Network management experience on four common faults of DHCP servers

The DHCP service has become an indispensable component in the network. DHCP can be seen in the Windows server operating system or in the IOS software of the Cisco router. We can also see the importance of the DHCP service to the enterprise network. There will be a lot of content about DHCP features and working principles on the network. This is not the focus of my discussion. I will talk about the DHCP server from the opposite perspective. The following lists four common DHCP Service faults to help you solve problems encountered during the use of the DHCP service.

Fault 1: The address pool on the DHCP server is used up.

In practice, all IP addresses are generally not used as IP addresses that can be allocated by the DHCP server. This is mainly because some application servers in the enterprise must use fixed IP addresses. Such as printer servers and file servers. In addition, some services also require a fixed IP address for the client. For example, if you want to use the financial management software, you must use a fixed IP address for the client. If you use an automatically obtained IP address, you cannot connect to the financial management software server. Therefore, a considerable number of enterprise hosts use fixed IP addresses. Therefore, on the DHCP server, an IP address is specified as the IP address that can be allocated.

When no resources are available in the IP address pool, that is, all the IP addresses that can be allocated have been allocated, the IP address cannot be assigned to the new client, as a result, the newly connected client fails due to the absence of an IP address. The depletion of this IP address may be caused by many reasons.

The lease term configured by the Administrator may be long. In this case, some unused clients (such as obsolete old computers) still occupy IP addresses. This is called taking care of the pitfalls. As a result, the old client does not use an IP address, but the new client cannot obtain the IP address. In this case, a better solution is to shorten the lease term. For example, reduce it to three days. The specific time can be determined according to your own needs. The only objective is to minimize the occupation of unused Client IP addresses and improve the IP address usage.

Of course, there may also be a small number of original IP address pools. For example, in the beginning, only some hosts in the enterprise use DHCP to allocate addresses. Later, for some reason, most hosts were changed to IP addresses obtained from DHCP servers. In this case, the IP address resources on the original DHCP server may not be enough. In this case, you need to adjust the number of IP addresses.

Fault 2: IP addresses in the DHCP server address pool conflict with fixed IP addresses.

It can be said that most enterprises adopt a combination of fixed IP addresses and automatically assigned IP addresses. In this case, there will be a problem: Is there a duplicate between a fixed IP address and an automatically assigned IP address? In practice, this is also a common error. Generally, this error is caused by the following two conditions.

First, IP address conflicts caused by external clients. Enterprise networks often have some foreign customers. If the computer of some employees is broken, the company will be handed over to IT personnel for repair. If some customers carry their laptops, they need to access the Internet through the enterprise network. If these computers use a fixed IP address, the IP address may conflict with the IP address assigned by the enterprise DHCP server. This causes network faults. For this reason, I suggest that you restrict network rules for future clients. For example, you can restrict the MAC address. For MAC addresses added to the enterprise network, you must use the policy of automatically allocating IP addresses instead of using fixed IP addresses. To prevent conflicts with existing enterprise IP addresses.

Second, the IP address pool configuration of the DHCP server is incorrect. For example, a server uses a fixed IP address of 192.168.0.2. However, when setting the DHCP scope, the IP address is not excluded, but is allowed to be allocated. In this case, if the DHCP server configures the IP address, the IP address may conflict. This may cause network disconnection. Therefore, when configuring the DHCP server scope, you need to make reasonable planning for the enterprise's IP address. Generally, it is best to divide a continuous IP address segment for allocation of fixed IP addresses. Instead of selecting several IP addresses as fixed IP addresses. If this operation is performed, an IP address conflict may occur. When planning a fixed IP address, you must keep a certain margin to ensure future needs.

Fault 3: Application Service conflict caused by Multiple DHCP servers

In practice, the DHCP service exists in multiple applications. For example, a vro has the DHCP service function, which is also used in Windows Server systems. Generally, registration is not required when the DHCP service is deployed. That is, after a device that has started the DHCP service is connected to the network, the service will be automatically declared. Then the client that receives the information will use the DHCP server to allocate IP addresses (usually based on the first-come-first-served principle ). Therefore, if an enterprise has Multiple DHCP servers in its network, a conflict may occur.

In general, this should be treated separately. If Multiple DHCP services (such as cross-subnet deployment) need to be deployed out of IP Address Allocation, this is also possible. Note that, if there are multiple DHCP servers in the LAN, the address pools in Different server scopes cannot overlap. Otherwise, an IP address conflict occurs when the two DHCP servers give the two clients the same IP address.

If it is an unauthorized DHCP server, the network administrator must strictly prohibit it. If you are using a Windows domain environment, you can register the DHCP server and use it. In practice, you must pay attention to unintentional behaviors. Generally, the DHCP service is enabled for a vrodhcp by default. If you want to add a vro in the network, it is best to configure the vro in a single-host environment and disable the dhcp function of the vrodhcp. Prevent conflicts with existing DHCP servers.

Fault 4: MAC address and IP Address binding

In addition to completing their jobs, the current DHCP software also provides many additional support, such as binding IP addresses to MAC addresses. In other words, in the DHCP system, specifying an IP address can only be used for a MAC address. That is, the IP address corresponds to the MAC address of the client. In fact, this is equivalent to a fixed IP address. When this feature is used with the DHCP service, it may cause some faults.

First, the IP address pool resources may be insufficient. If enterprises need to upgrade their systems, a large number of computers will be eliminated at one time. In this case, the IP address is bound to the MAC address. The MAC addresses of different hosts are different. In this case, the IP address may be insufficient and cannot be assigned to the new host. At this time, even if the lease is shortened, the problem cannot be solved. For this reason, it is recommended that the IP address and MAC Address binding function should not be used together with the IP address automatically allocated by DHCP.

Second, give the host a meaningful name. In actual work, it is possible that an enterprise has multiple office buildings. In this case, some simple maintenance tasks are generally managed remotely. If a host suffers a sudden power failure, the time is incorrect. At this time, if I know the name (or IP address) of this host, I can directly use tools such as Putty for remote maintenance without running around. However, the IP address is automatically assigned. Therefore, you may not be able to precisely locate the faulty client through IP addresses. If you know the host name, you can still find the IP address of this client through the host name for remote maintenance. For enterprise DHCP applications, it is recommended that the network administrator plan the host name of each client. For example, the code can be performed by department + employee number. Then, you can locate the client by host name instead of IP address. The IP address may change, but the host name will not change.

It should be noted that, even if the configuration file, including the DHCP server, is backed up. In particular, proper backup before policy adjustment is a magic weapon for timely troubleshooting when a fault occurs. The planning of IP addresses should also have written documents. This document can only be used to solve problems in the future troubleshooting.