Mesh Deployment Modes

This chapter describes the mesh deployment modes and contains the following sections:

Wireless Mesh Network

In a Cisco wireless outdoor mesh network, multiple mesh access points comprise a network that provides secure, scalable outdoor wireless LAN.

The three RAPs are connected to the wired network at each location and are located on the building roof. All the downstream access points operate as MAPs and communicate using wireless links (not shown).

Both MAPs and RAPs can provide WLAN client access; however, the location of RAPs are often not suitable for providing client access. All the three access points in are located on the building roofs and are functioning as RAPs. These RAPs are connected to the network at each location.

Some of the buildings have onsite controllers to terminate CAPWAP sessions from the mesh access points but it is not a mandatory requirement because CAPWAP sessions can be back hauled to a controller over a wide-area network (WAN).


Note

CAPWAP over CAPWAP is not supported. AP in local mode connected on the RAP or MAP ethernet port is not a supported configuration.


Wireless Backhaul

In a Cisco wireless backhaul network, traffic can be bridged between MAPs and RAPs. This traffic can be from wired devices that are being bridged by the wireless mesh or CAPWAP traffic from the mesh access points. This traffic is always AES encrypted when it crosses a wireless mesh link such as a wireless backhaul.

AES encryption is established as part of the mesh access point neighbor relationship with other mesh access points. The encryption keys used between mesh access points are derived during the EAP authentication process.

Universal Access

You can configure the backhaul on mesh access points to accept client traffic over its 802.11a radio. This feature is identified as Backhaul Client Access in the controller GUI (Monitor > Wireless). When this feature is disabled, backhaul traffic is transmitted only over the 802.11a or 802.11a/n radio and client association is allowed only over the 802.11b/g or 802.11b/g/n radio. For more information about the configuration, see the “Configuring Advanced Features” section on page 159.


Note

In rel 8.2 and higher the backhaul is also supported on 2.4 GHz.


Point-to-Multipoint Wireless Bridging

In the point-to-multipoint bridging scenario, a RAP acting as a root bridge connects multiple MAPs as non-root bridges with their associated wired LANs. By default, this feature is disabled for all MAPs. If Ethernet bridging is used, you must enable it on the controller for the respective MAP and for the RAP.

Figure 1. Point-to-Multipoint Bridging Example. This figure shows a simple deployment with one RAP and two MAPs, but this configuration is fundamentally a wireless mesh with no WLAN clients. Client access can still be provided with Ethernet bridging enabled, although if bridging between buildings, MAP coverage from a high rooftop might not be suitable for client access.

Point-to-Point Wireless Bridging

In a point-to-point bridging scenario, a 1500 Series Mesh AP can be used to extend a remote network by using the backhaul radio to bridge two segments of a switched network. This is fundamentally a wireless mesh network with one MAP and no WLAN clients. Just as in point-to-multipoint networks, client access can still be provided with Ethernet bridging enabled, although if bridging between buildings, MAP coverage from a high rooftop might not be suitable for client access.

If you intend to use an Ethernet bridged application, we recommend that you enable the bridging feature on the RAP and on all MAPs in that segment. You must verify that any attached switches to the Ethernet ports of your MAPs are not using VLAN Trunking Protocol (VTP). VTP can reconfigure the trunked VLANs across your mesh and possibly cause a loss in connection for your RAP to its primary WLC. An incorrect configuration can take down your mesh deployment.

Figure 2. Point-to-Point Bridging Example

For security reasons the Ethernet port on the MAPs is disabled by default. It can be enabled only by configuring Ethernet bridging on the Root and the respective MAPs. To enable Ethernet bridging using the controller GUI, choose Wireless > All APs > Details for the AP page, click the Mesh tab, and then select the Ethernet Bridging check box.


Note

The overall throughput of backhaul radio decreases by half for each hop of a mesh tree. When the Ethernet-bridged clients are used in MAPs and heavy traffic is passed, it may result in a high throughput consumption, which may cause the downlink MAPs to disassociate from the network due to throughput starvation.


Ethernet bridging has to be enabled for the following two scenarios:

When you want to use the mesh nodes as bridges.

When you want to connect Ethernet devices such as a video camera on the MAP using its Ethernet port.

Ensure that you enable Ethernet bridging for every parent mesh AP taking the path from the mesh AP in question to the controller. For example, if you enable Ethernet bridging on MAP2 in Hop 2, then you must also enable Ethernet bridging on MAP1 (parent MAP), and on the RAP connecting to the controller.

To configure range parameters for longer links, choose Wireless > Mesh. Optimum distance (in feet) should exist between the root access point (RAP) and the farthest mesh access point (MAP). Range from the RAP bridge to the MAP bridge has to be mentioned in feet.

The following global parameter applies to all mesh access points when they join the controller and all existing mesh access points in the network:

Range: 150 to 132,000 feet

Configuring Mesh Range (CLI)

Procedure

Introduction to Flex+Mesh in release 8.8

Below is a general FlexMesh architecture. The CAPWAP AP is in Flex connect + Bridge mode with a wired uplink to the core network in a 'Root' AP or RAP mode. The AP is still managed by a centralized controller over CAPWAP. The AP however is capable of moving to a standalone mode where the AP will be able to continue to serve the 802.11 clients depending on the data switching method of the WLAN configured on the AP. The data can be centrally or locally switched. When data is centrally switched, all data is sent to the WLC that does the switching further. In a local switch network, the data is sent to RAP where RAP switches locally on the wired uplink. There is no difference to the central and local switched WLAN configurations and functionality between a Flex-Connect and Flex+Mesh mode AP.



Mesh COS AP that support new 8.8 Feature

1562 supported mesh in 8.4 release. 1542 AP (1542D and 1542I) models supported Mesh in 8.5 release. All these Aps should be able to support Flex Mesh as Flex Mesh designed on 1542 should be applicable to 1562 as well since Flex Mesh is a platform independent feature.

Flex Mesh feature is supported on the IOS based Mesh AP even prior to release 8.8; however in rel 8.8 this feature is officially supported on the COS based Mesh AP and supported by TAC starting with rel 8.8. In addition, the IPv6 is now supported on the COS based Mesh APs.

There are two new SKUs of 1542 that are developed. The AP1540 Series released in 8.5 meets most of the technical requirement, but does not have external antennas. AP1542E2 and AP1542E4 are hardware variants of 1541D/I AP. The 1542E2 is a dual band mode AP with dual-radio dual-band 2.4GHz (802.11b/g/n, 20MHz) & 5 GHz (802.11a/n/acW2, 20/40/80 MHz). The 1542E4 is a single band mode AP with antenna A and B to support 2.4G and C and D to support 5G. The Aps support minimum 2 TX & 2 RX chains, 2 spatial streams. AP expected to support minimum 22 dBm (2.4 GHz) and 24 dBm (5 GHz) conducted transmit output power per TX. Basic new PID additions and power table changes for this new platform will be done on both AP and WLC. New power tables for -D(INDIA) with external antennas.

Flexible Antenna port configuration

The above HW changes have requirements for SW changes as well. The AP needs to support a flexible antenna port configuration. SW changes are done to let the user configure the antennas to support either in a single band mode or dual band mode. Software configurable Single Band Vs Dual Band mode. This is similar to the 1532 AP configuration. The user can configure the antenna band modes using WLC CLI or GUI.

Flex Mesh AP Running Modes

Flex mesh COS AP can be running in connected or standalone mode. Standalone mode in flex connect will undergo some changes to inherit standalone functionality for a mesh network. There is also another mode called 'abandoned' mode discussed below in this section of the guide.

Connected Mode

A COS Flex Mesh AP (Root AP or Child Mesh AP) is considered to be in connected mode when it can access and join the WLC and can exchange periodic keep alive messages with WLC. In this mode, Flex Mesh AP will be able support locally and centrally switched WLAN's. It shall allow regular client and Child mesh APs to join.

Standalone Mode

A COS Flex Mesh AP, is considered to be in standalone mode if it loses connection to the controller but it can access the local gateway. In this mode, the COS Flex+Mesh AP will disable all the centrally switched WLANs, and shall keep the locally switched WLANs up and running. It will also allow the new clients to join on local switched WLANs using local authentication as long as the authentication server is reachable in the local network. Child mesh APs will NOT be allowed to join in this mode.

Abandoned Mode or Persistent SSID Mode

A COS Flex+Mesh AP is in abandoned mode when it can no longer access the gateway IP and has no connectivity to the local network. Possible scenarios are:

  • AP is still not locked on to any uplink wired or wireless.

  • A wireless uplink has been established but has not been authenticated.

  • An uplink is established and authenticated, but IP address has the gateway IP has not been configured.

  • An uplink is established, authenticated and also IP address and gateway IP has been configured, but the gateway is not reachable for over a minute.

Neither Child Mesh APs nor the clients are allowed to join in this mode. Local as well as centrally switched WLANs will be disabled. AP may still be scanning for an uplink in this mode so no beacons will be transmitted during this time.


Note

For flex mesh COS APs, in abandoned mode, reboot timer shall be enabled so the AP will have rebooted after 40 minutes, if it does not transition to either standalone mode or connected mode.


Mode/State transitions in the Flex Mesh COS APs

  • Flex Mesh mode COS AP will always boot up in abandoned mode, in which it would need to scan for the uplink (wired or radio).

  • Once a new uplink is selected either during initial stage or during inter gateway roaming scenario, it is expected that the authentication should pass and the CAPWAP connection needs to be formed within 2 minutes, else the selected parent will be blacklisted. This function should be same as a regular Mesh mode COS AP.

  • If a Flex mesh AP has a valid CAPWAP connection and it loses the CAPWAP connection it will transition to standalone mode, and will stay in standalone mode, as long as the gateway is reachable. A Flex Mesh AP will keep track of the IP mode (IPV6 or IPV4) used for the last successful CAPWAP connection and with track the reachability of the GW for that IP mode.

  • For Flex mesh AP in standalone mode, Mesh control will start a timer (20 second) to periodically refresh the ARP entry for GW IP (IPV4 or IPV6) and to also query the GW reachability status from the Path Control Protocol. PCP will maintain the gateway reachability status from that AP either reported by the Root AP via PCP messages or if it is Root AP by doing an ARP lookup for the gateway IP address. If the GW is unreachable for over a minute, the Flex Mesh AP will blacklist the parent and will transition to abandoned mode and will re-scan for a new uplink.

  • To come out of the abandoned mode, AP must connect to the WLC and transition to the connected mode. Transition from abandoned mode directly to standalone mode is not supported and needs to be considered in future design enhancements.

Design considerations for Flex AP in standalone mode:

  • When the Flex AP is in standalone mode, it will stick to the same parent and will NOT try to discover or roam to a better neighbor, even if it is a preferred parent. The reason is that there is no guarantee that the security will pass with the new parent and the roaming will be successful. If the security fails, the perspective parent may get blacklisted unnecessarily. It is best to consider standalone roaming once standalone security is supported for Mesh APs in future design enhancements.

  • BGN timer will be stopped in standalone mode. So, if the child mesh AP is in standalone mode and it joins a parent with a different BGN and goes back into standalone mode after that, BGN timer will be stopped so that the child Mesh AP does not go into re-scan mode after 15 minutes (BGN timer expiry).

  • In standalone mode, reboot timer will be stopped so that the AP does not reboot after 40 minutes, in the absence of a CAPWAP connection.

  • After moving back to connected mode, from standalone mode, best neighbor selection timer and BGN timer will be restarted, so allow the child mesh AP to roam to the best possible neighbor.

Special standalone mode for COS Flex RAPs

In this mode the SSID will be broadcasted always (Persistent SSID). In addition, after reboot, when this special Persistent mode is enabled, Flex Mesh RAP should be able to start broadcasting the SSID even if the gateway is not reachable.

Existing Flex-connect AP mode design

  • Locally switched WLANs are stored in config.flex file and Flex-connect AP broadcasts the local WLAN SSIDs as long as it is standalone mode.

  • On boot up Flex-connect AP would only start broadcasting the locally switched WLANs if the gateway provisioned.

  • If for a COS Flex connect AP, gateway information is removed at some point, it moves out of the standalone mode and stops broadcasting the locally switched SSIDs and waits for gateway to be provisioned again.

  • Once the gateway is provisioned, Flex AP again transitions into the standalone mode and starts broadcasting the locally switched SSIDs again.

  • Without a valid gateway, flex-connect AP eventually stops broadcasting SSIDs, since the local network is not reachable so no reason to connect the clients.

Parts of the existing Flex-connect AP mode design is used to retain WLAN configuration during reboot and to be able to start broadcasting Local SSIDs etc. However, for Flex RAP we have a special standalone mode requirement for NBN deployment as stated below:

  • Flex RAP should be able to boot up directly into the standalone mode and start broadcasting SSIDs, even if the gateway is not reachable.

  • Flex RAP will continue to be in standalone mode and keep broadcasting SSIDs if the gateway was reachable earlier and becomes unreachable at some point.

  • Even if the Flex RAP cannot support any real clients, it still needs to broadcast SSID so that the operator can check if the AP is UP and running.

Design considerations to support new requirement

  • Flex RAP should join the controller at least once to download the WLAN configuration that gets stored in the config.flex file. This WLAN is a local switched one.

  • Once the configuration is stored in the config.flex file, it will become persistent across the reboots and AP does not need to join WLC again as long as the configuration is not erased.

  • A new configuration that is needed for the RAP to maintain the wired link is supported and will be stored in mesh configuration file i.e. "strict_wired_uplink".

  • If the following conditions are true, FLEX Mesh AP will broadcast the local WLANs stored in flex configuration file even if the gateway is not reachable.

    • AP is a Flex Mesh Root AP

    • AP is configured with strict_wired_uplink as true.

  • A new AP CLI command will be supported to configure a Flex Mesh AP as a strict wired AP.

    # CAPWAP ap mesh strict-wired-uplink <true/false>

  • New configuration parameter "strict_wired_uplink" will be stored in config.mesh file in storage directory so that it is persistent across the reboots. Default value of this parameter will be false.

  • Strict wired uplink configuration is only valid if the AP is configured as Flex-Mesh Root AP. For all other AP modes and for Mesh AP role, strict wired uplink configuration will not be effective, even if configured.

  • When strict wired uplink is true for Flex Mesh Root AP:

    • Wired uplink will be immediately selected on mesh restart.

    • Wired uplink will never be blacklisted

    • CAPWAP up timer will not run

    • Mesh Reboot timer will not run

    • Seek of the wired adjacency will always return true, even if the interface is down

    • Wireless backhaul can never be selected as an uplink

    • Wireless backhaul can still be used as downlink to provide connectivity to the Mesh child nodes

  • To avoid issues due to gateway configuration checks, static IP and gateway must be configured on the Flex RAP (even if it just a dummy IP or gateway).

    • Having Static IP and Gateway configuration will allow the Flex RAP to transition into standalone mode after reboot even when there is no connectivity to the local network (i.e. no DHCP server to provision IP and gateway). Flex RAP will then continue broadcasting the locally switched SSIDs even in absence of any network connectivity.

    • If the IP and gateway are not valid, and once AP has connectivity to DHCP server, DHCP IP overwrites the static IP configuration and DHCP IP and gateway configuration takes over.

  • A Simple WLC CLI to enable/disable the 'Persistent SSID' feature will be provided. The WLC and AP should have communication for this configuration to take effect.

  • The AP 'show mesh config' will also dump the current status of this feature.

Configuring Mesh Enhancements

Procedure


Step 1

As indicated in the explanation above the RAP has to be configured to be in an Persistent Transmit of the SSI mode. This configuration option is available from the CLI mode only.



Step 2

To verify that the mode is enabled execute a "show mesh config" command and the "strict wired uplink" should show as Enabled.



Step 3

As indicated above for the persistent SSID to function and to avoid issues due to gateway configuration checks, static IP and gateway must be configured on the Flex RAP (even if it just a dummy IP or gateway). Having Static IP and Gateway configuration will allow the Flex RAP to transition into standalone mode after reboot even when there is no connectivity to the local network (i.e. no DHCP server to provision IP and gateway). Flex RAP will then continue broadcasting the locally switched SSIDs even in absence of any network connectivity.

If the IP and gateway are not valid, and once AP has connectivity to DHCP server, DHCP IP overwrites the static IP configuration and DHCP IP and gateway configuration takes over.




Steps for testing RAP Persistent Mode in Rel 8.8

In order to test the setup best is to configure one RAP with perrsistant SSID or in abandoned mode and one in a regular RAP mode. Connect a client to both RAP and observe behavior when RAPs lose their connectivity to the controller.

  • Client with Persistent Mode enabled should maintain connectivity to the RAP, since RAP continues to transmit the SSID.

  • Client that connected to the regularly configured RAP will lose connectivity since SSID will stop being transmitted.

Introduction to Additional Mesh Features in release 8.8

This section of the deployment guide introduces the few new Mesh or Outdoor AP features in the release 8.8.

The purpose of this document is to provide configuration guidance of the following features:

  1. “Lawful Intercept (LI)” and Monitoring

  2. Whitelisting of specific URLs

  3. Captive Portal Configuration

  4. Policy Enforcement and Quota Management

“Lawful Intercept” (LI) and Monitoring in Rel 8.8

Certain Cisco customers have plans to deploy Cisco Wifi Mesh solution across very large geographical areas with a Flex+Mesh (with local switching) tree. A RAP (Root Access Point) with a wired backhaul to the centralized WLC shall forms a mesh tree serving wireless clients. Lawful Intercept feature is a process of lawful interception and monitoring of mobile phones, landlines, and wireless internet traffic if administration decides to set up a Centralized Monitoring System (CMS) .

There will be a mesh network in Flex+Mesh mode setup and as part of LI, export the client flow information for each flow will be provided.

RAP will do the NAT/PAT as well as LI record generation and sending to LI server via WLC. For all flows, a record will be created in the NAT/PAT. At that point, RAP will create the Syslog record for that flow. RAP will send those Syslog packets through CAPWAP-DATA to the WLC.


Note

Any peer to peer client traffic within mesh tree which does not go via RAP (only MAP handles locally) will not be considered to be reported to LI server.


WLC will update the syslog packet with its own MAC and IP and will forward the Syslog packets to the Syslog server in the network. These packets will not be encrypted.

This will be the typical workflow:

  1. Admin has to configure Syslog server config.

    Either IPv4 or IPv6 is only supported.

    If IPv6 is configured then WLC should be IPv6 enabled.

    The existing “config ap syslog global “ command will be functional.

  2. LI will be enabled/disabled only Globally.

    Prerequisite for this will be Syslog server config.

  3. AP saves the syslog server configuration (IP address and enable/disable) received from WLC on RAP.

  4. IPv4 packets to be NAT/PAT(in case of internal DHCP) on the packets.

    IPv6 packets and also IPv4 packets (in case of external DHCP) will:

    1. Identify flows based on packet Source/dest IP/port.

    2. Save the flows in a FlowTable entry.

  5. LI Reporter element will:

    1. Receive and save new flow records pushed by NAT element/FlowTable element.

    2. It will run a periodic timer (typically 1 minute).

    3. c. On expiry of this timer, all the flow records in its table are flushed and converted to syslog records containing both v4 and v6 flows together. Syslog format is given in next section.

  6. Only at the beginning of the flow creation, it will be sent. Subsequently, no other flow records will be sent.

  7. AP will form the syslog packet

  8. 8) WLC will recognize whether it is LI packet or not.

    Update the contents,

    IP: Dst IP: LI IP (v4 or v6)

    Source IP: Mgmt IP

    Dst Mac: GW Mac

    Source Mac: Mgmt Mac

    UDP Source Port: 514

    UDP Dest Port: 514

  9. Based on the Inner IP packet, WLC will update the Mgmt IP.

    If it is IPv4 then Mgmt IP will be updated.

    If it is IPv6 then Mgmt IPv6 will be updated

  10. WLC will not store any records.

  11. Stats will be recorded for the incoming messages from AP.

    Stats will also be recorded for the outgoing messages from WLC to syslog server.

    Also, other stats if the packet is dropped.

  12. Whenever show command is executed it will show the log.

Syslog Format for Netflow Collector

The syslog record is then encapsulated inside UDP/IP header from AP to LI server based on the config received from WLC.

A Syslog record will be formatted as below:

“syslog header+’:’+ LI Header +’:’+ LI Record 1+’|’+ LI Record 2 +’|’+….”

Syslog Header

  • Facility: Syslog facility code.

  • Severity: Syslog Severity.

  • Timestamp: Time at which the AP sends out the syslog message. This is sent in human readable date format : mmm dd yyyy hh:mm:ss

  • Hostname: Name of the AP (RAP Name)

  • Tag: The Tag field is a string that signifies what type of message is carried in the payload. (AP_LI_V4_FLOW/ AP_LI_V6_FLOW)

LI Header:

“VVTTTTTTTTMMMMMMMMMMMM”

  • VV: Version, currently it is always “01”

  • TTTTTTTT: Time in seconds when this logging is created, Hex values

  • MMMMMMMMMMMM: AP’s mac address. (RAP mac address)

LI record (for IPv4):

“MMMMMMMMMMMM AAAAA’A’A’A’BBBBCCCCCCCCCC’C’C’C’C’C’C’C’DDDDDDDDTTTTTTTTHHHHHHHH”

  • MMMMMMMMMMMM : Client MAC Address (6 bytes)

  • AAAA–source port in HEX (2 bytes)

  • A’A’A’A’ – NAT source port in HEX (2 bytes) (this will be same as above for no-nat case)

  • BBBB–dest port in HEX (2 bytes)

  • CCCCCCCC – source ip address in HEX (4 bytes)

  • C’C’C’C’C’C’C’C’ – NAT source ip address in HEX (4 bytes) (this will be same as above for no-nat case)

  • DDDDDDDD–dest ip address in HEX (4 bytes)

  • TTTTTTTT-Time in seconds, time when flow was created (4 bytes)

  • HHHHHHHH–RAP IP in HEX (4 bytes or 16 bytes)

LI record (for IPv6) (will not have NAT ip and ports):

“MMMMMMMMMMMM AAAABBBB CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDTTTTTTTTHHHHHHHH”

  • F – 0 for IPv4, 1 for IPv6

  • MMMMMMMMMMMM : Client MAC Address (6 bytes)

  • AAAA – source port in HEX (2 bytes)

  • BBBB – dest port in HEX (2 bytes)

  • CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC– source ipv6 address in HEX (16 bytes)

  • DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD – dest ipv6 address in HEX (16 bytes)

  • TTTTTTTT - Time in seconds, time when flow was created (4 bytes)

  • HHHHHHHH – RAP IP in HEX (4 bytes or 16 bytes)

CLI Configuration and Show Commands

New command will be added for LI enable and Disable.

(Cisco Controller) >config flexconnect lawful-interception ?
disable        Disable Lawful-Interception.
enable         Enable Lawful-Interception.
syslog         Configure Lawful-Interception syslog.
timer          Configure Lawful-Interception timer value. Timer is periodic interval [60sec - 600sec] 

Pre-requisite: Ap Syslog should be configured else it will not allow.

  1. Existing command is modified to reflect LI changes.

    # config ap syslog host global <ipv4/ipv6>

    Pre-requisite: If IPv6 is trying to configure we need to check whether IPv6 is enabled and Management is configured with IPv6 address.

  2. There is a new show command to show the stats.

    (Cisco Controller) >show flexconnect lawful-interception ?
    summary        Display Lawful-Interception summary.
    Example of the LI show command on the controller:
    (Cisco Controller) >show flexconnect lawful-interception sum
    Lawful Interception Status: Disabled
    Lawful Interception Timer: 60       
    Lawful Interception IPv4 Addr: 192.201.1.1
    Lawful Interception IPv6 Addr: Not Configured
    

Note

There will be show commands on AP to display the configured LI server IP and status.


Example of the show LI command on the AP.

AP-2802#show lawful-intercept 
Enable: false
Interval(sec): 60
AP IPv4 Address: 1.5.39.108
AP IPv6 Address: ::
Max records: 15
syslog src ip: 192.201.1.2
syslog src ipv6: ::syslog 
src mac: 00:01:02:03:04:09
extlog server ip: 0.0.0.0
extlog server ipv6: ::
extlog server mac: 00:8E:73:56:24:C7
ap name: AP-2802

GUI Configuration of LI

To configure Lawful Intercept from the Controller GUI interface follow the steps below:

Procedure


Step 1

On the controller Management tab configure Logs>Config the IP address of the Log Server.

Step 2

On the Controller tab choose Lawful Intercept and then enable it with the Log Server IP address configured. Hit Apply.


Whitelisting of specific URLs in Rel 8.8

The feature of Whitelist specific URLs on the controller / AP so that users can access those specific sites without having connectivity to the internet. No authentication is mandated to access the whitelisted URLs.

  • Customer device associates to “XXXX” SSID

  • A client gets the IP address and moves to "webauth" required state for HTTP and HTTPS sites

  • A client is able to access the whitelist websites even without authentication; for example, Providing location specific information and other details to the user

  • Unique whitelist URL (based on flex group) for specific GP's based on the local rural policy

  • When the user attempts to navigate to other websites not configured in the whitelist walled garden profile, the user is redirected back to the login page.

  • Once a user is authenticated, he has access to the internet (non-whitelist websites)

Above feature was addressed with DNS-PreAuth ACL feature implemented in 8.7 release (DNS-ACL). A max of 20 domain names can be configured and snooped IP addresses (max 64) will be sent to WLC to assist client roam across APs in webauth_reqd state. Clients shall use these URLs without any authentication as data traffic to/from the snooped IPs (of preconfigured URLs) are allowed from AP.

As https encrypted packets do not give clear-text URL name to allow/deny access in client’s webauth_reqd state, IP address snooping is required to address this requirement.

Admin has to configure a preAuth ACL with a list of whitelisted URLs and map it to a FlexConnect Group assigned to a specific location or users.

The above feature configuration is documented in the 8.7 and 8.8 Flex Connect Deployment Guide at the link below: https://www.cisco.com/c/en/us/td/docs/wireless/controller/technotes/8-7/Flex_7500_DG.html#pgfId-167660

Captive Portal Configuration in Rel 8.8

This feature is to allow users to have multiple splash pages per SSID (Flex group/VLAN based). When users at specific locations are segregated based on VLANs however, in the same SSID (XXXX) will be broadcasted by the WLAN hence need the feature which can support multiple splash pages on single SSID.

Use-case :

  • Customer device associates to “XXXX” SSID

  • A client gets the IP address and moves to "webauth" required state for HTTP and HTTPS sites

  • Customized Captive Portal through external Web-auth is presented to the user based on the AP group configuration

Scaling is to be taken care of in this scenario. If there are many remote locations connected to one WLC and each location will require its own captive portal. For example, WLC 8540 can support 6000 APs. One remote location can have have ~ 5-6 APs so ~ 1000 locations can be connected to one WLC8540 hence WLC will support 1000 splash pages in order to support one splash page for each remote location.

Presently WLC supports external redirect-URL configuration per SSID. This new feature will allow multiple external re-direct URLs for a single SSID, either FlexConnect group (OR) AP Group should be taking the configuration input of external re-direct URL and apply to clients behind the APs mapped to the group.

CLI Configuration and Show

(WLC)config wlan apgroup custom-web global enable/disable <apgroup_name>

(WLC)config wlan apgroup custom-web ext-webauth-url add <ext-webauth-url> <apgroup_name>

 (WLC)config wlan apgroup custom-web ext-webauth-url delete <apgroup_name>


Configured redirect-URL shall be listed in existing show dump:

(WLC)show wlan apgroups 

GUI Configuration of Captive Portal

To configure a captive portal from the controller GUI follow the steps below:

Procedure


Step 1

From the WLAN tab chose Advanced>AP Groups and then create a Flex Group and than select a Flex Connect group to apply the Captive portal .

Step 2

Enable “custom web override” and enter the “External WebAuth URL”

Note 

In this feature, you can create multiple groups with different Captive Portal per the same WLAN and overwrites External Webauth URL configured at Global WLAN level.


Policy Enforcement and Quota Management in rel 8.8

For Quota management - WLC should accept the Radius user authorization change request to allocate different quota to the same user without disconnecting the user.

This feature is supported in:

  • Local, Bridge (Central Switching)

  • Flexconnect, Flex+Bridge (Local Switching)

Feature Use-case:

  • A client has 2 GB plan to access the internet

  • AP is monitoring the bandwidth usage and reporting the statistics to the controller (Bandwidth monitoring)

  • The controller sends the Interim update to the radius server for IPv4 and/or IPv6 (Dual Stack clients)

  • As soon as the given Quota is exhausted, Radius sends CoA to change the policy to a different default plan - (CoA override)

  • A client gets moved to a new plan without actually being disconnected from the network – (Applying new policy on the fly)

Dynamic Policy from AAA

  • 802.11 clients are allotted QoS policy and data rate limits on authenticating with AAA Server

  • WLC does not support ‘run-time' policy enforcement as the client gets new policies during full authentication

  • RFC-5176 allows dynamic rate limiting using Change-of-Authorization(CoA) request / response

  • End clients get provisioned with maximum allotted quota by Service providers based on prepaid / postpaid data plans

  • External billing servers notify AAA on reaching maximum data limit per client basis

Feature Implementation on the WLC

In order to get the new policy/quota enforcement, following enhancements are implemented in WLC:

  1. WLC periodically sends Accounting-Interim to AAA with client statistics.

  2. On reaching the maximum quota allotted per client, AAA will send a CoA-Request with service-type set to “Authorize Only” along with a state parameter.

  3. WLC will respond with CoA-NAK with service-type set to “Authorize-Only” and with state parameter unmodified.

  4. WLC will also send an Access-Request to AAA with service-type set to “Authorize-Only” along with state parameter as received in CoA-Request.

  5. Access-Request shall have the same format of holding other session attributes / NAS as received in CoA-Request.

  6. AAA will respond with Access-Accept with the new policy on rate/bandwidth enforcement.

  7. WLC will forward these new QoS parameters to AP using existing AP_AAA_QOS_PARAMS_PAYLOAD .

  8. AP will apply the new QoS values to the flex local switched client.

  9. There will not be any Disassociation / De-Authentication message sent from WLC or AP to the end client.

Configuration from GUI

Procedure


Step 1

Configure Authentication Server from Security>Radius>Authentication and select “Support for CoA” as shown in the example below.

Step 2

On the WLAN choose the AAA override option as shown below.