Cisco Cisco 2000 Series Wireless LAN Controller Manuel Technique
GUI, or with greater ease, through WCS templates.
Basic REAP Priming Functions
After 1030 REAPs are connected to a WLC within the mobility group where REAPs connect to when placed
at remote sites, this information can be provided:
at remote sites, this information can be provided:
Required REAP Settings
A list of IP addresses for the WLC in the mobility group (provided automatically upon controller/AP
connection)
connection)
•
REAP AP mode (APs must be configured to operate in REAP mode in order to perform REAP
functionality)
functionality)
•
Optional REAP Settings
Statically assigned IP addresses (an optional setting input on a per−AP basis)
•
Primary, secondary, and tertiary WLC names (an optional setting input on a per−AP basis or via WCS
templates)
templates)
•
AP name (an optional informational setting input on a per−AP basis)
•
AP location information (an optional informational setting input on a per−AP basis or via WCS
templates)
templates)
•
REAP−to−Controller Link Requirements
When you plan to deploy REAPs, a few basic requirements need to be remembered. These requirements
concern the speed and latency of the WAN links REAP LWAPP control traffic will traverse. The 1030 LAP is
intended to be used across WAN links, such as IP Security tunnel, Frame Relay, DSL (non PPPoE) and leased
lines.
concern the speed and latency of the WAN links REAP LWAPP control traffic will traverse. The 1030 LAP is
intended to be used across WAN links, such as IP Security tunnel, Frame Relay, DSL (non PPPoE) and leased
lines.
Note: The 1030 REAP LWAPP implementation assumes a 1500 byte MTU path between the AP and the
WLC. Any fragmentation that takes place in transit due to a sub−1500 byte MTU leads to unpredictable
results. Therefore, the 1030 LAP is not suited for environments, such as PPPoE, where routers proactively
fragment packets to sub−1500 bytes.
WLC. Any fragmentation that takes place in transit due to a sub−1500 byte MTU leads to unpredictable
results. Therefore, the 1030 LAP is not suited for environments, such as PPPoE, where routers proactively
fragment packets to sub−1500 bytes.
WAN link latency is particularly important because every 1030 LAP sends, by default, heartbeat messages
back to controllers every 30 seconds. After heartbeat messages are lost, the LAPs send 5 successive
heartbeats, once every second. If none are successful, the LAP determines that controller connectivity is
severed and the 1030s revert to standalone REAP mode. While the 1030 LAP can tolerate large latencies
between itself and the WLC, it is necessary to ensure that latency does not exceed 100ms between the LAP
and the controller. This is due to client−side timers that limit the amount of time clients wait before the timers
determine an authentication has failed.
back to controllers every 30 seconds. After heartbeat messages are lost, the LAPs send 5 successive
heartbeats, once every second. If none are successful, the LAP determines that controller connectivity is
severed and the 1030s revert to standalone REAP mode. While the 1030 LAP can tolerate large latencies
between itself and the WLC, it is necessary to ensure that latency does not exceed 100ms between the LAP
and the controller. This is due to client−side timers that limit the amount of time clients wait before the timers
determine an authentication has failed.
REAP Limitations
Although the 1030 AP is designed to be managed centrally and to provide WLAN service during WAN link
outages, there are some differences between what services the REAP offers with WLC connectivity and what
it can provide when connectivity is severed.
outages, there are some differences between what services the REAP offers with WLC connectivity and what
it can provide when connectivity is severed.
WLANs
While the 1030 REAP can support up to 16 WLANs (wireless profiles that contain a Service Set Identifier
[SSID] each, along with all security, QoS, and other policies), each with its own Multiple Basic Service Set
[SSID] each, along with all security, QoS, and other policies), each with its own Multiple Basic Service Set