ACOS 4.1.

1-P11
Configuring Scaleout
for A10 Thunder® Series and AX™ Series
29 May 2019
© 2019 A10 NETWORKS, INC. CONFIDENTIAL AND PROPRIETARY- ALL RIGHTS RESERVED
Information in this document is subject to change without notice.

PATENT PROTECTION
A10 Networks products are protected by patents in the U.S. and elsewhere. The following website is provided to satisfy the virtual pat-
ent marking provisions of various jurisdictions including the virtual patent marking provisions of the America Invents Act. A10 Net-
works' products, including all Thunder Series products, are protected by one or more of U.S. patents and patents pending listed at:

https://www.a10networks.com/company/legal-notices/a10-virtual-patent-marking

TRADEMARKS
A10 Networks trademarks are listed at:

https://www.a10networks.com/company/legal-notices/a10-trademarks

CONFIDENTIALITY
This document contains confidential materials proprietary to A10 Networks, Inc. This document and information and ideas herein may
not be disclosed, copied, reproduced or distributed to anyone outside A10 Networks, Inc. without prior written consent of A10 Net-
works, Inc.

A10 NETWORKS INC. SOFTWARE LICENSE AND END USER AGREEMENT

Software for all A10 Networks products contains trade secrets of A10 Networks and its subsidiaries and Customer agrees to treat Soft-
ware as confidential information.

Anyone who uses the Software does so only in compliance with the terms of the End User License Agreement (EULA), provided later in
this document or available separately. Customer shall not:

1. Reverse engineer, reverse compile, reverse de-assemble, or otherwise translate the Software by any
means.
2. Sub-license, rent, or lease the Software.

DISCLAIMER
This document does not create any express or implied warranty about A10 Networks or about its products or services, including but not
limited to fitness for a particular use and non-infringement. A10 Networks has made reasonable efforts to verify that the information
contained herein is accurate, but A10 Networks assumes no responsibility for its use. All information is provided "as-is." The product
specifications and features described in this publication are based on the latest information available; however, specifications are sub-
ject to change without notice, and certain features may not be available upon initial product release. Contact A10 Networks for current
information regarding its products or services. A10 Networks’ products and services are subject to A10 Networks’ standard terms and
conditions.

ENVIRONMENTAL CONSIDERATIONS
Some electronic components may possibly contain dangerous substances. For information on specific component types, please con-
tact the manufacturer of that component. Always consult local authorities for regulations regarding proper disposal of electronic com-
ponents in your area.

FURTHER INFORMATION
For additional information about A10 products, terms and conditions of delivery, and pricing, contact your nearest A10 Networks loca-
tion, which can be found by visiting www.a10networks.com.
Table of Contents

Configuring Scaleout .................................................................................................................... 5

Scaleout Overview ..................................................................................................................5
What is Scaleout? .........................................................................................................................................5
Scaleout Benefits ..........................................................................................................................................6
Load Balancing As You Grow ..............................................................................................................6
Appropriately Provisioned Network ....................................................................................................6
Surge Relief .............................................................................................................................................6
Scaleout Topology ........................................................................................................................................6
Understanding Traffic Distribution........................................................................................9
Distributing Traffic from Upstream Devices ...........................................................................................9
Using NAT to Handle Return Traffic ...................................................................................................... 10
Election of the Cluster Master .............................................................................................10
Scaleout Configuration Example .........................................................................................10
Configure VCS on Each Device ........................................................................................................ 12
Configure Scaleout on the vMaster.....................................................................................14
Configure Scaleout IP NAT Pool Group on the vMaster ............................................................. 16
Configure SLB on the vMaster ......................................................................................................... 16
Graceful Addition and Removal of Devices.........................................................................17
Graceful Addition of a Device to the Cluster ........................................................................................ 17
Graceful Shutdown of a Device in the Cluster ..................................................................................... 18
Scaleout Configuration Notes .............................................................................................19

Config Commands: Scaleout ...................................................................................................... 21

Scaleout Global Configuration Commands.........................................................................22
scaleout .......................................................................................................................................... 22
Scaleout Configuration Commands ....................................................................................22
cluster-devices .............................................................................................................................. 23
device-groups ............................................................................................................................... 23
follow-vcs ....................................................................................................................................... 24
local-device .................................................................................................................................... 24
service-config ................................................................................................................................ 24
Scaleout Local Device Configuration Commands ..............................................................25
id ...................................................................................................................................................... 25
priority ............................................................................................................................................ 26
Scaleout show Commands ..................................................................................................26
show scaleout ............................................................................................................................... 26

page 3
ACOS 4.1.1-P11 Configuring Scaleout
Contents

page 4
 ACOS 4.1.1-P11 Configuring Scaleout

Configuring Scaleout

This chapter describes the Scaleout feature.

The following topics are covered:

• Scaleout Overview

• Understanding Traffic Distribution

• Election of the Cluster Master

• Scaleout Configuration Example

• Graceful Addition and Removal of Devices

• Scaleout Configuration Notes

Scaleout Overview
This section contains the following:

• What is Scaleout?

• Scaleout Benefits

• Scaleout Topology

What is Scaleout?
Scaleout is a solution where multiple ACOS devices form a cluster to provide the same set of services.
The Scaleout technology enables ADC services to be provided across multiple devices for load distribu-
tion. The services provided by a virtual IP or virtual port can be spanned across multiple devices that
form a Scaleout cluster.

It also provides an option to dynamically add or remove devices to the cluster. When the number of
devices in the cluster changes the traffic gets rebalanced. This may cause traffic to be serviced by a
different device in the cluster.

page 5
ACOS 4.1.1-P11 Configuring Scaleout
Scaleout Overview

Scaleout Benefits
This section describes some use cases where Scaleout can be successfully implemented.

• Load Balancing As You Grow

• Appropriately Provisioned Network

• Surge Relief

Load Balancing As You Grow

A small organization may begin with a few ACOS devices to meet its initial needs. However, over time,
as the organization grows, its ACOS capacity requirements grow with it. A given device may not be able
to handle all the traffic directed to a VIP configured on the device in terms of throughput. Scaleout
allows you to add and group devices together in a cluster, and configure the VIP to be load balanced
across multiple devices in the cluster, thereby allowing the VIP to scale beyond the capacity of a single
device. The new device can also support new VIPs or additional functionalities that are not scaled out
to all the devices.

Appropriately Provisioned Network

Not all network resources have to be available one hundred percent of the time. Usually, network utili-
zation has peak times throughout the day. For example, fixed-interval software updates (Microsoft's
Patch Tuesday) result in predictable network peaks. Enterprises often do not need all their resources
available outside of working hours. In such cases, fewer ACOS devices can be enabled and other parts
of the network can be switched off during the off-peak hours. This decision can be made based on both
the time of the day, the amount of user traffic, or CPU utilization. This results in a much more efficient
network, power usage saving and reduced cooling requirements.

Scaleout allows new devices to be provisioned in or out of the cluster (see ““Graceful Addition and
Removal of Devices” on page 17) seamlessly to accommodate peak utilization scenarios.

Surge Relief
In situations where an ACOS device is suddenly presented with a flow of traffic that causes it to work
near full capacity, the load of other traffic can be temporarily distributed across other ACOS devices in
the Scaleout cluster. When the load reduces, the service can easily be rolled back to the devices origi-
nally handling that service.

Scaleout Topology
Figure 1 illustrates a sample basic Scaleout topology.

page 6
 ACOS 4.1.1-P11 Configuring Scaleout
Scaleout Overview

NOTE: A Scale Out cluster requires a minimum of three devices to be opera-

tional, up to a maximum of eight devices. When a majority of the devices
in the cluster are down, then all service nodes are removed from the clus-
ter and the service goes down.

FIGURE 1 Sample Scaleout Topology

The elements in Figure 1 are described below in Table 1.

TABLE 1 Explanation of the Sample Scaleout Topology

Element Description
Scaleout Cluster The group of ACOS devices configured to provide Scaleout functionalities.

To configure the devices that belong to the cluster, use the cluster-devices com-
mand.
Cluster Service The cluster service relates to a specific type of service provided by a VIP or virtual port
being scaled out across multiple devices (cluster nodes).

Each cluster service can be provided by different sets of devices in the cluster. For
example, in the example above, HTTP traffic to a single VIP is scaled out across Device
1 and Device 2, and TCP traffic to a single virtual port is scaled out across Device 2 and
Device 3.

Use the template scaleout command to identify the services that you want to scale
out. For more information, see the Command Line Interface Reference, or “Scaleout Con-
figuration Example” on page 10.

page 7
ACOS 4.1.1-P11 Configuring Scaleout
Scaleout Overview

TABLE 1 Explanation of the Sample Scaleout Topology

Element Description
Traffic Map Each ACOS device in the cluster has what is known as a traffic map keeping track of the
active and the standby ACOS device for each traffic bucket. Sessions get synced
between the Active and the Standby devices for each traffic bucket. And traffic gets
internally redirected using the traffic map for the bucket.

This traffic map is updated and synced across the ACOS devices whenever there is any
change to the cluster (device gets added/removed).

The incoming traffic streams are classified and distributed to the devices within a clus-
ter for processing. Incoming traffic is distributed based on a hash of the source IP. See
“Understanding Traffic Distribution” on page 9 for more information.

A traffic map shows how the incoming traffic is distributed across devices. You can use
the show scaleout trafifc-map command to view information about the traffic
maps on each device. For more information, see show scaleout.
Cluster Master A single device in the cluster that keeps track of all other devices in that cluster.

The cluster master is responsible for programming the various network elements to
classify and re-distribute traffic as required among the various devices in the cluster. It
also keeps track of service nodes leaving and joining the cluster and takes appropriate
actions. See Election of the Cluster Masterfor more information.
Service Nodes Each device in the Scaleout cluster is a service node. All service nodes process incom-
ing traffic and each node acts as a traffic classification and distribution engine. The
upstream router or switch may forward traffic to one of the devices in the cluster based
on routing or ARP responses. The devices inspect the packet and may redirect it to
other nodes within the cluster if the state for handling that packet is on another node.

The devices also serve (provide) the corresponding virtual service for the packet’s desti-
nation. The mapping of traffic segments and their owners is created by the cluster mas-
ter and pushed to all service nodes.

The Cluster Master and the Service Nodes are also more generically referred to as Clus-
ter nodes.
Device Group A group of devices in the cluster that provide a particular cluster service.

Each VIP, for example, does not need to be scaled out to all the service nodes in a clus-
ter. A subset of cluster nodes can be defined as a “device group,” which can then be
bound to a VIP being scaled. The service nodes in the device group will handle traffic
related to that VIP.

For example, Device 1 and Device 2 form a device group for HTTP traffic on the VIP.
Similarly, Device 2 and Device 3 form another device group for TCP traffic on the VIP.

NOTE: Throughout this document, devices in the cluster are referred to as

“ACOS devices”. However, Scaleout is only supported on physical A10
Thunder Series and vThunder devices. Scaleout is not supported on AX
Series devices.

page 8
 ACOS 4.1.1-P11 Configuring Scaleout
Understanding Traffic Distribution

Understanding Traffic Distribution

Figure 2shows a topology illustrating traffic distribution in a Scaleout topology. This type of traffic dis-
tribution is called Symmetric Distribution of Traffic.

FIGURE 2 Scaleout Traffic Distribution - Symmetric Distribution

Distributing Traffic from Upstream Devices

Traffic from upstream devices is distributed across the devices in the cluster based on a hash of the
client IP address.

The entire set of traffic destined for a cluster service is classified into multiple subsets, or buckets:

• Each traffic bucket is assigned to a single ACOS device.

• Each ACOS device can have multiple buckets assigned to it.

• Different buckets from different cluster services may be assigned to a single ACOS device. This is
useful for properly distributing traffic destined for a particular cluster service across service
nodes.

page 9
ACOS 4.1.1-P11 Configuring Scaleout
Election of the Cluster Master

The cluster devices may be connected to the rest of the network via an upstream router or a switch.
Based on how that device forwards the traffic (for example, a router might have a default route for a
VIP in a different subnet or ARP entry for a VIP in the same subnet), it will reach one of the cluster
nodes. The cluster node may need to redirect the incoming packet based on a hash of the client IP if
that node is not the owner of the set of traffic for the client IP.

All cluster nodes must be Layer-2 connected with high throughput links so that any redirection of pack-
ets can be done efficiently.

Using NAT to Handle Return Traffic

For the return traffic from the servers, a device-level NAT pool is used. Each incoming flow is source-
NAT’ed to ensure that the return traffic always reaches the node that owns that traffic set. Source NAT
configuration is mandatory for Scaleout to work. NAT pool configuration has been enhanced to include
a device-id so that it can be bound to a given device. Each device in the cluster will have its own Source
NAT pool.

To configure NAT pools for Scaleout, all NAT pools must be configured on all devices. The example in
“Configure Scaleout IP NAT Pool Group on the vMaster describes this configuration.

Election of the Cluster Master

Each ACOS device can be configured with a priority value that is used during the election of the cluster
master. Devices with higher priority have precedence over devices with lower priorities. Among devices
that have equal priority value, any one of them can potentially be selected as the master.

If you have VCS configured, you can also use the follow-vcs command to use the priorities and device
IDs already configured for VCS.

Scaleout Configuration Example

This section shows how to configure Scaleout in a cluster of three devices using VCS incremental sync.

Figure 3 illustrates a sample topology that will serve as the basis for the configuration example in this
section.

page 10
 ACOS 4.1.1-P11 Configuring Scaleout
Scaleout Configuration Example

FIGURE 3 Topology for the Scaleout Configuration Example

NOTE: This example makes the assumption that basic network configuration
for your device is already in place; this aspect of the configuration is not
covered in this example.

The overall steps for the configuration are:

1. Configure VCS on each device. (See Configure VCS on Each Device.)

For Scaleout to function, the Scaleout-related configuration on all devices in the cluster must be
duplicated and synchronized. To accomplish this, you can use VCS to automatically synchronize
the configurations on all devices. Alternatively, if you choose not to use VCS, then you must manu-
ally replicate the configuration on all devices.
2. Set up the Scaleout configuration on the vMaster. (See Configure Scaleout on the vMaster.)
With VCS configured and enabled, configuration changes on the vMaster (Device 3) are automati-
cally synchronized to the vBlades (Device 2 and Device 1).

page 11
ACOS 4.1.1-P11 Configuring Scaleout
Scaleout Configuration Example

3. Set up the Scaleout NAT Pool Group on the vMaster. (See Configure Scaleout IP NAT Pool Group
on the vMaster.)
4. Set up the SLB configuration on the vMaster. (See Configure SLB on the vMaster.)

Configure VCS on Each Device

This section contains the following:

• Configure VCS on Device 1

• Configure VCS on Device 2

• Configure VCS on Device 3

Configure VCS on Device 1

To configure VCS on device 1, use the commands in this section.

1. Specify the VRRP-A device ID and set ID using the following commands:
ACOS(config)# hostname ACOS1
ACOS1(config)# vrrp-a common
ACOS1(config-common)# set-id 8
ACOS1(config-common)# device-id 1
ACOS1(config-common)# enable
ACOS1(config-common)# exit

2. Enable VCS:
ACOS1(config)# vcs enable

3. Configure the floating IP address for the virtual chassis:

ACOS1(config:1)# vcs floating-ip 10.6.14.7 /24
The changed configuration of aVCS will take effect only after 'vcs reload'

4. Configure the VCS profile for the device:

ACOS1(config:1)# vcs device 1
ACOS1(config:1-device:1)# enable
ACOS2(config:2-device:1)# interface management
ACOS1(config:1-device:1)# priority 100
ACOS1(config:1-device:1)# exit
ACOS1(config:1)# write memory
Building configuration...
Write configuration to primary default startup-config
[OK]
ACOS1(config:1)# vcs reload
ACOS1#

page 12
 ACOS 4.1.1-P11 Configuring Scaleout
Scaleout Configuration Example

Configure VCS on Device 2

To configure VCS on device 2, use the commands in this section.

1. Specify the VRRP-A device ID and set ID using the following commands:
ACOS(config)# hostname ACOS2
ACOS2(config)# vrrp-a common
ACOS2(config-common)# set-id 8
ACOS2(config-common)# device-id 2
ACOS2(config-common)# enable
ACOS2(config-common)# exit

2. Enable VCS:
ACOS2(config)# vcs enable

3. Configure the floating IP address for the virtual chassis:

ACOS2(config:2)# vcs floating-ip 10.6.14.7 /24
The changed configuration of aVCS will take effect only after 'vcs reload'

4. Configure the VCS profile for the device:

ACOS2(config:2)# vcs device 2
ACOS2(config:2-device:2)# enable
ACOS2(config:2-device:2)# interface management
ACOS2(config:2-device:2)# priority 110
ACOS2(config:2-device:2)# exit
ACOS2(config:2)# write memory
Building configuration...
Write configuration to primary default startup-config
[OK]
ACOS2(config:2)# vcs reload
ACOS2#

Configure VCS on Device 3

To configure VCS on device 3, use the commands in this section.

1. Specify the VRRP-A device ID and set ID using the following commands:
ACOS(config)# hostname ACOS3
ACOS3(config)# vrrp-a common
ACOS3(config-common)# set-id 8
ACOS3(config-common)# device-id 3
ACOS3(config-common)# enable
ACOS3(config-common)# exit

2. Enable VCS:
ACOS3(config)# vcs enable

3. Configure the floating IP address for the virtual chassis:

ACOS3(config:3)# vcs floating-ip 10.6.14.7 /24

page 13
ACOS 4.1.1-P11 Configuring Scaleout
Configure Scaleout on the vMaster

The changed configuration of aVCS will take effect only after 'vcs reload'

4. Configure the VCS profile for the device:

ACOS3(config:3)# vcs device 3
ACOS3(config:3-device:3)# enable
ACOS2(config:2-device:3)# interface management
ACOS3(config:3-device:3)# priority 120
ACOS3(config:3-device:3)# exit
ACOS3(config:3)# write memory
Building configuration...
Write configuration to primary default startup-config
[OK]
ACOS3(config:3)# vcs reload
ACOS3#

Configure Scaleout on the vMaster

This section contains the following:

• Local Cluster Configuration

• Cluster Node Configuration

• Device Groups Configuration

Local Cluster Configuration

Configure the Scaleout cluster. In a VCS environment, you can use the follow-vcs command to use the
device ID and priority settings from your VCS configuration.

ACOS1-vMaster[8/1](config:3)# scaleout 1
ACOS1-vMaster[8/1](config:3-scaleout:1)# follow-vcs
ACOS1-vMaster[8/1](config:3-scaleout:1)#

If you are using VCS and do not necessarily want the vMaster device to be the same as the cluster
master, do not use the follow-vcs command.

Without VCS, you must manually specify the Scaleout device ID and priority settings. For example, in
Device 3:

ACOS3(config)# scaleout 1
ACOS3(config-scaleout:3)# local-device
ACOS3(config-scaleout:3-local-device)# priority 120
ACOS3(config-scaleout:3-local-device)# id 3
ACOS3(config-scaleout:3-local-device)# exit

page 14
 ACOS 4.1.1-P11 Configuring Scaleout
Configure Scaleout on the vMaster

ACOS3(config-scaleout:3)# exit
ACOS3(config)#

Cluster Node Configuration

This section shows how to configure each node in the cluster. On the vMaster, enter the following con-
figuration:

ACOS1-vMaster[8/1](config:3)# scaleout 1
ACOS1-vMaster[8/1](config:3-scaleout:1)# cluster-devices
ACOS1-vMaster[8/1](config:3-scaleout:1-cluster-devices)# device-id 1
ACOS1-vMaster[8/1](config:3-scaleout:1-cluster-devices-de...)# ip 192.168.230.56
ACOS1-vMaster[8/1](config:3-scaleout:1-cluster-devices-de...)# exit
ACOS1-vMaster[8/1](config:3-scaleout:1-cluster-devices)# device-id 2
ACOS1-vMaster[8/1](config:3-scaleout:1-cluster-devices-de...)# ip 192.168.230.57
ACOS1-vMaster[8/1](config:3-scaleout:1-cluster-devices-de...)# exit
ACOS1-vMaster[8/1](config:3-scaleout:1-cluster-devices)# device-id 3
ACOS1-vMaster[8/1](config:3-scaleout:1-cluster-devices-de...)# ip 192.168.230.58
ACOS1-vMaster[8/1](config:3-scaleout:1-cluster-devices-de...)# exit
ACOS1-vMaster[8/1](config:3-scaleout:1-cluster-devices)# exit
ACOS1-vMaster[8/1](config:3-scaleout:1)#

In non-VCS environments, this configuration must be manually duplicated on all devices in the cluster.

NOTE: The IP address of each device represents the management IP address of

the device.

Device Groups Configuration

This section shows how to configure the device groups; in our example, we have only one device group,
with all three devices in the cluster serving a single VIP. On the vMaster, enter the following configura-
tion:

ACOS1-vMaster[8/1](config:3)# scaleout 1
ACOS1-vMaster[8/1](config:3-scaleout:1)# device-groups
ACOS1-vMaster[8/1](config:3-scaleout:1-device-groups)# device-group 1
ACOS1-vMaster[8/1](config:3-scaleout:1-device-groups-devi...)# device-id 1 to 3
ACOS1-vMaster[8/1](config:3-scaleout:1-device-groups-devi...)# exit
ACOS1-vMaster[8/1](config:3-scaleout:1-device-groups)# exit
ACOS1-vMaster[8/1](config:3-scaleout:1)# exit
ACOS1-vMaster[8/1](config:3)#

page 15
ACOS 4.1.1-P11 Configuring Scaleout
Configure Scaleout on the vMaster

Configure Scaleout IP NAT Pool Group on the vMaster

This section provides the configuration to set up the NAT pool group on the vMaster. For more informa-
tion about NAT pools in Scaleout configurations, see Using NAT to Handle Return Traffic.

ACOS1-vMaster[8/1](config:3)# ip nat pool p1 2.2.2.240 2.2.2.240 netmask /24 scaleout-

device-id 1
ACOS1-vMaster[8/1](config:3)# ip nat pool p2 2.2.2.241 2.2.2.241 netmask /24 scaleout-
device-id 2
ACOS1-vMaster[8/1](config:3)# ip nat pool p3 2.2.2.242 2.2.2.242 netmask /24 scaleout-
device-id 3
ACOS1-vMaster[8/1](config:3)# ip nat pool-group pg
ACOS1-vMaster[8/1](config:3-pool-group:pg)# member p1
ACOS1-vMaster[8/1](config:3-pool-group:pg)# member p2
ACOS1-vMaster[8/1](config:3-pool-group:pg)# member p3
ACOS1-vMaster[8/1](config:3-pool-group:pg)# exit
ACOS1-vMaster[8/1](config:3)#

In non-VCS environments, this configuration must be manually duplicated on all devices in the cluster.

Configure SLB on the vMaster

This section contains the following:

• Configure the Service Templates

• Associate the Templates with the Services

Configure the Service Templates

To configure the service templates:

ACOS1-vMaster[8/1](config:3)# scaleout 1
ACOS1-vMaster[8/1](config:3-scaleout:1)# service-config
ACOS1-vMaster[8/1](config:3-scaleout:1-service-config)# template adc1
ACOS1-vMaster[8/1](config:3-scaleout:1-service-config-tem...)# device-group 1
ACOS1-vMaster[8/1](config:3-scaleout:1-service-config-tem...)# exit
ACOS1-vMaster[8/1](config:3-scaleout:1-service-config)# exit
ACOS1-vMaster[8/1](config:3-scaleout:1)#

Associate the Templates with the Services

After configuring a Scaleout service template, apply it to the service. In our example, the template
“adc1” will be applied to the VIP and it’s service. Specify “pg” for the NAT pool group that was config-
ured in “Configure Scaleout IP NAT Pool Group on the vMaster.

page 16
 ACOS 4.1.1-P11 Configuring Scaleout
Graceful Addition and Removal of Devices

ACOS1-vMaster[8/1](config:3)# slb virtual-server s1_vs 1.1.1.100

ACOS1-vMaster[8/1](config:3-slb vserver)# template scaleout adc1
ACOS1-vMaster[8/1](config:3-slb vserver)# port 80 http
ACOS1-vMaster[8/1](config:3-slb vserver-vport)# source-nat pool pg

Graceful Addition and Removal of Devices

Devices can be gracefully added or removed from the cluster.

• Graceful Addition of a Device to the Cluster

• Graceful Shutdown of a Device in the Cluster

Graceful Addition of a Device to the Cluster

By default, all device additions to a cluster are graceful; either when a new device joins an existing clus-
ter, or a previously disabled device is re-enabled.

In both cases:

1. The device notifies the cluster master that it wants to join the cluster.
2. The cluster master generates a new potential bucket map with the new device as part of the map-
pings.
If any of the existing devices in the cluster have open sessions for which this new device will be the
new active device, the sessions are synced to this device.
3. Finally, after a short delay, the new device takes over the sessions and starts processing the traffic
actively.

Suppose we want to add a device to the Scaleout cluster defined in Figure 3. To accomplish this:

1. Add the device-specific configuration for device 4 (the device being added to the cluster) to each
device in the cluster:
ACOS4(config)# scaleout 1
ACOS4(config-scaleout:4)# device-context 4
ACOS4(config-scaleout:4)# local-device
ACOS4(config-scaleout:4-local-device)# priority 4
ACOS4(config-scaleout:4-local-device)# id 4

ACOS4(config)# scaleout 1
ACOS4(config-scaleout:4)# cluster-devices
ACOS(config-scaleout:4-cluster-devices)# device-id 4

page 17
ACOS 4.1.1-P11 Configuring Scaleout
Graceful Addition and Removal of Devices

ACOS(config-scaleout:4-cluster-devices-de...)# ip 10.6.14.14

2. Add the IP NAT pool for device 4 and make it a member of the NAT pool group pg. The updated
NAT pool group should look like this on all devices:
ACOS4(config:1)# ip nat pool p1 20.0.0.11 20.0.0.11 netmask /24 scaleout-device-id 1
ACOS4(config:1)# ip nat pool p2 20.0.0.12 20.0.0.12 netmask /24 scaleout-device-id 2
ACOS4(config:1)# ip nat pool p3 20.0.0.13 20.0.0.13 netmask /24 scaleout-device-id 3
ACOS4(config:1)# ip nat pool p4 20.0.0.14 20.0.0.14 netmask /24 scaleout-device-id 4

ACOS4(config:1)# ip nat pool-group pg

ACOS4(config:1-pool-group:pg)# member p1
ACOS4(config:1-pool-group:pg)# member p2
ACOS4(config:1-pool-group:pg)# member p3
ACOS4(config:1-pool-group:pg)# member p4

3. Configure device 4 so that its configuration contains the same cluster information as the other
devices in the cluster. You will need to add the configuration for device 3, device 2, and device 1 on
device 4.

Graceful Shutdown of a Device in the Cluster

In a graceful shutdown scenario:

1. The device notifies the cluster master that it will be shut down. This triggers the cluster master to
assign a new owner for each bucket.
2. The old owner (device going to be shut down) performs a session sync to the new owner for all
Layer 4 connections.
3. The cluster master updates the bucket mappings and pushes the configuration to all active
devices in the cluster.
4. The admin should wait for all layer 7 sessions on the device to close, at which point it is safe to
shut down the device.

A CLI example of a graceful shutdown is show below:

ACOS(config)# scaleout 64
ACOS(config-scaleout:64)# cluster-devices
ACOS(config-scaleout:64-cluster-devices)# device-id 3
ACOS(config-scaleout:64-cluster-devices-d...)# disable

page 18
 ACOS 4.1.1-P11 Configuring Scaleout
Scaleout Configuration Notes

Scaleout Configuration Notes

Below are some additional features and limitations for this feature:

• The devices providing the services in the Scaleout feature must be configured together in a clus-
ter. Up to eight devices can be configured in the cluster, and all devices must be Layer 2 adjacent.
• Each device in the cluster must have unique IP and MAC addresses.

• Only a single cluster is supported.

• Only IPv4 services are available.

• Virtual IP and virtual port elasticity are supported.

• Support for Layer 2 redirection, Layer 4, Fast-HTTP, and basic Layer 7 functionality is provided.
Network SNMP is not supported.
• Support for per-device NAT is available.

• Support for session synchronization is available.

• Support for Scaleout control plane connectivity is available through the management port or data
ports.

page 19
ACOS 4.1.1-P11 Configuring Scaleout
Scaleout Configuration Notes

page 20
 ACOS 4.1.1-P11 Configuring Scaleout

Config Commands: Scaleout

This chapter contains the following:

• Scaleout Global Configuration Commands

• Scaleout Configuration Commands

• Scaleout Local Device Configuration Commands

• Scaleout show Commands

page 21
ACOS 4.1.1-P11 Configuring Scaleout

Scaleout Global Configuration Commands

The scaleout command is used at the global configuration level to enter Scaleout configuration mode.

scaleout
Description Enter Scaleout configuration mode.

Syntax [no] scaleout cluster-id

Replace cluster-id with the ID of the cluster you are creating or configuring.
Each ACOS device can only be a member of one cluster.

Mode Configuration mode

Example Enter Scaleout configuration mode for cluster 3.

ACOS(config)# scaleout 3
ACOS(config-scaleout:3)#

Scaleout Configuration Commands

This section describes the command available in Scaleout configuration mode.

To enter Scaleout configuration mode, use the scaleout command:

ACOS(config)# scaleout 3
ACOS(config-scaleout:3)#

The following commands are available:

• cluster-devices

• device-groups

• follow-vcs

• local-device

• service-config

page 22
 ACOS 4.1.1-P11 Configuring Scaleout

cluster-devices
Description Enter Scaleout cluster configuration mode to configure devices in a cluster.

Syntax cluster-devices

This command places you in another sub-configuration mode, where you

can use the following command to add devices to the cluster:

Mode Scaleout configuration mode.

Example The following example shows how to enter cluster configuration mode for
cluster 3, then add three devices to the cluster. The IP addresses of the
devices represent each device’s management IP address. In a VCS environ-
ment, you can also use the floating IP of the virtual chassis:

ACOS(config)# scaleout 3
ACOS(config-scaleout:3)# cluster-devices
ACOS(config-scaleout:3-cluster-devices)# device-id 1
ACOS(config-scaleout:3-cluster-devices-de...)# ip 192.168.230.56
ACOS(config-scaleout:3-cluster-devices-de...)# device-id 2
ACOS(config-scaleout:3-cluster-devices-de...)# ip 192.168.230.57
ACOS(config-scaleout:3-cluster-devices-de...)# device-id 3
ACOS(config-scaleout:3-cluster-devices-de...)# ip 192.168.230.58

device-groups
Description Enter device groups configuration mode to configure Scaleout device
groups.

Syntax [no] device-groups

This command places you in another sub-configuration mode, where you

can use the following command to configure device groups in the cluster:
[no] device-group num

After this command, use the device-id command to add devices to the
device group:
[no] device-id device-id [to device-id]

You can specify a single device or a range of devices using the to option.

Mode Scaleout configuration mode.

Example The following example shows how to enter device groups configuration
mode for cluster 3, then creates a device group and adds a range of devices
(1 to 3) to the group:

page 23
ACOS 4.1.1-P11 Configuring Scaleout

ACOS(config)# scaleout 3
ACOS(config-scaleout:3)# device-groups
ACOS(config-scaleout:3-device-groups)# device-group 1
ACOS(config-scaleout:3-device-groups-devi...)# device-id 1 to 3

follow-vcs
Description Follow VCS priority and device ID configuration settings.

Syntax [no] follow-vcs

Mode Scaleout configuration mode

Example The following example shows how to disable the follow VCS priority.

ACOS(config)# scaleout 3
ACOS(config-scaleout:3)# no follow-vcs

local-device
Description Enter Scaleout local device configuration mode to configure the local device
in a cluster.
Syntax [no] local-device

Mode Scaleout configuration mode.

Usage After using this command, the commands in “Scaleout Local Device Config-
uration Commands” on page 25 are available.

Example The following example shows how to enter local device configuration mode
for cluster 3 and set the priority of the local device:

ACOS(config)# scaleout 3
ACOS(config-scaleout:3)# local-device
ACOS(config-scaleout:3-local-device)# pripority 1

service-config
Description Enter service configuration mode to configure Scaleout templates.

Syntax service-config

This command places you in another sub-configuration mode, where you

can use the following command to create a template:
[no] template name

page 24
 ACOS 4.1.1-P11 Configuring Scaleout

Replace name with the name of the template.

Mode Scaleout configuration mode.

Example The following example shows how to enter service configuration mode for
cluster 3, and then create a template called “so_template” with a bucket
count of 128:

ACOS(config)# scaleout 3
ACOS(config-scaleout:3)# service-config
ACOS(config-scaleout:3-service-config)# template so_template
ACOS(config-scaleout:3-service-config-tem...)# bucket-count 128

Scaleout Local Device Configuration Commands

This section describes the command available in Scaleout local device configuration mode.

To enter this configuration mode, use the local-device command:

ACOS(config)# scaleout 3
ACOS(config-scaleout:3)# local-device
ACOS(config-scaleout:3-local-device)#

The following commands are available:

• id

• priority

id
Description Assign an ID to the local device.

Syntax [no] id num

Replace num with an ID for the device.

Default No default ID is assigned.

Mode Scaleout local device configuration

Example The following example shows how to assign an ID of 4 to the local device.

ACOS(config)# scaleout 3

page 25
ACOS 4.1.1-P11 Configuring Scaleout

ACOS(config-scaleout:3)# local-device
ACOS(config-scaleout:3-local-device)# id 4

priority
Description Assign a priority to the local device.

Syntax [no] priority num

Replace num with an ID for the device.

Default No default priority is assigned.

Mode Scaleout local device configuration

Example The following example shows how to assign a priority of 150 to the local
device.

ACOS(config)# scaleout 3
ACOS(config-scaleout:3)# local-device
ACOS(config-scaleout:3-local-device)# priority 150

Scaleout show Commands

The show scaleout command can be used to view information about your Scaleout configuration.

show scaleout
Description View Scaleout information.

Syntax show scaleout

[fixed-nat {inside-user address | nat-address address}]
[nat-pool [word]]
[statistics]
[traffic-map name
[vport-num]
[check-data-plane]

page 26
 ACOS 4.1.1-P11 Configuring Scaleout

[src-ip ipv4-addr/mask-length]
]

Parameter Description
fixed-nat Show Scaleout fixed NAT mapping:

• inside-user - IPv4 or IPv6 address

• nat-address - IPv4 address
nat-pool Show Scaleout NAT pool resources:

• word - NAT pool name

Using the nat-pool command will show all the NAT pool
names, addresses, active devices, and user groups. Using
the command plus the optional name, such as nat-pool
p1, will show only the resources for the specified pool.
statistics Show Scaleout statistics.
traffic-map View all traffic-maps, or view a traffic-map for a specific
service name.

• vport-num - Virtual port number for the service.

• check-data-plane - Check to see if the traffic map is

present in the data plane; this command is used for
debugging purposes only.

• src-ip - View active and standby status for the speci-

fied source IP address.

Mode All

Example The following example shows sample output for the show scaleout com-
mand:

ACOS# show scaleout

Role - Cluster Master

Device 6 - Active
Device 7 - Active
Device 8 - Active (Local)

page 27
ACOS 4.1.1-P11 Configuring Scaleout

The following table describes the fields in this output:

Field Description
Role Shows the role of the local device (the device on which you are
executing the show scaleout command:

• Cluster Master - the local device has the highest priority of all
devices in the cluster.

• Standby Node - the local device does not have the highest pri-
ority of the devices in the cluster.

• Unknown Node - the local device is not recognized as being

part of a Scaleout cluster.
Device Shows a list of devices in the cluster. The device marked with
“(Local)” is the local device; the device from which you are run-
ning the show scaleout command.

Example The following example shows sample output for show scaleout traffic-map
with a specific service s1_vs:

ACOS# show scaleout traffic-map s1_vs

Virtual Server=s1_vs :

Bucket Active Device Standby Device New Act Device New Stby Device
0 8 6 - -
1 8 6 - -
2 6 8 - -
3 6 8 - -
4 6 7 - -
5 6 7 - -
6 6 7 - -

For bucket 0, the active device is device 8, and the standby device is device
6.

Example The following example shows sample output for show scaleout traffic-map
with a specific source IP address:

ACOS# show scaleout traffic-map s1_vs src-ip 1.1.1.2

Virtual Server=s1_vs :Bucket Active Device Standby Device New Act Device New Stby
Device
2 6 8 - -

Example The following example shows sample output for show scaleout traffic-map
with a specific service s1_vs:

page 28
 ACOS 4.1.1-P11 Configuring Scaleout

ACOS# show scaleout traffic-map s1_vs

Example The following example shows sample output for show scaleout nat-pool.

ACOS4-vMaster[8/4](config:4)# show scaleout nat-pool

cp1 Address Active Device User-Group
-----------------------------------------------------------------
50.50.50.1 1 1
50.50.50.2 4 2
50.50.50.3 1 3
50.50.50.4 3 4
50.50.50.5 1 5
50.50.50.6 4 6
50.50.50.7 2 7
50.50.50.8 4 8
50.50.50.9 2 9
50.50.50.10 1 10

Example The following example shows sample output for show scaleout fixed-nat
inside-user.

ACOS4-vMaster[8/4](config:4)# show scaleout fixed-nat inside-user 2001:abcd::1

Address Active Device User-Group
------------------------------------------------------------------------------
2001:abcd::1 1 1

Example The following example shows sample output for show scaleout fixed-nat
nat-address.

ACOS4-vMaster[8/4](config:4)# show scaleout fixed-nat nat-address 50.50.50.7

Address Active Device User-Group
-----------------------------------------------------------------
50.50.50.7 2 7

page 29
ACOS 4.1.1-P11 Configuring Scaleout

page 30
 ACOS 4.1.1-P11 Configuring Scaleout

page 31
 CONTACT US
3 a10networks.com/contact

ACOS 4.1.1-P11 CONFIGURING SCALEOUT 29 MAY 2019