Browse Prior Art Database

A METHOD TO INTELLIGENTLY HANDLE DHCP TRAFFIC ON A NETWORK WHEN DHCP SERVERS ARE NOT REACHABLE

IP.com Disclosure Number: IPCOM000244168D
Publication Date: 2015-Nov-18
Document File: 6 page(s) / 120K

Publishing Venue

The IP.com Prior Art Database

Related People

Anantha Narayanan: AUTHOR [+2]

Abstract

Mechanisms and systems are provided to intelligently prune or discard aDynamic Host Configuration Protocol DHCP message that is sent from a DHCP clientto a DHCP server when the DHCP server cannot be reached or when the DHCP serverdoes not properly respond to the DHCP request A relay agent is provided whichperiodically determines the operational status of a plurality of DHCP servers byperiodically polling the DHCP servers DHCP requests sent to a non active DHCPserver are discarded and only DHCP requests sent to DHCP servers that are classified as active are forwarded This allows reducing overall network traffic since DHCPmessages are prevented form being sent to inactive DHCP servers

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 30% of the total text.

Page 01 of 6

A METHOD TO INTELLIGENTLY HANDLE DHCP TRAFFIC ON A NETWORK WHEN DHCP SERVERS ARE NOT REACHABLE

  AUTHORS: Anantha Narayanan Sheltan Thayananth

CISCO SYSTEMS, INC.

ABSTRACT

    Mechanisms and systems are provided to intelligently "prune" or discard a Dynamic Host Configuration Protocol (DHCP) message that is sent from a DHCP client to a DHCP server when the DHCP server cannot be reached, or when the DHCP server does not properly respond to the DHCP request. A relay agent is provided which periodically determines the operational status of a plurality of DHCP servers by periodically polling the DHCP servers. DHCP requests sent to a "non-active" DHCP server are discarded and only DHCP requests sent to DHCP servers that are classified as "active" are forwarded. This allows reducing overall network traffic since DHCP messages are prevented form being sent to "inactive" DHCP servers.

DETAILED DESCRIPTION

    Typical Enterprise environments include a HQ and multiple Branch locations; where DHCP servers are provided at the HQ and Customer Premises Equipment (CPE) at the Branch locations act as a DHCP relay. All the consumer devices like PCs and phones at the Branch locations will relay DHCP messages through the CPE (DHCP relay agent) to the DHCP server. CPE devices can be Cisco routers running IOS/IOSXE.

    For example, a network of a large enterprise environment may include one HQ and 100s of branches. One Branch location can have up to 500 DHCP endpoints assuming that 250 people use 2 client devices (e.g., a phone and a laptop).

    A Relay agent is an Internet host or router that passes DHCP messages between DHCP clients and DHCP servers. The Relay agent is configured to communicate with a plurality of DHCP servers. DHCP Relay agents forward DHCP packets between DHCP

Copyright 2015 Cisco Systems, Inc.
1


Page 02 of 6

clients and DHCP servers. These DHCP Relay agents point to DHCP servers at the head quarter (HQ). DHCP Relay agents are applicable to Service Provider (SP) managed environments and SP Wi-Fi™ (hotspot) deployments.

    In a conventional system, in the event of a DHCP server failure (failure point), DHCP messages such as "DHCP Discover" or "DHCP Renew" are transmitted through the network from all devices that require DHCP service. Based on a timeout, it can be determined whether the DHCP server is down or unreachable.

    After a timeout occurs, the conventional DHCP protocol suggests resending the DHCP messages in a Unicast or Broadcast mode until an Internet Protocol (IP) address lease time is reached. This may generate significant network traffic.

    The following failure points can be defined that may result in a DHCP server not being available: (1) a standard Service Level Agreement (SLA) for a Wide Area Network (WAN) may exists that has a downtime requirement of 4-9 hours per year (Multiprotocol Label Switching (MPLS) with 99.95% SLA and Internet with 99.90%), (2) there may be an upgrade or planned maintenance of the network and the servers, and...