Thoughts on DHCP Server load balancing

DHCP Server Load balancer is particularly concerned in some enterprises. Here, we will provide the Cisco DHCP Server Load balancer solution and provide you with the following reference. In large-scale data exchange over networks, especially for local networks, DHCP is under great pressure. Many enterprises always invest a lot of manpower and financial resources in device management, in order to ensure the normal operation of the network. In fact, we can solve this problem through the Server Load balancer function, so let's take a look at the specific content through the Cisco solution.

Cisco DHCP service solution

For an intranet with nearly a thousand information points, manual Address allocation will carry a huge management burden and maintenance costs, static Address allocation is not feasible especially when user identity authentication and dynamic VLAN division are implemented on the network. Therefore, DHCP Dynamic Address Allocation Scheme is generally used for large-scale local area networks, however, the traditional DHCP address allocation scheme has many problems in terms of security, reliability, and load balancing capabilities. Cisco's innovative DHCP address allocation scheme CNR (Cisco Network Register) it can effectively solve the above problems and support other services such as TFTP and DNS ｡

1) distributed design of DHCP Server Cisco D

Two Cisco dhcp cnr servers are deployed on the two core Switches of the network. The two DHCP servers are connected through dual NICs. In addition, the hcp cnr Server can achieve load balancing and failover, 80% of the IP address pool is the responsibility of the two servers, and 20% of the address pool is the responsibility of the two DHCP servers on the other floor ｡

80/20 rules for DHCP allocation:

To avoid duplicate address allocation, a 80/20 rule is usually adopted. A local DHCP Server is deployed, which is responsible for 80% of a specific address range, remote Deployment of another DHCP Server is responsible for 20% of a specific address range. If the IP address range assigned to a CIDR block is 10.1.1.0/24, then 10.1.1.1-10.1.1.200 is the responsibility of the local DHCP Server, 10.1.1.201-10.1.1.253 the premise that the remote DHCP Server is responsible for the 80/20 rule is based on the following assumptions:

When the local DHCP Server fails, because the addresses of the two DHCP servers are allocated Real-Time Database Synchronization operations, the lease period of many hosts that have obtained the IP address does not expire and you do not need to apply for an address, only a few newly connected hosts need to apply for IP addresses, which are assigned by the remote DHCP Server ｡

2) DHCP Server load balancing and Redundancy

Traditional simple DHCP redundancy measures

Generally, two DHCP servers are deployed in the center. The two DHCP servers do not have any redundancy protocol. To prevent different clients from getting duplicate IP addresses, assign different address pools to the two servers ｡

Problems with simple DHCP redundancy:

◆ Insufficient IP Address Space

When one DHCP Server fails, only the address space of the other Server provides services. To prevent IP address conflicts, the address pools of the two servers must be different, therefore, the other address space can only be allocated to half of a network segment ｡

◆ PC Connections cannot always provide online connections, which may be interrupted

When a DHCP Server fails, when the IP address of the client obtained from this Server expires, it cannot get a new IP address for continued use, it will interrupt the connection and restart the DISCOVERY process, the network connection is interrupted for a period of time ｡

Cisco DHCP Failover Protocol

To solve the above problems, Cisco submitted a draft to the IETF and applied for IETF considerations as the standard. Currently, Cisco's failover protocol has become the foundation for the ietf dhcp Working Group to build the standard DHCP Redundcy protocol ｡

The Chief Engineer of Cisco Syetems, author of draw.ietf-dhc-failover-12.txt, is currently the Chairman of the IETFDHCP Working Group. In this working group model, there are protocol interactions between Primary DHCP Server and Secondary DHCP Server, the Secondary Server polls the Primary Server to check whether the Server is working. If the Server is working properly, the Seconday Server does not respond to the DHCP request sent by the Client, the Primary Server synchronously updates its DHCP database to the Secondary Server. the Cisco Network Registrar6.2 software uses the Cisco DHCP Safe Failover Protocol to achieve DHCP Server redundancy ｡

DHCP Server load balancing

RFC 3074 defines an algorithm for implementing DHCP Server load balancing Based on MAC addresses. It can send DHCP requests from different MAC addresses to different DHCP servers, therefore, the load balancing of DHCP Server is realized. Cisco DHCP Server supports RFC 3074, so it can achieve redundancy switching and load balancing ｡