Enterasys Networks X-PeditionTM manual RAI Requirements on the XSR, How RAI Components Work

Utilizing the Command Line Interface

DHCP client over the LAN:

•Operational over an Ethernet interface only on the lowest slot/card/port only.

•Uses the options field for TFTP server, IP address, host name and config file.

•Optionally uses Reverse DNS if options are not populated.

At a branch site, the XSR supports the following features over a PPP IETF serial interface:

•Operational on Serial NIM interfaces only - Lowest slot/card/port only.

•Supports standard physical Serial media-types.

•Supports Sync/Async interface type.

•Supports the following clock rates (in bps) for the Async interface:

–9600, 14400, 33300, 28800, 57600, and 115200 baud.

• Supports PPP and MLPPP negotiation via LCP.

• Supports IP address negotiation via IPCP.

RAI Requirements on the XSR

The branch XSR retrieves the configuration file by means of a TFTP client. The TFTP transfer requires a specific file name and unique local IP address to communicate with a remote server.

The branch XSR must get the local IP address from the central site. How it is done differs between RAI methods. FR RAI uses the Bootp server residing on the central site node, RAI DHCP gets the local IP through DHCP negotiation, while PPP-based RAIs facilitate PPP IP address negotiation.

At the end of the process, the interface on the branch XSR is configured with an IP address and can communicate with the TFTP server.

The file name is the name of the startup file that is transferred from the TFTP server to the branch XSR. It derives from the XSR hostname. As in the case of the local IP address, the branch XSR must get the hostname from the central site. RAI uses the rDNS for this purpose. DNS servers map nodenames/domains and IP addresses - typically, you provide a hostname and the DNS server returns the IP address. But rDNS lookup used by RAI offers an IP address and DNS returns the hostname.domain. RAI ADSL is an exception to this process since is not required to perform rDNS during the RAI process because it uses the serial number of the XSR as the name of the startup- config file. In the case of RAI over Ethernet, where DHCP is configured to provide the configuration file, no rDNS is required because config-fileis the file name.

In general, accessing the DNS or TFTP server requires the client (in this case, the branch XSR) to know the IP address of the server. Since the branch router has no configuration and no knowledge of the server address, it must broadcast a request for DNS or TFTP access. For RAI to work correctly, the terminator of the connection (other end of the FR DLCI, PPP or PPPoE point-to-point connection) is required to channel the broadcast to a specific address. This is done by entering the ip helper-addresscommand at the central site.

How RAI Components Work

Frame Relay (Remote Router)

The FR interface keeps the FR link alive via proper protocol handling of LMI. If the connection fails, or new DLCIs are added to the list, the FR interface will notify RAI of the changes.

XSR User’s Guide 2-7