R Series Application Brief

Digium Application Brief
R-Series Failover Appliances

This application brief summarizes strategies for implementing high availability/
failover and outlines the components necessary to implement a redundant/
high availability Asterisk solution using the R-Series appliances
from Digium. The R-Series devices handle the physical redirection
of PSTN circuits in the event of a catastrophic hardware or software
failure. Tightly integrated with monitoring and resource management
software, the R-Series can switch PSTN traffic within seconds of a
service interruption. A pair of Asterisk-based servers cross-connected through an
Digium is the creator,
sponsor, and innovative
force behind Asterisk,
the industrys first and
worlds most popular
open source telephony
software. Additionally,
Digium provides a variety
of VoIP communication
solutions that fit the needs
of small, medium, and large
businesses. Digiums product
lines include commercial
business phone systems, as
well as software, hardware,
and other components
needed to create powerful
custom communications
solutions.
R-Series appliance can achieve 99.99% availability.
Application Summary
When high availability business communication
services are required, the R-Series appliances,
Asterisk and additional open-source tools can
be converged to create a completely redundant and mirrored clustered environment for
24x7 availability. What is high availability (HA)?
It can be defined as an approach to communications system design which emphasizes
service availability. In true HA systems, core
services are available at all times. For users of
business communications systems, this would
include key PBX services such as inbound and
outbound calling, voice mail, automated attendants, call queues and conferencing.
For the Server Administrator, high availability

means more than one Asterisk system, a
primary and secondary, clustered for the
purpose of sensing a failure of the primary.
Once a failure of the primary is detected,
whether hardware or software, the backup
communication server will assume
responsibility of all communication services.
For true high availability, the backup
communications server must restore all
services such as voice mail, dial plans, IP
addressing and much more intact.
Deployment
The following diagram shows an R-Series
failover appliance managing PSTN connections to a pair of mirrored Asterisk Servers.
Server A is the primary sever while Server B is
a hot standby unit.
Digium Application Brief
R-Series Failover Appliances

The two servers are either identical, or sufficiently
comparable such that either server can act as the
primary at any given time. This scenario describes
that happens when a fatal hardware or software
failure is experienced on the machine serving as
the primary.
Components:

n
n
n
n
PSTN connection (analog or digital lines)

R-Series R800 or R850 unit
Server A
Server B
Preconditions:
This example uses the R850 appliance and digital
(T1) PSTN service
n Server A and Server B are running the software
configuration described below
n Server A is acting as the primary node
Asterisk is running on Server A
The R-Series unit has the T1(s) routed to
Server A
VoIP traffic is directed to an IP address bound
to Server A
n Server B is acting as a standby node for Server A
n
Put second diagram here
Trigger
Server A experiences a fatal software hardware
failure
General Flow
The host software on Server B detects that
Server A has left the cluster
n The host software on Server B instructs the
R-Series unit to direct the T1(s) to Server B
n The host software on Server B activates the
floating IP address locally
n The host software on Server B starts Asterisk
n Server A comes back online and rejoins the
cluster as a standby for Server B
n
Software Configuration
The software used for this method of automatic
failover includes Pacemaker, Corosync, and DRBD.
A brief description of each of these components is
listed below.
Pacemaker is a resource manager for clusters
(including clusters with just two nodes as outlined
in this example). The resource manager takes care
of starting and stopping services on the nodes
(computers) in the cluster. Think of Pacemaker as
the director it tells the other parts of the system
what to do.
Pacemaker is an Open Source, High

Availability resource manager suitable for
both small and large clusters. Hardware and
application failures can result in prolonged
downtime and impact your bottom line. In
the event of a failure, resource managers like
Pacemaker automatically initiate recovery
and make sure your application is available
from one of the remaining machines in the
cluster. Your users may never even know
there was a problem.
In the event of a hardware of software
failure on the primary server, the standby
server will resume communications with
all services (voice mail, IVR, etc.) intact.
www.clusterlabs.org
Digium Custom
Communications Solutions
Digium empowers users,
developers and integrators
to build custom telephony
solutions by offering a variety
of software, hardware, and
third-party components.
From basic voice applications
to sophisticated phone
systems, Digium makes it
possible for the world to
communicate at a fraction
of the cost of proprietary
solutions.
At the heart of these
offerings is Asterisk, the
powerful open source
telephony engine. Asterisk
is free software that turns
an ordinary computer
into a feature-rich voice
communications platform.
Its flexible architecture lets
you configure it as an IP PBX,
a voicemail server, IVR server,
VoIP gateway, call recorder,
automatic call distributor or
virtually any other voiceenabled application that you
can imagine.
Corosync is the messaging system that lets

all the nodes in the cluster reliably communicate with each other. In this example, Corosync sends system delivers status (health)
messages from one node to the other.
DRBD DRBD is a Linux utility that
synchronizes data between file systems on
different computers. It works much like
RAID-1 it mirrors the data on the drives
but over the LAN rather than on the same
system. In this example DRBD is used to
keep configuration data
DRBD refers to block devices designed

as a building block to form high
availability (HA) clusters. This is done
by mirroring a whole block device via
an assigned network. DRBD can be
understood as network-based RAID-1.
Failure Detection
Pacemaker manages resources in a cluster.
A resource agent (RA) is a utility used by
Pacemaker to take care of the details of
managing a resource. For our purposes,
Pacemaker is managing resources in a twonode cluster. Pacemaker will be configured to
periodically poll the resource agent in charge
of the Asterisk process. One of the required
interfaces of a resource agent is to be able to
return the status of the resource.
The cluster messaging layer, Corosync, is in
charge of determining cluster membership. If it
loses contact with another node for any reason,
it will decide that it has failed and failover will
be initiated by notifying the R-Series appliance
to redirect the PSTN services.
www.drbd.org
For detailed step-by-step configuration, please view the R-Series

manual at http://docs.digium.com/R800F/r-series-manual.pdf
Digium, Inc. 445 Jan Davis Drive NW, Huntsville, AL 35806, USA
Phone: +1 256-428-6000 Fax: +1 256-864-0464 www.digium.com
Copyright 2012 Digium, Inc. All rights reserved. Digium and Asterisk are registered trademarks of Digium, Inc.
All other trademarks are property of their respective owners. Version 1.0/6 August 2012

R Series Application Brief

Uploaded by

Copyright:

Available Formats

R Series Application Brief

Uploaded by

Document Information

Original Description:

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

R Series Application Brief

Uploaded by

Copyright:

Available Formats

Digium Application Brief

R-Series Failover Appliances

R-Series appliance can achieve 99.99% availability.

For the Server Administrator, high availability

Digium Application Brief

R-Series Failover Appliances

PSTN connection (analog or digital lines)

Put second diagram here

Pacemaker is an Open Source, High

Corosync is the messaging system that lets

DRBD refers to block devices designed

For detailed step-by-step configuration, please view the R-Series

You might also like