MobileFirstBaseDesignsLab ArubaOS8
MobileFirstBaseDesignsLab ArubaOS8
ArubaOS 8
Authors:
Ben Lowe
Andrew Tanguay
[Link]
Fax 408.227.4550
Contents
Contents
Topology ............................................................................................. 14
Recommended Architecture ................................................................................................ 14
Test Lab Network Topology ................................................................................................. 16
Clustering ........................................................................................... 35
Benefits and Considerations................................................................................................ 36
Local Management Switch ................................................................................................... 37
Cluster Roles ......................................................................................................................... 39
Change of Authorization ...................................................................................................... 43
Configuration ..................................................................................... 67
SW-TOR-LAB ......................................................................................................................... 68
SW-Core ................................................................................................................................ 71
SW-Access-01 ........................................................................................................................ 76
MM-01 Initial Setup .............................................................................................................. 77
MM-02 Initial Setup .............................................................................................................. 78
MM-01 Redundancy ............................................................................................................. 79
MM-02 Redundancy ............................................................................................................. 80
MM-01 Licenses .................................................................................................................... 81
MC-01 Initial Setup ............................................................................................................... 83
MC-02 Initial Setup ............................................................................................................... 85
MC-03 Initial Setup ............................................................................................................... 87
MC-04 Initial Setup ............................................................................................................... 89
MM-01 Remaining Configuration ........................................................................................ 91
Employee BYOD SSID Configuration ................................................................................... 98
HTTPS Server Certificate for MCs ...................................................................................... 102
BYOD Employee ClearPass Configuration......................................................................... 104
BYOD Employee Onboard Configuration .......................................................................... 124
PSK SSID Configuration ...................................................................................................... 132
AP Configuration ................................................................................................................ 133
Scope
The Validated Reference Design series documents focus on particular aspects of Aruba
technologies and deployment models. Together these guides provide a structured framework to
understand and deploy Aruba Wireless Local Area Networks (WLANs). The VRD series has four
document categories:
• Foundation guides explain the core technologies of an Aruba WLAN. These guides also
describe different aspects of planning, operation, and troubleshooting deployments
• Base Design guides describe the most common deployment models, recommendations,
and configurations
• Application guides build on the base designs. These guides deliver specific information
that is relevant to deploying particular applications such as voice, video, or outdoor campus
extension.
• Specialty Deployment guides involve deployments in conditions that differ significantly
from the common base design deployment models, such as high-density WLAN
deployments.
The Mobile First Base Designs Lab VRD is considered a Base Design guides within the VRD core
technology series. The design described in this document is intended for single stack IPv4 users
with small to medium size campus deployments. For additional details on designs for different
sized deployments please refer to the Reference Architecture chapter of the ArubaOS 8
Fundamentals Guide.
Reference Material
Readers should have a solid working understanding of basic wireless LAN concepts as well as the
Aruba technology explained in the foundation level guides in order to read this VRD. The following
resources will assist readers who require the knowledge necessary to digest this document in the
intended manner:
• For information on Aruba Mobility Controllers and deployment models, please refer to the
Aruba Mobility Controllers and Deployment Models Validated Reference Design
• The complete suite of Aruba technical documentation is available for download from the
Aruba Support Site. These documents present complete, detailed feature and functionality
explanations beyond the scope of the VRD series.
• For more training on Aruba products or to learn about Aruba certifications, please visit the
Aruba Training and Certification page. This page contains links to class descriptions,
calendars, and test descriptions.
• Aruba hosts a user forum site and user meetings called Airheads Community. The forum
contains discussions of deployment best practices, products, and troubleshooting tips.
Airheads is an invaluable resource that allows network administrators to interact with each
other and Aruba experts.
Conventions
Typographical Conventions
The following conventions are used throughout this manual to emphasize important concepts:
Bolded words indicate an option that should be selected in the Graphical User
Bolded > words Interface (GUI). The angled brackets indicate that the choices are part of a path
in the GUI.
Command text in this font will appear inside of a box and indicates commands
Command Text
that can be entered into the Command Line Interface (CLI).
In this example, you would type “send” at the system prompt exactly as shown,
followed by the text of the message you wish to send. Do not type the angle
brackets.
Graphical Icons
Lab Topology
Recommended Architecture
The Mobile First Base Designs Lab for ArubaOS 8 VRD will enable readers to quickly setup and
configure a state-of-the-art campus network powered by ArubaOS 8. The Aruba campus lab
network consists of the following key components:
• Data Center
• Wired LAN (collapsed core design)
Core/Aggregation
Access
• Demilitarized Zone (DMZ)
The data center (DC) contains all network services and applications along with two fully-redundant
virtual mobility masters (VMMs) and four mobility controllers (MC) configured as a cluster.
There are many other possible design scenarios for ArubaOS 8 other than the architecture described above. Please
refer to the “Controller Reference Architectures” section of ArubaOS 8 Fundamentals Guide for details.
The wired LAN employs a 2-tier design consisting of dual core switches with VSF (virtual switching
framework) redundancy directly connected to aggregation switches for access purposes. The
DMZ’s role in the campus topology design is to facilitate a separate domain of guest clients
through two MCs, a minimal aggregation layer, and a firewall at the LAN-Internet edge. The
topology as a whole utilizes device and link redundancy, hierarchical configuration, and guest
traffic isolation through the Multizone feature.
This same network architecture design is frequently used at Aruba for the testing and validation of
various network features. It is considered the Aruba base design for ArubaOS 8. Subsequent
chapters of this VRD will describe the network design in greater detail including device roles,
virtual local area network (VLAN) assignment, redundancy, hierarchical configuration, and guest
traffic isolation.
This topology uses existing services running in the corporate server farm. These services include
domain controller, Active Directory (AD), Domain Name System (DNS), Dynamic Host
Configuration Protocol (DHCP), and three Skype for Business (SfB) servers.
The topology depicted below represents the recommended architecture for ArubaOS which has
been thoroughly validated by Aruba Technical Marketing Engineering. The actual network which
was used for the validation in the TME lab is provided for reference purposes under the Test Lab
Network Topology section below.
ARUB AO S 8 MO BILITY
Mobility Controller Cluster (VRRP)
VIRTUAL MACHINES
CONTROLLERS
CONTROLLERS
ESXi SERVERS
MM-01 MM-02
MACHINES
MC-01 MC-02 MC-03 MC-04
AD/DNS/
(7210) (7210) (7210) (7210) CP AW
DHCP
Interne t
GATEWAY
GATEWAY
INTERNET
DMZ INET
DMZ INET
GATEWAY
GATEWAY
INTERNET
CORE
CORE
SW-CORE-STACK
FIREWALL FIREWALL-DMZ
(5400R)
AGGREGATION
AGGREGATION
AGGREGATION
AGGREGATION
DMZ
SW-AGG-1 DMZ SW-AGG-DMZ-1
(3810M) (2930F)
VRRP
DMZ STANDALONE
DMZ STANDALONE
ACCESS POINTS
ACCESS POINTS
CONTROLLERS
CONTROLLERS
DMZ SERVERS
DMZ SERVERS
VRRP
ARUB AO S 8 MO BILITY
Mobility Controller Cluster (VRRP)
VIRTUAL MACHINES
CONTROLLERS
CONTROLLERS
ESXi SERVERS
MM-01 MM-02
MACHINES
MC-01 MC-02 MC-03 MC-04
AD/DNS/
(7210) (7210) (7210) (7210) CP AW
DHCP
Aruba IT
GATEWAY
DMZ INET
DMZ INET
GATEWAY
CORE
CORE
SW-CORE-STACK SW-TOR-LAB
(2930F) FIREWALL-DMZ
(5400R)
AGGREGATION
AGGREGATION
AGGREGATION
AGGREGATION
DMZ
DMZ
SW-AGG-1 SW-AGG-DMZ-1
(3810M) (2930F)
VRRP
DMZ STANDALONE
DMZ STANDALONE
ACCESS POINTS
ACCESS POINTS
CONTROLLERS
CONTROLLERS
DMZ SERVERS
DMZ SERVERS
Hierarchical Configuration
ArubaOS 8 Controller Modes
Mobility Master
The concept of the Mobility Master (MM) is new to ArubaOS 8. MMs come in two variations: either
Virtual (VMM) or Hardware based (HMM). The MM is designed to run on an x86-based platform as
many of the features introduced in ArubaOS 8 require random access memory (RAM), central
processing unit (CPU), and storage space that are not supported by physical controllers. The MM
must be fully configured by an administrator similar to how a Master controller would be
configured in ArubaOS 6. Its primary role in the Mobile First architecture is to serve as the single
point of configuration and image management for the network. In addition, the MM can be
configured through a Northbound Application Programmable Interface (NBAPI). A VMM can be
installed on VMWare, KVM, or Hyper-V depending on which option is the most ideally suited for
the environment where it will be deployed.
HMMs and VMMs may alternatively be referred to as MM-HW and MM-VA meaning MM-Hardware and MM-Virtual
Appliance, respectively.
The design that was validated for this VRD consists of two MMs which were deployed as virtual
machines (VMs) with master-redundancy enabled. Aruba MMs rely on VRRP as their layer 2
redundancy mechanism. The entire configuration hierarchy is automatically synced from the
active MM to the standby MM with the exception of any configurations under the device
configuration node of the active MM.
As soon as VRRP and master redundancy are configured between the active and standby MMs the
entire configuration hierarchy is synced. Some of the databases that are synced to the standby
MM at periodic intervals when database synchronization is configured on the active MM are listed
below:
• WMS Database
• Local User Database
• Global AP Database
• AirGroup Database
• License Database
• CPSec Database
Once the active and standby MMs have performed their initial synchronization and reached a
stable state, any incremental configuration change that is committed and saved on the active MM
results in a configuration sync with the standby MM.
The exception to this behavior is any change made on the device configuration node of the active
MM (i.e. /mynode). These changes are not synced to the standby MM. The standby MM contains
its own version of the device configuration so any desired changes must be made directly on its
corresponding device configuration node (i.e. /mynode on the standby MM). Configuration
changes for other nodes in the hierarchy are not permitted on the standby MM.
Mobility Controller
The concept of the Mobility Controller or “MC “is also new to ArubaOS 8. An MC, in the past, has
also been known as Managed Node (MN), Managed Device (MD), or Mobility Device (MD) in some
Aruba documentation. An MC is similar to a Branch controller in ArubaOS 6 in the sense that it can
be adopted using zero touch provisioning (ZTP) and Aruba Activate. The last port of an MC is
enabled as a DHCP client on VLAN 4094 in its factory default configuration.
MMs cannot adopt an MC using DHCP Option 43 since MM certificate distribution is not supported with DHCP Options.
VMCs used as VPNCs do not support ZTP when the MM is behind the VPNC due to the lack of a Trusted Platform
Module (TPM). ZTP requires certificate-based authentication for the initial connection to the MM.
Unlike the ArubaOS 6 Local controllers, MCs can be fully managed by an MM or Mobility Controller
Master (MCM). In addition, unlike ArubaOS 6 Branch controllers, an administrator can configure
every feature of an MC. All 70xx series controllers and 72xx series controllers are shipped as MCs.
ArubaOS 8 also supports Virtual MCs (VMC). A VMC can be deployed either on VMWare, KVM, or
Hyper-V. MCs can be configured as Virtual Private Network Concentrators (VPNCs).
System Nodes
System level nodes are present on GUI of the MM by default and cannot be deleted. The system
nodes are as follows:
• MM – In the case of redundant MMs the configuration defined at this node is common for
both active and standby MMs
• Hostname (of MM) – Holds configuration for the actual MM
• Managed Network – Hierarchy under which all the user-defined nodes are created and
controllers are configured
User Nodes
User-defined nodes are created by administrators under the Managed Network system node. A
node hierarchy can be created under this node where the upper nodes hold common
configuration for all controllers. The configuration becomes more specific (based on region,
campus, or building) at lower levels of the hierarchy. The device nodes are defined at the very
bottom. The following examples demonstrate hierarchical group and device node definitions:
Up to four nested child nodes can be created under the Managed Network node. For example:
Numerous child nodes can be created under the same parent node. In addition, child nodes can
be freely moved to other nodes in the hierarchy as well as cloned from other nodes under the
same parent node.
In the example above the initial configuration is created on the user-defined Aruba node which
will be common to all the controllers in the organization. Since the Aruba-Central Node is farther
down the hierarchy it will receive its initial configuration from the Aruba node and then override
the RADIUS IP address. Similarly, the Cenral-Bldg-1 node and the 7210-1 device node will override
the dot1x role and the VLAN ID defined in the original configuration, respectively. The following
figures display some of the key elements which were configured using the Managed Network >
Aruba path:
Please refer to the Configuration Overrides section for a description of items which cannot be overridden.
The presence of the blue dot next to a configuration parameter indicates the value was
overridden such as in the case of a change to the configuration inherited from the parent node or
a blank value that was replaced. Clicking on the blue dot displays additional details about the
change and provides the option to either remove or retain the override.
In the figure below the Aruba-Central node inherited this configuration, however the IP of the
RADIUS server RAD-1 was changed to “[Link]”:
On the Central-Bldg-1 node father down, the presence of the blue dot indicates that the default
802.1X role in the authentication, authorization, and accounting (AAA) profile was changed to
“authenticated” and the configuration received from the parent node was overridden. Managed
Network > Aruba > Aruba-Central > Central-Bldg-1:
Lastly, the VLAN applied to the Virtual AP profile “aruba-employee” on the device node 7210-1 was
changed to “299”. Managed Network > Aruba > Aruba-Central > Central-Bldg-1 > 7210-1:
When the 7210 controller assigned to the Central-Blg-1 node contacts the MM for the first time,
its inherited configuration will result in the following changes:
Original Configuration Inherited Configuration
RADIUS Server IP [Link] [Link]
802.1X Default Role logon authenticated
VLAN 1 299
Licensing Pools
Licensing in ArubaOS 8 is managed centrally from the MM and the global license pool will be used
by default for all controllers under its management. However, if specific license pools need to be
dedicated, such as for a particular region, then custom license pools must be created on the MM
with the appropriate hierarchical node and license counts definitions.
Hierarchical configurations should be designed so that that configurations that are common to
the organization reside on the higher level nodes. The rest of the configuration will be inherited by
the lower nodes of the hierarchy as network requirements become more specific. E.g., a named
VLAN can be defined at a higher level of the hierarchy and then assigned with specific VLAN IDs at
the lower levels. Finally, configurations specific to individual controllers such as IP addresses,
physical and virtual interfaces, and cluster membership are configured at the device level nodes.
As a best practice, all configurations that are dependent on a single node should be always be
defined e.g., defining VLAN ID and VLAN interface parameters together in a common node.
Numerous overrides across many hierarchy levels should be avoided as it can make troubleshooting challenging.
Depth of Hierarchy
Up to four nested child nodes can be created under the Managed Network node. However, it is
recommended to create only as many nested nodes that are needed for purposes of
configuration management simplification.
Aruba strongly discourages placing any configuration on the /md (Managed Network) node under any
circumstances. Modifying configurations at this level will permanently alter the configuration for every child node
without any ability to determine the default settings. Configurations should always begin a level below the Managed
Network node.
Sites can be differentiated either physically or by type. In the example below, if the organization
“Aruba” acquired another company “Network-Co”, then we could simply define a new
configuration node under Managed Network called Network-Co parallel to the Aruba
Sunnyvale node.
Configuration Notes
• When manually bringing up MCs it is important to ensure that they have been whitelisted
on the MM under the appropriate configuration node
• When using ZTP to bring up MCs it is critical to ensure that the correct configuration node
and MM MAC address are configured on Activate.
• Verify that the MM has learned about the MCs from Activate and whitelisted them under
the configuration nodes that were specified in the Activate provisioning rule
• When specifying the MAC address of the MM for establishing an IPsec connection during
initial configuration of controllers, always ensure that the management port hardware MAC
address is used for a VMM and the hardware MAC address is used for an HMM
• When the MM registers with Activate, the correct MAC address is automatically populated.
If controllers are using ZTP to contact Activate and register with the MM, identify the MAC
address of the MM and select it from the dropdown list when configuring the provisioning
rule on Activate
The Mobile First architecture that has been validated and described in this document uses Tunnel Mode.
Tunnel Mode
When operating in the Tunnel forwarding mode, the AP handles all 802.11 association requests
and responses, but sends all 802.11 data packets, action frames, and EAPOL frames over a
Generic Routing Encapsulation (GRE) tunnel to the MC for processing. The MC then removes or
adds the GRE headers, decrypts or encrypts 802.11 frames, and applies firewall rules to the user
traffic as usual.
To achieve maximum performance benefits with Tunnel Mode, end-to-end jumbo frame support
should be enabled on a wired switch due to the increased aggregation introduced with the IEEE
802.11ac standard. Using Control Plane Security (CPSec) in Tunnel Mode is not mandatory. The
majority of production deployments utilize Tunnel Mode for AP forwarding where the AP sends
802.11 traffic to the controller. Control and data plane traffic between the AP and the MC is always
encrypted. Aruba recommends using Tunnel Mode with jumbo frames as a best practice as the
majority of traffic fits in a standard Ethernet frame and no special handling is required on the
wired network to achieve maximum aggregate performance.
Without end-to-end jumbo frames on the wired network, 802.11ac networks can experience
significant performance degradation in some cases. Although, it should be noted that this adverse
impact to performance is only noticed when the peak network performance is measured during
technology demonstrations. The day-to-day operations in real world production networks are
typically unaffected without jumbo frames turned on.
Mobile First Base Designs Lab for ArubaOS 8 Tunneling and Control Plane Security | 31
Figure 24 Tunnel Forwarding Mode
Decrypt-Tunnel Mode
Decrypt-tunnel mode allows an AP-client pair to take full advantage of Aggregated-Media Access
Control (MAC) Service Data Units (A-MSDUs) and Aggregated-MAC Packet Data Units (A-MPDUs)
without requiring the wired network to transport jumbo frames. APs perform decryption and de-
aggregation on themselves locally. It is mandatory to enable control plane security (CPSec)
between APs and Controllers when using Decrypt-tunnel Mode.
Decrypt-tunnel Mode does not provide end-to-end encryption. Only the control plane traffic between APs and MCs is
encrypted in Decrypt-tunnel Mode.
In Decrypt-tunnel mode the AP acts as a bridge between clients and the controller in addition to
performing encryption and decryption. The MC still acts as the aggregation point for terminating
data traffic. This allows the AP-Client pair to take advantage of A-MSDU and A-MPDU on the WLAN
radio side without requiring the wired network to transport the jumbo frames since the AP
performs all assembly aggregation and de-aggregation locally. The payload is then sent to the
controller for firewall processing and L2/L3 forwarding.
Decrypt-tunnel Mode is functionally equivalent to Tunnel Mode with jumbo frames enabled and is
typically used for technology demonstrations. It is important to keep in mind that the AP wireless
chipset performs cryptography for up to 50 clients which is offloaded to the AP hardware.
Scenarios involving more than 50 clients will likely experience minor performance degradation
due to this offload process.
Since CPSec is enabled by default, the MM certifies its MCs using its generated factory certificate
after booting up. MCs in turn certify their APs by signing their factory default certificates. Once the
APs are authorized through the CPSec whitelist and enter the certified-factory-cert state they will
initiate secure PAPI (UDP 8209 inside IPsec) communication with the controller, synchronize their
firmware, and download their configuration.
Clustering
Clustering is one of the key features introduced in ArubaOS 8 and was specifically designed to
capitalize on the MM architecture and deliver maximum value for mission-critical networks.
Clustering was developed to achieve the following objectives:
• Seamless Campus Roaming - Clients in a single large layer 2 domain will associate and
stay anchored to a single MC as they roam. Users will maintain the same subnet and IP
address regardless even if they roam across APs which are anchored to different
controllers. This enables mobility without compromising or sacrificing performance
• Stateful Client Failover - User traffic will remain uninterrupted and high value sessions
will be preserved in the event of a cluster member failure. Clients will not be required to re-
authenticate and there will be no adverse impact to performance. The impact to
performance will be mitigated to such an extent that users will not notice any degradation
in their performance and they will have no knowledge that a failure has even occurred
regardless of the applications they are currently utilizing
• Access Point and Client Load Balancing - APs and users are automatically load balanced
across controllers that are members of the cluster. This process prevents any one MC from
being disproportionately loaded in order to deliver and maintain optimal network
performance as well as to preserve capacity across all cluster members for new client
associations
• Live Upgrade - Aruba allows customers to perform in-service cluster upgrades which allow
improvements to be implemented without affecting performance while the network
remains fully operational. The Live Upgrade feature allows upgrades to be completely
automated. This is a key feature for customers with mission-critical networks that must
remain operational 24/7. Live upgrades can only be performed on MCs in a cluster and the
APs attached to them
All MCs in a cluster need to run the same software version so that APs that failover to a new controller will not
inadvertently upgrade to a new version.
The cluster capacity for each product line is detailed in the table below:
72xx 12
70xx 4
Virtual 4
While it is technically possible to combine 72xx and 70xx devices in the same cluster doing so is
strongly discouraged as a long term deployment option. Such a scenario is acceptable as a
temporary migration strategy however as a best practice cluster devices should always be
homogeneous. If different controller models are clustered together then all controller scalability
limits will be downgraded to the capabilities of the lowest controller model. E.g., if a cluster was
created with two 7240 controllers and one 7210 controller then the cluster’s scalability capacity
will be limited to that of three 7210 controllers.
Virtual and hardware controllers cannot be combined in a cluster under any circumstances.
The view above can be accessed through the GUI by navigating to the Cluster tab of the main
dashboard. Key statistics about a cluster can be seen under this tab including the number of
controllers, APs and clients in the cluster under management by the MM as well as the current AP
and client loads of the cluster members. The Cluster Members section at the bottom displays key
statistics pertaining to the MC which are members of the cluster including their IP address, model,
and which device is acting as the current cluster leader.
AP Anchor Controller
Anchoring is a concept that was introduced in ArubaOS 8 as part of the clustering feature set.
Anchoring and clustering are designed to achieve the following objectives:
• Enhance user mobility through seamless campus roaming
• Prevent any one MC from serving a disproportionate number of APs or users
• Enable redundancy scenarios creating fault tolerance for the cluster and minimizing the
impact of an MC failure
The AP Anchor Controller (AAC) can be thought of as the LMS for any AP that is anchored to it. Each
AP receives the IP address of the LMS and once they have been terminated they will remain
anchored until the cluster leader determines that they should be moved to a different cluster
member. An AP is anchored to its AAC in a three step process:
1. The AP establishes active tunnels with its AAC
2. The cluster leader dynamically assigns a standby AP Anchor Controller (S-AAC) for the AP
from one of the other cluster members
3. Once designated the AP established standby tunnels to the S-AAC
The AAC and S-AAC assignment process works similarly to how HA is configured however rather
than having to be manually configured the process is completely dynamic. Once the AAC is
designated for an AP the subsequent steps occur automatically. A visual representation of AAC
assignment is displayed in the figure below:
The AAC and S-AAC for an AP can be identified in the GUI of the MM by navigating to Dashboard >
Access Points:
While the view above indicates that the S-AAC for both APs is the same device ([Link]) it
should be noted that the S-AAC is assigned by the cluster leader and not all APs terminated on an
AAC will have the same S-AAC. It could just as easily be a different cluster member depending on
the determination made by the cluster leader based on the conditions in the cluster environment
at the time of assignment.
In order to remain anchored, a user must first be mapped to a UAC through a hashing algorithm
at the AP level. The MAC address of the client is examined and the hashing algorithm creates an
index which is then compared to a mapping table. The same mapping table is pushed to all APs by
the cluster leader to ensure UAC mapping consistency across the cluster. In addition, the cluster
leader will dynamically select a standby UAC (S-UAC) on a per-user basis for redundancy purposes.
An example of the hashing algorithm and UAC assignment process is displayed in the figure
below:
The UAC and S-UAC assignments for all associated clients can be identified in the GUI of the MM
by navigating to Dashboard > Clients:
The Active Controller and Standby Controller columns are not included in the standard view of the Clients page in the
GUI. They can be displayed by adding a customization to the page view.
This section describes the process of Dynamic Authorization to RADIUS as described in RFC-5176 and how RADIUS
communicates with Aruba controllers in a cluster. The Change of Authorization process was selected as a
representation of that communication sequence.
ArubaOS reserves VRRP instance IDs in the 220-255 range. When the master of each instance
sends RADIUS requests to the RADIUS server it injects the VIP of its instance into the message as
the NAS-IP by default. This ensures that CoA requests from the RADIUS server will always be
forwarded correctly regardless of which MC is the acting master for the instance. I.e. the RADIUS
server sends CoA requests to the current master of the VRRP instance and not to an individual
station. From the perspective of the server it is sending the request to the current holder of the
VIP address of the instance. The figure below depicts sample architecture that will be used for the
duration of the CoA section:
This sample network consists of a three-node cluster with three instances of VRRP. The AOS-
assigned VRRP ID range falls between 220 and 255 therefore the three instances in this cluster are
assigned the VRRP IDs of 220, 221, and 222. The priorities for the MCs in each instance are
dynamically assigned so that the master of the instance is assigned a priority of 255, the first
backup is assigned a priority of 235, and the second backup is assigned a priority of 215. The table
below outlines the priority assignments for each MC and each instance in the example network:
As demonstrated by the table, MC1 is the master of instance 220 with a priority of 255, MC2 is the
first backup with a priority of 235, and MC3 is the second backup with a priority of 215. Similarly,
MC2 is the master for instance 221 due to having the highest priority of 255, MC3 is the first
backup with a priority of 235, and MC1 is the second backup with a priority of 215. Instance 222
follows the same pattern as instances 220 and 221.
CoA Redundancy
The failure of a cluster node is an event that can adversely impact CoA operations if the network
doesn’t have the appropriate level of fault tolerance. If a user’s anchor controller fails, the RADIUS
server will push the CoA request to their UAC as usual with the assumption that it will enforce the
change and respond with an ACK. However, if a redundancy mechanism such as VRRP hasn’t been
implemented then the request will go unanswered and will not result in a successful change. In
such a scenario the users associated with the failed node will failover to their standby UAC as
usual. However, the UAC will never receive the change request from the RADIUS server since the
server has no awareness of cluster operations. VRRP instances must be implemented for each
node to prevent such an occurrence and maintain CoA operations in the cluster.
In the figure below MC1 is the master of instance 220 with MC2 serving as the first backup and
MC3 serving as the second backup. A client associated to MC1 has been fully authenticated using
802.1X with MC3 acting as the client’s standby UAC. When corresponding with ClearPass, MC1
automatically inserts VIP for instance 220 as the NAS-IP. From the perspective of ClearPass it is
sending CoA requests to the current master of instance 220.
If MC1 fails while the client is in session the AP where the client is associated will failover to MC2.
The client’s session moves over to MC3 since it was the standby UAC. MC3 then assumes the role
of UAC for the client. Since MC2 has a higher priority than MC3 in instance 220 it will assume the
role of Master and take ownership of the VIP.
Any CoA requests sent by ClearPass for the client will be addressed to the VIP for instance 220.
From the perspective of ClearPass, the VIP of instance 220 is the correct address for any CoA
request intended for the client in the example. Since MC1 has failed, MC2 is now the Master of
VRRP instance 200 and owns its virtual IP. When ClearPass sends a CoA request for the client,
MC2 will receive it and then forward it to all nodes in the cluster. Since our cluster only has three
nodes, in this case MC2 forwards the request to MC3.
After the change in the CoA request has been successfully implemented, MC3 will send a CoA-ACK
back to ClearPass.
Network Management
Topology Description
AirWave is a network monitor and management platform that adds controllability and visibility for
wired and wireless devices in any network from a single graphical interface. This kind of visibility
simplifies troubleshooting, for example tracking down slow DNS, and offers deep insight into
optimizing the network for specific applications such as Unified Communications and
Collaboration (UCC). AMP has other features like VisualRF to help us plan spectrum and RAPIDS to
protect the network from malicious attacks.
AMP collects data from network devices (Aruba and other brands) via protocols such as SNMP,
SSH, and ICMP so administrators can view network performance either historically or in real time.
Aruba controllers specifically use AMON to pass deeper information, such as AppRF, UCC, and
spectrum data. HTTP Secure (HTTPS) is used for all communications when implementing Aruba
Instant with AMP. Please refer to the 8.2.4 Best Practice Guide for additional information on data
acquisition methods.
This document assumes that an AirWave server has been installed and is reachable in the Data
Center. AirWave can be deployed either virtually or as a hardware appliance. Please refer to the
8.2 Installation Guide for additional details on the deployment methods for AirWave.
In the figure above the user role AdminEast has monitor access to devices in the Florida and New
York offices while AdminWest can see California and Oregon devices. The Architect role has global
access to all devices in the network.
Device Discovery
In this base design, we configure devices for monitoring using SNMPv2. With polling we can
manually onboard devices to AirWave. Routers, switches, and controllers require per-device SNMP
community and trap-host configuration so AirWave can poll and communicate with them. Standby
controllers can be discovered through the master controllers. APs are also discovered through
controllers. In order to communicate client information, firewall statistics, and RF data the
controllers must be configured to Enable AMON.
ZTP is another discovery option for Aruba devices using Aruba Activate or DHCP, not discussed in this VRD.
Full or partial configuration is another management scenario using Templates for Aruba devices. See Creating and
Using Templates section of the AirWave User Guide.
Mobile First Base Designs Lab for ArubaOS 8 Network Access Control | 53
Chapter 7
TME-MobileFirst-byod_employee
This SSID is for employee clients to connect their mobile phones, laptops, and other personal
devices to the corporate network. Traffic on this SSID is held in VLAN 225 and is handled by the
MCs. CP-Corp is configured as our RADIUS server and to host an Onboarding page.
The bring your own device (BYOD) deployment is a combination of 802.1X authentication and
onboarding portal to enforce the use of Extensible Authentication Protocol-Transport Layer
Security (EAP-TLS), certificate-based authentication.
Mobile First Base Designs Lab for ArubaOS 8 Lab Design and Addressing | 54
ClearPass Onboard relieves the burden of deploying certificates for every user. The TME-
MobileFirst-byod_employee wireless network allows a first-time corporate client to connect using
Extensible Authentication Protocol-Protected (EAP-PEAP) with a limited access user role called
onboard. During the onboarding process clients are redirected to a captive portal where they are
prompted to install a certificate and network profile. Post installation the client is able to
authenticate with EAP-TLS. The client then deauthenticates and then reconnects with the new
network profile using EAP-TLS. Upon recognizing clients connecting to the TME-MobileFirst-
byod_employee SSID using EAP-TLS ClearPass automatically profiles the user with the TME-
Mobilefirst-byod_employee user role. This role then grants the user full network access.
TME-MobileFirst-psk_corp
This SSID is for clients that do not support 802.1X authentication, such as printers, scanners, or
other application-specific machines. Traffic on this SSID is held in VLAN 226 and is handled by the
MCs, configured for Wi-Fi Protected Access 2-Pre-Shared Key (WPA2-PSK) authentication. The
TME-MobileFirst-psk_corp SSID could also potentially be used to accommodate IoT (Internet of
Things) devices.
TME-MobileFirst-Guest
This SSID is for guests visiting the physical campus to connect their mobile phones, laptops, and
so on, to access the Internet. Guest clients do not need access to the corporate network and are
therefore handled in the DMZ by DMZ-MCs and ClearPass-DMZ, the separate ClearPass server.
From the DMZ, guest clients can reach the internet by Network Address Translation (NAT) inside
the firewall. Using this SSID, guests can reach the Internet after a L3 captive portal self-registration
with Media Access Control (MAC) caching.
VLAN 999 holds the traffic on this SSID. VLAN 999 belongs in the DMZ, but is extended through L2
generic routing encapsulation (GRE) tunnels to the LCs, where all of the APs terminate.
2. The guest user's MAC address is added into the Endpoints Repository with an Unknown
status in the ClearPass server.
3. Guest user finishes registration, gets a valid username and password, and the user entry is
added into the guest user repository.
5. For example, say the MAC caching timeout on the ClearPass server is set to 8 hours. If the
end client were to drop off the guest network and connect back within 8 hours, in the next
1 hour, MAC authentication succeeds and the user is granted access. There is no need for
captive portal authentication again.
EAP-PEAP Authentication
EAP-PEAP authentication is usually username and password-based.
As EAP-PEAP and EAP-TLS flows have already been explained in the preceding sections this section will only outline
web-login redirect and device provisioning flows
Configuration
When building an ArubaOS 8 architecture certain steps must be followed in order to ensure
proper functionality. The wired network needs to be built first because the ArubaOS 8 controller-
based WLAN is an overlay and requires it to function. The network Core is configured first
followed by the Aggregation and Access layer devices. Once the wired network has been properly
configured the wireless components can be layered on top of it. The wireless network is
established from Data center level down to APs at the Access layer. The DMZ equipment is
configured last, again from top to bottom.
Upon completion of this chapter, the devices will be pingable and their Web user interfaces
(WebUI) will be functional. If replicating this configuration in a production or even in a test lab
environment it is important to keep the following points concerning ArubaOS switch configuration
in mind:
• The word “trunk” refers to a named group of physical links (or LAG), but it does not refer to
a “VLAN trunk”
• A VLAN added to a “VLAN trunk” is referred to as “tagged”. An “access” port on an ArubaOS
switch is “untagged”
• On an ArubaOS switch ports are added to a VLAN, whereas on an ArubaOS WLAN
Controller VLANs are added to a port
• Ports on a modular ArubaOS switches are labeled using the following convention: Chassis
number/module letter/port number if they are configured as part of VSF. E.g., port 20 on
module B installed on chassis 1 would be represented in the following manner: 1/B/20
SW-TOR-LAB(config)# vlan 88
Internal configuration, not
SW-TOR-LAB(vlan-88)# name 88-Interconnect_TOR-to-IT
necessary outside of the
SW-TOR-LAB(vlan-88)# ip address [Link] [Link]
Aruba test lab
SW-TOR-LAB(vlan-88)# exit
SW-TOR-LAB(config)# vlan 89
SW-TOR-LAB(vlan-89)# name 89-Servers Add VLAN 89 for Servers
SW-TOR-LAB(vlan-89)# tagged Trk2 tagged on Trk2
SW-TOR-LAB(vlan-89)# exit
SW-TOR-LAB(config)# vlan 90
Add VLAN 90 for controller
SW-TOR-LAB(vlan-90)# name 90-MobilityControllerMgmt
management
SW-TOR-LAB(vlan-90)# exit
SW-TOR-LAB(config)# vlan 91
Add VLAN 91 for switch
SW-TOR-LAB(vlan-91)# name 91-SwitchMgmt
management
SW-TOR-LAB(vlan-91)# exit
SW-TOR-LAB(config)# vlan 92
SW-TOR-LAB(vlan-92)# name 92-AccessPoint Add VLAN 92 for APs
SW-TOR-LAB(vlan-92)# exit
SW-TOR-LAB(config)# vlan 93
SW-TOR-LAB(vlan-93)# name 93-DMZ Add VLAN 93 for the DMZ
SW-TOR-LAB(vlan-93)# exit
SW-TOR-LAB(config)# vlan 96
Add VLAN 96 for PSK
SW-TOR-LAB(vlan-96)# name 96-PskClients
clients
SW-TOR-LAB(vlan-96)# exit
SW-TOR-LAB(config)# vlan 98
SW-TOR-LAB(vlan-98)# name 98-Interconnect_TOR-to-Core Add VLAN 98 to connect
the TOR switch to the
SW-TOR-LAB(vlan-98)# untagged Trk2
network Core untagged on
SW-TOR-LAB(vlan-98)# ip address [Link] [Link] Trk2
SW-TOR-LAB(vlan-98)# exit
SW-TOR-LAB(config)# vlan 99
SW-TOR-LAB(vlan-99)# name 99-Interconnect_TOR-to-DMZ-FW Add VLAN 99 to connect
SW-TOR-LAB(vlan-99)# untagged Trk3 the TOR switch to the DMZ
SW-TOR-LAB(vlan-99)# ip address [Link] [Link] untagged on Trk3
SW-TOR-LAB(vlan-99)# exit
SW-TOR-LAB(config)# interface 1
Specify all VLANs tagged on
SW-TOR-LAB(eth-1)# tagged vlan 88-99
Interface 1
SW-TOR-LAB(eth-1)# exit
SW-TOR-LAB(config)# interface 2
SW-TOR-LAB(eth-2)# disable Disable Interface 2
SW-TOR-LAB(eth-2)# exit
SW-TOR-LAB(config)# ip routing
SW-TOR-LAB(config)# exit
SW-TOR-LAB # write memory
VSF allows supported switches connected to each other through Ethernet connections (copper or
fiber) to behave like a single chassis switch.
Configuration guidelines:
• Supported for 5400R only (5406R, 5412R)
• 5400R with v3 modules, operating in v3-only mode
• Currently limited to 2 members (SW version 16.x.x or greater)
• Only same model switches can join a VSF system
• VSF links supported on 10G and 40G Ethernet interfaces only (no 1G)
• Each switch supports only 1 logical VSF link
• Logical VSF links can support up to 8 physical ports
• Physical ports can reside on different modules
• VSF is disabled on the switch by default
1. Ensure both switches are in default configuration state. Issue an ’erase all’ command on
both switches
HP-Switch-5406Rzl2# erase all
The system will be rebooted and all management module files except software images
will be erased.
Continue (y/n)? y
Continue (y/n)? y
3. Add the ports (10 Gig or faster) as the VSF link and connect the switches together. Enable
the VSF domain and reload. About one minute later, the second switch will automatically
setup and reload.
HP-Switch-5406Rzl2# config t
HP-Switch-5406Rzl2(config)# vsf member 1 link 1 A23,A24
All configuration on this port has been removed and port is placed in VSF mode.
Vsf-Port Port-State
-------- ------------
1/A23 Up: Connected to port 2/A23
1/A24 Up: Connected to port 2/A24
Vsf-Port Port-State
-------- ------------
2/A23 Up: Connected to port 1/A23
2/A24 Up: Connected to port 1/A24
SW-Core(config)# vlan 89
SW-Core(vlan-89)# name 89-Servers
Add VLAN 89 for Servers with
SW-Core(vlan-89)# tagged Trk1
ip address tagged on Trk1
SW-Core(vlan-89)# ip address [Link] [Link]
SW-Core(vlan-89)# exit
SW-Core(config)# vlan 90
SW-Core(vlan-90)# name 90-MobilityControllerMgmt Add VLAN 90 for controller
management with ip address
SW-Core(vlan-90)# untagged Trk11-Trk14
untagged on Trk11, Trk12,
SW-Core(vlan-90)# ip address [Link] [Link] Trk13, and Trk14
SW-Core(vlan-90)# exit
SW-Core(config)# vlan 91
SW-Core(vlan-91)# name 91-SwitchMgmt Add VLAN 91 for switch
SW-Core(vlan-91)# untagged Trk21 management with ip address
SW-Core(vlan-91)# ip address [Link] [Link] untagged on Trk21
SW-Core(vlan-91)# exit
SW-Core(config)# vlan 92
SW-Core(vlan-92)# name 92-AccessPoint
SW-Core(vlan-92)# tagged Trk21 Add VLAN 92 for APs with ip
address as a jumbo VLAN
SW-Core(vlan-92)# ip address [Link] [Link]
tagged on Trk21
SW-Core(vlan-92)# jumbo
SW-Core(vlan-92)# exit
SW-Core(config)# vlan 95
Add VLAN 95 for clients on
SW-Core(vlan-95)# name 95-ByodClients the BYOD SSID with ip
SW-Core(vlan-95)# tagged Trk11-Trk14 address tagged on Trk11,
SW-Core(config)# vlan 96
SW-Core(vlan-96)# name 96-PskClients Add VLAN 96 for clients on
the PSK SSID with ip address
SW-Core(vlan-96)# tagged Trk11-Trk14
tagged on Trk11, Trk12,
SW-Core(vlan-96)# ip address [Link] [Link]
Trk13, and Trk14 with DHCP
SW-Core(vlan-95)# ip helper-address [Link] helper address
SW-Core(vlan-96)# exit
SW-Core(config)# vlan 98
SW-Core(vlan-98)# name 98-Interconnect_Core-to-TOR Add VLAN 98 for connectivity
to the TOR switch with ip
SW-Core(vlan-98)# untagged Trk1
address untagged on Trk1
SW-Core(vlan-98)# ip address [Link] [Link] with DHCP helper address
SW-Core(vlan-98)# exit
SW-Core(config)# ip routing
SW-Access-01(config)# vlan 91
SW-Access-01(vlan-91)# name 91-SwitchMgmt Add VLAN 91 for switch
SW-Access-01(vlan-91)# untagged Trk1 management with ip address
SW-Access-01(vlan-91)# ip address [Link] [Link] untagged on Trk1
SW-Access-01(vlan-91)# exit
SW-Access-01(config)# vlan 92
SW-Access-01(vlan-92)# name 92-AccessPoint Add VLAN 92 for AP
management untagged on
SW-Access-01(vlan-92)# untagged 3-48
ports 3-48 and tagged on
SW-Access-01(vlan-92)# tagged Trk1 Trk1
SW-Access-01(vlan-92)# exit
SW-Access-01(config)# exit
Saving Configuration...
Saving Configuration...
Partial configuration for /mm/mynode
------------------------------------
Contents of : /flash/ccm/partial/1/p=sc=[Link]
vrrp 89
ip address [Link]
description 89-MM-Redundancy
priority 100
vlan 89
no shutdown
!
master-redundancy
master-vrrp 89
peer-ip-address [Link] ipsec fb871cf3a675e00c07e947912d86ccdd85808916fe2976ae
!
Configuration Saved.
Limits updated.
The limit for Access Points has been constrained to the platform
limit [499]
(mm01) [mm] (config) #license add ljrSk1hn-TabYSOOl-VYN86W95- Add the LIC-PEF license (Policy
7/d01lwr-HH4yjYq/-3P0zwReC-BzyU04go-7PIJ1ZKO-0vuOlCjS-OtY Enforcement Firewall)
Please make sure to enable the feature bit to have the license
take effect.
(mm01) [mm] (config) #license add ++CJGGtV-w4HfSCma-5s9lgR3s- Add the LIC-RFP license (RF
mNWVVGNO-LOHvgVYY-4fh1MhTY-Gy8OfgYq-tEIn6Zn1-s5b7SOwf-KBI Protect)
Please make sure to enable the feature bit to have the license
take effect.
(mm01) [mm] (config) #license add npMVaDJl-sgXS/PLO-noI9Rnis- Add the SUBX-WebCC license
RQPAJl0m-363eANDJ-UBDYlc6x-rFByHqbD-N4nZwY4Z-dXntPFxu-qcQ (Web Content Classification)
Please make sure to enable the feature bit to have the license
take effect.
(mm01) ^[mm] (License root(/) pool profile) #pefng-licenses- Enable the LIC-PEF license
enable feature bit
Please ensure to add licenses before enabling feature bit.
(mm01) ^[mm] (License root(/) pool profile) #rfp-license-enable Enable the LIC-RFP license
feature bit
Please ensure to add licenses before enabling feature bit.
(mm01) ^[mm] (License root(/) pool profile) #webcc-license-enable Enable the SUBX-WebCC
Please ensure to add licenses before enabling feature bit. license feature bit
The device is reaching out to receive updates for WebCC must have internet access.
Are you sure that you want to stop auto-provisioning and start full setup dialog? (yes/no):
yes
Are you sure that you want to stop auto-provisioning and start full setup dialog? (yes/no):
yes
Are you sure that you want to stop auto-provisioning and start full setup dialog? (yes/no):
yes
Are you sure that you want to stop auto-provisioning and start full setup dialog? (yes/no):
yes
(mm01) [mm] (config) #database synchronize period 60 Adjust the database synchronization period
(mm01) ^[mm] (config) #write memory
(mm01) [mm] (config) #configuration node /md/Aruba Create the /md/aruba and
(mm01) [mm] (config) #configuration node /md/Aruba/campus /md/aruba/campus nodes
(mm01) [mm] (config) #localip [Link] ipsec aruba123 Configure secure communication
between MM-01 and MC-01
(mm01) ^[mm] (config) #localip [Link] ipsec aruba123 Configure secure communication
between MM-01 and MC-02
(mm01) ^[mm] (config) #localip [Link] ipsec aruba123 Configure secure communication
between MM-01 and MC-03
(mm01) ^[mm] (config) #localip [Link] ipsec aruba123 Configure secure communication
between MM-01 and MC-04
(mm01) ^[mm] (config) #write memory
Wait for the tunnels to be built between the MCs and the MM before proceeding with the configuration.
(mm01) [Aruba] (config) #ntp server [Link] Designate servers for all
(mm01) ^[Aruba] (config) #snmp-server community aruba123 devices managed by
MM-01
(mm01) ^[Aruba] (config) #snmp-server host [Link] version 2c
aruba123
(mm01) ^[md Aruba] (config) #mgmt-server primary-server [Link]
profile default-amp
(mm01) ^[campus] (Classic Controller Cluster Profile "campus- Configure cluster VRRP
cluster") #controller [Link] vrrp-ip [Link] vrrp-vlan 90 settings
(mm01) ^[campus] (Classic Controller Cluster Profile "campus-cluster") #controller
[Link] vrrp-ip [Link] vrrp-vlan 90
(mm01) ^[campus] (Classic Controller Cluster Profile "campus-cluster") #controller
[Link] vrrp-ip [Link] vrrp-vlan 90
(mm01) ^[campus] (Classic Controller Cluster Profile "campus-cluster") #controller
[Link] vrrp-ip [Link] vrrp-vlan 90
(mm01) ^[campus] (Classic Controller Cluster Profile "campus-cluster") #redundancy
(mm01) ^[campus] (Classic Controller Cluster Profile "campus-cluster") #active-ap-lb
(mm01) ^[campus] (Classic Controller Cluster Profile "campus-cluster") #exit
(mm01) ^[aruba] (config) #ip access-list session MobileFirst- Prevent BYOD clients from
byod_employee-deny-client-as-dhcp-server acting as a DHCP server
(mm01) ^[aruba] (config-submode) #user any udp 68 deny
(mm01) ^[aruba] (config-submode) #ipv6 user any icmpv6 rtr-adv
deny
(mm01) ^[aruba] (config-submode) #exit
(mm01) ^[aruba] (config) #ip access-list session MobileFirst- Allow all other traffic to and
byod_employee-allowall from BYOD clients
(mm01) ^[aruba] (config-submode) #any any any permit
(mm01) ^[aruba] (config-submode) #ipv6 any any any permit
(mm01) ^[aruba] (config-submode) #exit
(mm01) ^[aruba] (config) #user-role onboard Create the onboard user role
(mm01) ^[aruba] (config-submode) #access-list session and apply the appropriate
mobilefirst-allow-captiveportal position 3 ACLs
(mm01) ^[aruba] (config-submode) #access-list session logon-control position 4
(mm01) ^[aruba] (config-submode) #access-list session captiveportal position 5
(mm01) ^[aruba] (config-submode) #exit
(mm01) ^[aruba] (config) #aaa authentication-server radius CP-Corp Designate RADIUS server
(mm01) ^[aruba] (RADIUS Server "CP-Corp") #host [Link]
(mm01) ^[aruba] (RADIUS Server "CP-Corp") #key aruba123
(mm01) ^[aruba] (RADIUS Server "CP-Corp") #mac-delimiter colon
(mm01) ^[aruba] (RADIUS Server "CP-Corp") #exit
(mm01) ^[aruba] (config) #aaa server-group CP-Corp Define AAA server Group
(mm01) ^[aruba] (Server Group "CP-Corp") #auth-server CP-Corp
(mm01) ^[aruba] (Server Group "CP-Corp") #exit
(mm01) [aruba] (config) #aaa authentication dot1x MobileFirst- Define 802.1X authentication
byod_employee profile
(mm01) [aruba] (802.1X Authentication Profile "MobileFirst-byod_employee") #exit
(mm01) ^[aruba] (config-submode) #aaa profile MobileFirst- Define AAA profile for
byod_employee MobileFirst-byod_employee
(mm01) ^[aruba] (AAA Profile "MobileFirst-byod_employee") #initial-role onboard
(mm01) ^[aruba] (AAA Profile "MobileFirst-byod_employee") #dot1x-default-role onboard
(mm01) ^[aruba] (AAA Profile "MobileFirst-byod_employee") #rfc-3576-server [Link]
(mm01) ^[aruba] (AAA Profile "MobileFirst-byod_employee") #radius-accounting CP-Corp
(mm01) ^[aruba] (AAA Profile "MobileFirst-byod_employee") #authentication-dot1x
MobileFirst-byod_employee
(mm01) ^[aruba] (AAA Profile "MobileFirst-byod_employee") #dot1x-server-group CP-Corp
(mm01) ^[aruba] (AAA Profile "MobileFirst-byod_employee") #exit
(mm01) ^[aruba] (config) # ap-group CampusAP Create the AP group and add
(mm01) ^[aruba] (AP group "CampusAP") #virtual-ap MobileFirst- it to the virtual AP profile
byod_employee
(mm01) ^[aruba] (AP group "CampusAP") #exit
(mm01) ^[aruba] (config) #write memory
In the Mobile First Base Designs Lab for ArubaOS 8 test network CP-Corp is configured as seen in
Administration > Server Manager > Server Configuration > cp-corp.
Next the Server Certificate must be uploaded. Errors will prevent the certificate from being
imported if the Trust List requirement is not met first.
Upon successful import, there will be a confirmation message. You will have to log out of
ClearPass Policy Manager and log in again.
The VIP addresses for all 4 controllers in the campus cluster must also be added for CoA.
Authentication Methods
Two authentication methods will be added: byod_employee EAP-TLS and byod_employee EAP-
PEAP. Navigate to Configuration >Authentication > Methods and click Add.
Enter the details for byod_employee EAP-PEAP and click Save when finished.
Authentication Sources
The Active Directory must be added as an authentication source. Navigate to Configuration >
Authentication > Sources and click Add.
Service Templates
Navigate to Configuration > Start Here. Scroll to the bottom and click Onboard.
3. Select a wireless controller. The choices will be the devices that were configured at the
beginning of this section
4. Click Next
When the Onboard Service Templates is complete, a summary of added content appears:
1. Ensure that the value field in the Onboard Post-Provisioning profile is an exact match for
the name of the user role that was configured on the controllers
2. Click Save when done
Enforcement Policies
Once the profiles have been properly configured it is time to edit the enforcement policies. The
content that the Service Template added must be edited. Two new enforcement profiles now exist
in Configuration > Enforcement > Policies:
A few adjustments will need to be made. Click the name of a policy to begin editing.
3. Access is allowed for every day of the week In the Onboard Pre-Auth Enforcement Policy.
This policy does not need to be edited
4. Click Cancel to leave the screen
From the Configuration > Services menu select byod_employee Onboard Authorization.
1. Add the Corporate AD source created in the Authentication Sources section
2. Click Save
The Summary tab for the Onboard Provisioning service shows an overview.
Configuration Profile
In the pane on the left side of the page navigate to Onboard > Deployment and Provisioning >
Configuration Profiles.
1. Click Create new configuration profile
2. Enter the following details:
Name: byod_employee Configuration Profile
Networks: byod_employee Network Settings
3. Click Save Changes
Provisioning Settings
Navigate to Onboard > Deployment and Provisioning > Provisioning Settings.
1. Click Create new provisioning settings
2. Enter the following details:
Name: byod_employee Provisioning Settings
Organization: Aruba Networks TME
3. Click Next
4. The Supported Devices tab will be left default
5. Click Next
The Onboard Client tab is mainly used to enter configuration for older/legacy OS X devices.
12. Enter the following details on the Onboard Client tab:
Provisioning Address: cp-corp (requires DNS resolution)
13. Click Save Changes
14. The Sponsorship Confirmation page can be left as default
Navigate to Onboard > Deployment and Provisioning > Provisioning Settings and click Test.
(mm01) ^[aruba] (config) #aaa profile MobileFirst-psk Define the AAA profile
(mm01) ^[aruba] (AAA Profile "MobileFirst-psk") #authentication-dot1x MobileFirst-psk
(mm01) ^[aruba] (AAA Profile "MobileFirst-psk") #initial-role MobileFirst-authenticated
(mm01) ^[aruba] (AAA Profile "MobileFirst-psk") #exit
(mm01) [aruba] (config) #wlan ssid-profile MobileFirst-psk Define the SSID profile
(mm01) ^[aruba] (SSID Profile "MobileFirst-psk") #essid TME-MobileFirst-psk
(mm01) ^[aruba] (SSID Profile "MobileFirst-psk") #wpa-passphrase aruba123
(mm01) ^[aruba] (SSID Profile "MobileFirst-psk") #opmode wpa2-psk-aes
(mm01) ^[aruba] (SSID Profile "MobileFirst-psk") #exit
(mm01) ^[aruba] (config) #wlan virtual-ap MobileFirst-psk Define the virtual AP profile
(mm01) ^[aruba] (Virtual AP profile "MobileFirst-psk") #aaa- and add the AAA and SSID
profile MobileFirst-psk profiles to the virtual AP
(mm01) ^[aruba] (Virtual AP profile "MobileFirst-psk") #ssid- profile
profile MobileFirst-psk
(mm01) ^[aruba] (Virtual AP profile "MobileFirst-psk") #vlan 96
(mm01) ^[aruba] (Virtual AP profile "MobileFirst-psk") #exit
AP Configuration
(mm01) [aruba] (config) #whitelist-db cpsec add mac-address
a8:bd:27:c4:ae:7e Whitelist the APs, assign them
names, and add them to the
(mm01) [aruba] (config) #whitelist-db cpsec add mac-address
a8:bd:27:c4:b0:8a CampusAP group
Are you sure that you want to stop auto-provisioning and start full setup dialog? (yes/no):
yes
Network Configuration
(FW-DMZ) [mynode] #configure terminal Enter configuration mode
Enter Configuration commands, one per line. End with CNTL/Z
(FW-DMZ) ^[mynode] (config) #vlan 999 Add VLAN 999 for guest
(FW-DMZ) ^[mynode] (config-submode) #description 999-GuestClients clients with ip address
(FW-DMZ) ^[mynode] (config-submode) #interface vlan 999
(FW-DMZ) ^[mynode] (config-submode) #ip address [Link]
[Link]
(FW-DMZ) ^[mynode] (config-submode) #ip nat inside
(FW-DMZ) ^[mynode] (config-submode) #exit
(FW-DMZ) ^[mynode] (config) #interface gigabitethernet 0/0/2 Add interfaces 0/0/2 and
(FW-DMZ) ^[mynode] (config-submode) #lacp group 1 mode active 0/0/3 into LAG 1
(FW-DMZ) ^[mynode] (config-submode) #lldp transmit
(FW-DMZ) ^[mynode] (config-submode) #lldp receive
(FW-DMZ) ^[mynode] (config-submode) #exit
Are you sure that you want to stop auto-provisioning and start full setup dialog? (yes/no):
yes
(mc-dmz01) ^[mynode] (config) #interface vlan 999 Add ip address for VLAN 999
(mc-dmz01) ^[mynode] (config-submode) #ip address [Link] (required for captive portal
[Link] redirection)
(mc-dmz01) ^[mynode] (config-submode) #exit
(mc-dmz01) ^[mynode] (config) #interface gigabitethernet 0/0/0 Add interface 0/0/0 and 0/0/1
(mc-dmz01) ^[mynode] (config-submode) #lacp group 0 mode active into LAG 0
(mc-dmz01) ^[mynode] (config-submode) #lldp transmit
(mc-dmz01) ^[mynode] (config-subconfig-submode) #exit
Are you sure that you want to stop auto-provisioning and start full setup dialog? (yes/no):
yes
(mc-dmz02) ^[mynode] (config) #vlan 999 Add VLAN 999 for Guest
clients
(mc-dmz02) ^[mynode] (config-submode) #description 999-GuestClients
(mc-dmz02) ^[mynode] (config-submode) #exit
(mc-dmz02) ^[mynode] (config) #interface vlan 999 Add ip address for VLAN 999
(mc-dmz02) ^[mynode] (config-submode) #ip address [Link] (required for captive portal
[Link] redirection)
(mc-dmz02) ^[mynode] (config-submode) #exit
(mc-dmz02) ^[mynode] (config) #interface gigabitethernet 0/0/0 Add interface 0/0/0 and 0/0/1
(mc-dmz02) ^[mynode] (config-submode) #lacp group 0 mode active into LAG 0
(mc-dmz02) ^[mynode] (config-submode) #lldp transmit
(mc-dmz02) ^[mynode] (config-submconfig-submode) #exit
MC-DMZ-01 Licenses
(mc-dmz01) [mynode] #configure terminal Enter configuration mode
Please make sure to enable the feature bit to have the license
take effect.
(mc-dmz01) [mm] (config) #show license-pool-profile-root Verify that license has been
License root(/) pool profile enabled
----------------------------
Parameter Value
--------- -----
enable PEFNG feature Enabled
enable RFP feature Disabled
enable ACR feature Disabled
enable WebCC feature Disabled
(mc-dmz01) [mm] (config) #database synchronize period 60 Set the database synchronization interval
(mc-dmz01) ^[mm] (config) #ntp server [Link] Set the NTP server
(mc-dmz01) ^[mm] (config) #netdestination guest-dmz-external- Create alias for the captive
captive-portal portal
(mc-dmz01) ^[mm] (config-submode) #host [Link]
(mc-dmz01) ^[mm] (config-submode) #exit
(mc-dmz01) ^[mm] (config) #netdestination guest-dmz-internal-net Create alias for the internal
(mc-dmz01) ^[mm] (config-submode) #network [Link] [Link] DMZ network
(mc-dmz01) ^[mm] (config-submode) #network [Link] [Link]
(mc-dmz01) ^[mm] (config-submode) #exit
(mc-dmz01) ^[mm] (config) #ip access-list session guest-dmz- Permit http and https traffic to
allow-external-captive-portal captive portal
(mc-dmz01) ^[mm] (config-submode) #user alias guest-dmz-external-
captive-portal svc-http permit
(mc-dmz01) ^[mm] (config-submode) #user alias guest-dmz-external-captive-portal svc-https
permit
(mc-dmz01) ^[mm] (config-submode) #exit
(mc-dmz01) ^[mm] (config) #ip access-list session guest-dmz-block Block client traffic to the
internal network
(mc-dmz01) ^[mm] (config-submode) #user alias guest-dmz-internal-net any deny
(mc-dmz01) ^[mm] (config-submode) #exit
(mc-dmz01) ^[mm] (config) #ip access-list session guest-dmz- Permit http and https traffic
authenticated
(mc-dmz01) ^[mm] (config-submode) #any any svc-http permit
(mc-dmz01) ^[mm] (config-submode) #any any svc-https permit
(mc-dmz01) ^[mm] (config-submode) #exit
(mc-dmz01) ^[mm] (config) #ip access-list session guest-dmz-drop- Deny all traffic not explicitly
all permitted by other ACLs
(mc-dmz01) ^[mm] (config-submode) #user any any deny log position 1
(mc-dmz01) ^[mm] (config-submode) #exit
(mc-dmz01) ^[mm] (config) #user-role guest-dmz Create guest-dmz user role and apply ACLs
(mc-dmz01) ^[mm] (config-submode) #access-list session guest-dmz-cplogout position 3
(mc-dmz01) ^[mm] (config-submode) #access-list session logon-control position 4
(mc-dmz01) ^[mm] (config-submode) #access-list session guest-dmz-block position 5
(mc-dmz01) ^[mm] (config-submode) #access-list session guest-dmz-authenticated position 6
(mc-dmz01) ^[mm] (config-submode) #access-list session guest-dmz-drop-all position 7
(mc-dmz01) ^[mm] (config-submode) #exit
(mc-dmz01) ^[mm] (config) #aaa authentication-server radius CP-DMZ Designate RADIUS server
(mc-dmz01) ^[mm] (RADIUS Server "CP-DMZ") #host [Link]
(mc-dmz01) ^[mm] (RADIUS Server "CP-DMZ") #key aruba123
(mc-dmz01) ^[mm] (RADIUS Server "CP-DMZ") #mac-delimiter colon
(mc-dmz01) ^[mm] (RADIUS Server "CP-DMZ") #exit
(mc-dmz01) ^[mm] (config) #aaa rfc-3576-server [Link] Designate RFC 3576 server
(mc-dmz01) ^[mm] (RFC 3576 Server "[Link]") #key aruba123
(mc-dmz01) ^[mm] (RFC 3576 Server "[Link]") #exit
(mc-dmz01) ^[mm] (config) #aaa server-group CP-DMZ Define AAA server group
(mc-dmz01) ^[mm] (Server Group "CP-DMZ") #auth-server CP-DMZ
(mc-dmz01) ^[mm] (config) #aaa authentication captive-portal Define captive portal profile
guest-dmz
(mc-dmz01) ^[mm] (Captive Portal Authentication Profile "guest-dmz") #login-page
[Link]
(mc-dmz01) ^[mm] (Captive Portal Authentication Profile "guest-dmz") #welcome-page
/auth/[Link]
(mc-dmz01) ^[mm] (Captive Portal Authentication Profile "guest-dmz") #no guest-logon
(mc-dmz01) ^[mm] (Captive Portal Authentication Profile "guest-dmz") #redirect-pause 3
(mc-dmz01) ^[mm] (Captive Portal Authentication Profile "guest-dmz") #server-group CP-DMZ
(mc-dmz01) ^[mm] (Captive Portal Authentication Profile "guest-dmz") #default-role guest-
dmz-logon
(mc-dmz01) ^[mm] (Captive Portal Authentication Profile "guest-dmz") #exit
(mc-dmz01) ^[mm] (config) #aaa profile guest-dmz Define the AAA profile
(mc-dmz01) ^[mm] (AAA Profile "guest-dmz") #initial-role guest-dmz-logon
(mc-dmz01) ^[mm] (AAA Profile "guest-dmz") #mac-default-role guest-dmz
(mc-dmz01) ^[mm] (AAA Profile "guest-dmz") #radius-accounting CP-DMZ
(mc-dmz01) ^[mm] (AAA Profile "guest-dmz") #rfc-3576-server [Link]
(mc-dmz01) ^[mm] (AAA Profile "guest-dmz") #authentication-mac guest-dmz
(mc-dmz01) ^[mm] (AAA Profile "guest-dmz") #mac-server-group CP-DMZ
(mc-dmz01) ^[mm] (AAA Profile "guest-dmz") #exit
(mc-dmz01) ^[mm] (config) #wlan ssid-profile MobileFirst-guest Define the SSID profile
(mc-dmz01) ^[mm] (SSID Profile "MobileFirst-guest") #essid TME-MobileFirst-guest
(mc-dmz01) ^[mm] (SSID Profile "MobileFirst-guest") #opmode opensystem
(mc-dmz01) ^[mm] (SSID Profile "MobileFirst-guest") #exit
(mc-dmz01) ^[mm] (config) #wlan virtual-ap MobileFirst-guest Define the virtual AP profile
(mc-dmz01) ^[mm] (Virtual AP profile "MobileFirst-guest") #aaa-profile guest-dmz
(mc-dmz01) ^[mm] (Virtual AP profile "MobileFirst-guest") #ssid-profile MobileFirst-guest
(mc-dmz01) ^[mm] (Virtual AP profile "MobileFirst-guest") #vlan 999
(mc-dmz01) ^[mm] (Virtual AP profile "MobileFirst-guest") #forward-mode tunnel
(mc-dmz01) ^[mm] (Virtual AP profile "MobileFirst-guest") #exit
(mc-dmz01) ^[mm] (config) #whitelist-db cpsec add mac-address Whitelist the APs, assign them
a8:bd:27:c4:ae:7e names, and add them to the
(mc-dmz01) ^[mm] (config) #whitelist-db cpsec add mac-address CampusAP group
a8:bd:27:c4:b0:8a
(mc-dmz01) ^[mm] (config) #whitelist-db cpsec add mac-address a8:bd:27:c4:ae:24
(mc-dmz01) ^[mm] (config) #whitelist-db cpsec add mac-address a8:bd:27:c4:ae:0c
(mc-dmz01) ^[mm] (config) #whitelist-db cpsec add mac-address a8:bd:27:c4:af:08
Saving Configuration...
Configuration Saved.
In the Mobile First Base Designs Lab for ArubaOS 8 test network CP-Corp is configured as seen in
Administration > Server Manager > Server Configuration > cp-dmz.
Next the Server Certificate must be uploaded. Errors will prevent the certificate from being
imported if the Trust List requirement is not met first.
Upon successful import, there will be a confirmation message. You will have to log out of
ClearPass Policy Manager and log in again.
General
1. Navigate to the General tab
2. Enter guest-dmz as the Name Prefix
3. Click Next
Access Restrictions
1. Navigate to the Access Restrictions tab
2. Enter the following values:
Enforcement Type: Aruba Role Enforcement
Captive Portal Access: guest-dmz-logon
Maximum number of devices allowed per user: 3
Guest Access: guest-dmz
3. Everything else can be leave blank or in their default states.