01
Chapter
13






Configuring

High

Availability
Understanding
ANM

High

Availability
Understanding
ANM

High

Availability
ANM
high

availability

(or

fault

tolerance)

ensures

that

your

network

services

and

applications

are
always
available.

High

availability

(HA)

provides

seamless

switchover

of

flows

in

case

an

ANM

server
becomes
unresponsive

or

a

critical

host

or

interface

fails.

High

availability

uses

two

ANM

nodes,

where
one
node

is

the

active

node

and

the

ot

her

i

s

the

standby

node.
The
ANM

high

availability

features

are

as

follows:

•
Automatic
determination

of

node
status,
whether
active
or

standby,

using

heartbeat

counts.

•
Designation
of

the

virtual

IP

address

(VIP),

which

is

associated

with

the
active
node.

•
Near
real-time
replication
of

ANM

configuration

and

events
after
a

failover

occurs.

•
Automatic
inspection
of
certificate/key
presence

on

HA

peer
upon
SSL

certificate
or
key

import.
During
normal

operation,

ANM

high

availability

performs

the

following

actions:

•
The
t

wo

nodes

constantly
exchange
heartbeat

packets
over
bot

h

i

nterfaces.

•
Database
operations

that

occur

on

the

active

node’s

database

are

replicated
on
the

standby

node’s
database.

•
The
monitor
function
ensures

that

the

necessary

processes

are

running
on
both

the

active

and
standby
node.

For

example,

not

al

l

processes

necessarily

run

on

t

he

standby

node,

so

after

a

node
changes
from

active

to

standby,

ANM

high

availability

function

stops

certain

processes

on

the
standby
node.
When
you

log

into

ANM,

you

log

in

using

a

virtual

IP

address

(VIP)

that

associates

with

the

active

node.
The
VIP

is

the

only

IP

address

you

need

to

remember.

If

the

current

active

node

fails,

the

standby

node
takes
over

as

the

active

node

and

the

VIP

automatically

associates

with

the

node

that

has

just

become
active.
When

a

failover

occurs

and

the

standby

node

becomes

the

active

node,

all

existing

web

sessions
are
lost.

In

addition,

there

is

a

slight

delay

while

the

standby

node

takes

over

as

the

active

node.

After
the
switchover

is

complete

and

the

ANM

fully

initializes,

you

can

log

into

ANM

using

the

same

VIP.

All
ANM
functions

remain

the

same.
ANM
uses

heartbeat

counts

t

o

determine

when

a

failover

should

occur.

Because

both

nodes

ar

e
constantly
sending

and

receiving

heartbeat

packets,

if

heartbeat

packets

are

no

longer

being

received

on
a
node,

its

peer

node

is

determined

to

be

dead.

If

this

peer

node

was

the

active

node,

then

the

standby
node
takes

over

as

the

active

node.

The

VIP

automatically

associates

with

the

newly

active

node,

and

the
monitoring
process

starts

any

necessary

processes

on

t

he

newly

active

node

that

were

not


al

ready
running.
Similarly,
if

you

manually

issue

a

failover

to

cause

the

active

node

to

become

the

standby

node,

the
heartbeat
process

disassociates

the

VIP

from

the

node

and

tells

the

monitoring

function

to

stop

processes
that
are

not

normally

run

on

the

standby

node.
Related
Topics

•
Understanding
ANM

High

Availability
Processes,
page
13-3

•
Configuring
ANM
High

Availability

Overview,

page
13-3

•
CLI
Commands

for


ANM

High

Availability
Processes,
page
13-4

•
Recovering
From

an

HA

Database

Replication

Failure,

page
13-6