WO2012109841A1

WO2012109841A1 - Method and apparatus for keeping network connection alive under a dynamic address configuration

Info

Publication number: WO2012109841A1
Application number: PCT/CN2011/076435
Authority: WO
Inventors: 丁赟
Original assignee: 中兴通讯股份有限公司
Priority date: 2011-02-17
Filing date: 2011-06-27
Publication date: 2012-08-23
Also published as: CN102647290B; CN102647290A

Abstract

The present invention provides a method for keeping network connection alive under a dynamic address configuration based on dynamic host configuration protocol (DHCP), including: after a client accesses, based on DHCP, a network, it monitors the current network connection; when a network connection failure is detected, the current network configuration in use is released, and the re-access to the network based on DHCP is performed. The present invention also provides an apparatus for keeping network connection alive under a dynamic address configuration based on DHCP. By monitoring the network connection, the present invention enables the client to make response and adjustment to the real-time situation of the current network environment. The robustness of the network configuration and the self-repair capability for the connection failures are improved. Therefore, the energy of a network manager to artificially find and solve the failures is saved, and the time that a client waits for the network connection to be repaired artificially is also saved.

Description

Method and device for network connection keep-alive under dynamic address configuration

Technical field

The present invention relates to the field of network connection networking configuration of network devices, and in particular, to a method and apparatus for network connection keep-alive under dynamic address configuration.

Background technique

In today's network device networking configuration, Dynamic Host Configuration Protocol (DHCP) is a very widely deployed deployment method. It is mostly used for corporate network networking within the enterprise, education network networking within the school, etc. . It belongs to the application layer and is based on the Bootstrap Protocol (B00TP). Similar to the Reverse Address Resolution Protocol (RARP), B00TP is used in relatively static environments where each host has a permanent network connection. The administrator creates a B00TP configuration file to define the B00TP parameters for each host. However, when the host frequently moves and the actual number of hosts exceeds the available host IP address, this static mapping method that only provides the host ID to the host parameters is no longer applicable. For this purpose, DHCP is added, which adds automatic allocation. The ability to utilize addresses and additional configuration options. DHCP extends B00TP from two ways: First, DHCP allows the host to use a message to obtain all the configuration information it needs, that is, the protocol for transmitting configuration information. Second, DHCP allows the host to obtain the IP address quickly and dynamically. That is, the mechanism for dynamically allocating IP addresses.

DHCP is built on the client-server model, where the specified DHCP server assigns network addresses and passes configuration parameters to dynamically configured hosts. A host can only function as a server when the system administrator explicitly configures the host as a DHCP server.

DHCP supports three types of address allocation: In automatic allocation mode, DHCP assigns a permanent IP address to the host. In dynamic allocation mode, DHCP assigns a time-limited IP address to the host. When the time arrives or the host explicitly indicates that the address is abandoned. This address can be used by other hosts again. The IP address of the host in the manual allocation mode is specified by the network administrator. DHCP only tells the host the specified IP address. In these three modes, only the dynamic allocation method can re-apply the IP address that has been assigned to the host but if the host is no longer used at this time. Use. In this way, assigning an address to a host that temporarily accesses the network or in a group does not need to be permanent

Dynamic allocation is especially useful when sharing a limited set of IP addresses in a host with an IP address. If a new host wants to permanently access a network and the IP address of the network is very limited, in order to reclaim the IP address when the host is eliminated in the future, dynamic allocation is also a good choice.

In the dynamic allocation mode of DHCP, the DHCP server will inform the client (PC, modem, and any device accessing the network through the DHCP protocol) to the network configuration parameters, and inform the time limit of its use, that is, the lease. If the client uses the DHCP protocol to obtain an IP address and access the network, it must exchange a series of access packets with the DHCP server, and it is in one of six states. As shown in Figure 1, the process is as follows:

1. When the client starts for the first time, it first enters the INITIALIZE state. In order to start obtaining an IP address, it will contact all DHCP servers on the local network. For this reason, the client will broadcast a DHCP in its subnet. DISCOVER message, which contains the value of the network address and lease option. If the client and the DHCP server are not in a subnet, the BOOTP relay agent forwards the message to the DHCP server. After the client sends the DHCP DISCOVER broadcast message, it will go to the SELECT state.

2. After receiving the DHCP DISCOVER broadcast message, all DHCP servers on the local network broadcast a DHCP OFFER message containing the available network address and other configuration information if it can respond to the client. Therefore, The client may receive zero or more DHCP OFFER response messages.

3. The client in the SELECT state receives DHCP from one or more DHCP servers.

After the OFFER responds to the message, it must select one (such as the first arriving response) according to certain conditions, and send a DHCP REQUEST message to the selected DHCP server to negotiate the lease with the request. The text must contain a server ID to indicate the selected server, and then the client enters the REQUEST state. If the client does not receive the DHCP OFFER packet before the timer expires, the DHCP DISCOVER broadcast packet is resent.

4. After receiving the DHCP REQUEST message, the DHCP server can satisfy the request in the request message. In order to confirm that the request has been accepted and start the lease, a DHCP ACK (DHCP acknowledgment) message is sent back as a response. Send a DHCP NAK back to the request (DHCP non-DHCP Confirm) the message as a response. The address that the DHCP server provides to a client in the DHCP OFFER packet cannot be provided to another client. However, if the DHCP server does not receive the DHCP REQUEST packet from the client, the DHCP server can provide the address in the DHCP OFFER packet to the client. Provided to another client.

After receiving the DHCP ACK packet containing the configuration parameters, the client checks the configuration parameters provided in the packet, configures it, and then transfers to the BOUND state. At this point, the client can start using this. Address, if there is an error in the message, send a DHCP DECLINE message and transfer to the INITIALIZE state to restart the process. If the client receives the DHCP NAK packet, it also enters the INITIALIZE state and restarts the entire process.

6. If the client in the BOUND state no longer needs to assign an IP address to it at a certain time, the DHCP protocol allows it to terminate the lease without waiting for the lease to expire. This is especially meaningful when the DHCP server can provide fewer IP addresses than the number of hosts connected to the network. If the client no longer needs an IP address and terminates the lease in time, the DHCP server can reclaim the address and assign it to other clients. , thereby greatly improving the utilization of network resources. In order to terminate the lease early, the client can release the address by sending a DHCP RELEASE message to the DHCP server. After the release message is sent, the client must stop using the address and cannot use this address to send other datagrams. According to Figure 1, the host that issued the DHCP RELEASE message leaves the BOUND state, and must restart the network access procedure from the INITIALIZE state before using the IP address.

When a DHCP client gets an address, it moves to the BOUND state. After entering the BOUND state, the client sets three timers: Control lease updates, rebinds, and expiration. When a DHCP server assigns an address to a client, it can specify a certain value for the timer; if the server does not specify a timer value, the client uses the default value. The default value for the first timer is usually half of the total lease. When the first timer expires, the client must attempt to update the lease. To request an update, the client sends a DHCP REQUEST message to the server that obtained the lease, moves to the update state and waits for a response. The DHCP REQUEST message contains the IP address that the client is using and requests the server to extend the lease for this address. The server can respond to the client's update request in one of two ways: Instruct the client to stop using the address or agree to the client to continue using the address. If the server agrees to send A DHCP ACK packet is sent, and the DHCP ACK packet may also contain a new value of the client timer. If the server does not agree to continue to use, it sends a DHCP NAK message, so that the client immediately stops the use of the address. After receiving the DHCP ACK packet, the client returns to the BOUND state and continues to use the address. If the DHCP NAK message is received, the client immediately stops using the address and returns to the INITIALIZE state.

The second timer starts after the _C li _en t enters the BOUND state. The default value is 87.5% of the total lease. After the client sends the request message, it keeps updating status and waits for the response of the server. If the second timer arrives The client has not received a response from the server, and the client moves from the update state to the re-binding state. At the time of the transfer, the client assumes that the original server is unavailable and starts broadcasting the DHCP REQUEST message to any server on the local network. Any server that can serve the client may respond affirmatively (ie, extend the lease), or negate (ie, refuse to continue using the address). If the client receives a positive response, it returns to the BOUND state and resets both timers. If the client receives a negative response, it moves to the INITIALIZE state and stops using the IP address. It must reacquire a new address before continuing to use the IP address.

If the client does not receive a response from any server after the third timer expires after the client rebinds the status broadcast request message, the client must stop using the IP address, return to the INITIALIZE state, and start applying again. new address.

In the DHCP protocol, since all the network configurations of the client are sent by the DHCP server, the client does not actively change its network configuration as long as the DHCP server does not fail and the network deployment does not change. Some problems have arisen, consider the following two networking scenarios: If the DHCP server is also the gateway server assigned to the client, then once this server fails, the client configured by it will not be able to access the Internet, and will continue until the client enters. When the state is re-bound, the client will consider that the original server is unavailable, and then resend the DHCP DISCOVER message to start a new round of network access. At this time, if there are other DHCP servers available on the network, the client You can get a new network configuration and access the network normally. In this scenario, the worst case occurs when the client just gets the network configuration and the DHCP server fails. At this point, if the client does not restart, it must wait for seven-eighth leases. In the actual network, the lease Generally, it is 24 hours, so the user will not be able to access the Internet for almost a whole day.

If the DHCP server is separated from the gateway server assigned to the client, then once the gateway If the server fails, the situation is even worse: Because the DHCP server is not aware of whether the gateway server is in normal working condition, the DHCP server will not adjust the configuration once the gateway server fails, then all the DHCP server configuration Clients can't access the Internet, and there is no way for the client to fix itself (this is different from scenario 1), even if there are other DHCP servers and gateway servers available on the network.

Summary of the invention

The technical problem to be solved by the present invention is to overcome the problems and defects in the prior art that the client cannot perform automatic monitoring and repair when the gateway device fails, and provide a network connection keep-alive method in the DHCP deployment environment. And equipment.

In order to solve the above problem, the present invention provides a network connection keep-alive method under DHCP dynamic address configuration, including:

After the client accesses the network through the dynamic host configuration protocol, it monitors the current network connection. When the network connection fault is detected, the current network configuration is released, and the network access is re-established through the dynamic host configuration protocol.

Preferably, the foregoing method may further have the following feature: the monitoring the connection of the current network comprises: periodically monitoring the connection of the current network.

Preferably, the method may further include the following features, the method further includes: the client further maintaining a blacklist, and rejecting the blacklist when the client performs network access through a dynamic host configuration protocol. The services provided by the Dynamic Host Configuration Protocol server.

Preferably, the foregoing method may further have the following features, the method further comprising: adding, by the client, the currently associated dynamic host configuration protocol server to the blacklist when the network connection is faulty.

Preferably, the method may further include the following feature, the method further includes: when the dynamic host configuration protocol server on the blacklist meets a preset aging condition, the dynamic host configuration protocol server is removed from the blacklist Deleted.

The invention also provides a network connection keep-alive device in a DHCP dynamic address configuration, comprising: a dynamic host configuration protocol module, which is configured to: perform network connection through a dynamic host configuration protocol Enter

a network connection monitoring module, which is configured to: monitor the connection of the current network after accessing the network;

The network configuration adjustment module is configured to: when the network connection monitoring module detects a network connection failure, release the currently used network configuration, and instruct the dynamic host configuration protocol module to re-access the network through the dynamic host configuration protocol.

Preferably, the foregoing apparatus may further have the following feature, the network connection monitoring module is configured to: periodically monitor a connection of a current network.

Preferably, the device may further have the following features, the device further includes a blacklist management module, and the blacklist management module is configured to: maintain a blacklist;

The dynamic host configuration protocol module is further configured to: refuse to accept the service provided by the dynamic host configuration protocol server on the blacklist when performing network access through the dynamic host configuration protocol.

Preferably, the device may further have the following features: The blacklist management module is further configured to: when the network connection monitoring module detects a network connection failure, add the currently associated dynamic host configuration protocol server to the blacklist.

Preferably, the foregoing apparatus may further have the following feature: the blacklist management module is further configured to: when the dynamic host configuration protocol server on the blacklist meets a preset aging condition, the dynamic host configuration protocol server is removed from the Remove from the blacklist.

Compared with the prior art, the network connection monitoring module and the network configuration adjustment module are mainly added, so that the client can react and adjust the real-time status of the current network environment, thereby improving the network configuration. Robustness and self-healing capabilities for connection failures saves network administrators the time to find faults and solve problems and the time users wait for network connections to be repaired. BRIEF abstract

Figure 1 is a state transition diagram when a DHCP client obtains an IP address using the DHCP protocol to access the network.

2 is a networking diagram of a DHCP client with a session keep-alive mechanism in a real-world application; 3 is a network connection keep-alive device in the DHCP dynamic address configuration of the present invention; FIG. 4 is a workflow diagram of a DHCP client with a session keep-alive mechanism.

Preferred embodiment of the invention

The present invention uses the following technical solutions:

As shown in Figure 2, 3:

On the server side, implement the DHCP protocol module and network deployment;

On the client side include:

The DHCP protocol module is used to implement the function of the client accessing the network in the DHCP environment; the network access is performed through the dynamic host configuration protocol;

The network connection monitoring module is configured to monitor the current network connection after accessing the network; the network configuration adjustment module is configured to perform network configuration adjustment strategies when the network connection monitoring module detects a network connection failure, including: releasing the current use The network configuration instructs the DHCP protocol module to re-access the network through the Dynamic Host Configuration Protocol.

The network connection monitoring module is used to periodically monitor the connection of the current network.

The device further includes a blacklist management module, configured to maintain a blacklist, and add the currently associated dynamic host configuration protocol server to the blacklist when the network connection monitoring module detects a network connection failure.

The dynamic host configuration protocol module is further configured to: when the network access is performed through the dynamic host configuration protocol, refuse to accept the service provided by the dynamic host configuration protocol server on the blacklist.

The blacklist management module is further configured to: when the dynamic host configuration protocol server on the blacklist meets a preset aging condition, delete the dynamic host configuration protocol server from the blacklist.

The server-side DHCP protocol module indicates that a server configured with a DHCP protocol needs to be deployed to provide a DHCP service on the network, and multiple DHCP servers and their associated gateway devices need to be deployed to provide a redundant backup mechanism, because only one set is provided. In the network environment of the service device, if the service device fails, the client has no way to adjust the network configuration itself. Repair with the network connection. The client's DHCP protocol module implements the DHCP client function on the client side to interface with the DHCP server. The network connection monitoring module is responsible for monitoring the connection status with the assigned gateway server after the client obtains the network configuration information through the DHCP protocol and successfully accesses the network. If the connection is found to be faulty, the network configuration adjustment module is notified to the current The network configuration is adjusted to achieve the purpose of repairing the network connection and being able to access the Internet again. The invention provides a network connection keep-alive method under DHCP dynamic address configuration, which includes: after the client accesses the network through the dynamic host configuration protocol, monitors the current network connection; when the network connection fault is detected, releases the currently used network configuration. , Re-network access through the Dynamic Host Configuration Protocol.

Among them, the current network connection is periodically monitored.

The method further includes: the client further maintains a blacklist, and refuses to accept the service provided by the dynamic host configuration protocol server on the blacklist when the client accesses the network through the dynamic host configuration protocol. ;

The method further includes: when the client detects a network connection failure, adding the currently associated dynamic host configuration protocol server to the blacklist.

The method further includes deleting the dynamic host configuration protocol server from the blacklist when the dynamic host configuration protocol server on the blacklist meets a preset aging condition.

The network connection keep-alive method in the DHCP deployment environment of the present invention includes the following steps: Step A. Deploy two or more network service devices on the local network, where the network service device refers to a DHCP server and a matching gateway. Server

Step B. The DHCP client performs a network access attempt through the DHCP protocol.

Step C. If the DHCP client successfully obtains the network configuration parameters and accesses the network, the network connection monitoring module is started to monitor the connection status with the assigned gateway server. If the network access attempt fails, return to step B;

Step D. If the network connection monitoring module finds that the connection with the assigned gateway server is normal, step E is performed, otherwise step F is performed;

Step E. Perform normal maintenance of the client-side DHCP protocol, and restart the network connection monitoring module to perform network connection monitoring after a certain period of time, and return to step D; Step F. If the current gateway server is unavailable, the network configuration adjustment module sends a DHCP RELEASE message to the currently associated DHCP server, releases the currently used network configuration, and adds the DHCP server to the blacklist for later network connection. In the attempt, the service provided by the user will not be accepted again, and the process proceeds to step B again.

The implementation of the technical solution will be further described in detail below with reference to the accompanying drawings.

As shown in Figure 4, the DHCP client with the keep-alive mechanism is implemented as follows: Step 1: First, when the DHCP client starts, it creates a blacklist to record the faulty DHCP server and then starts sending DHCP. DISCOVER message;

The aging condition is set. When the DHCP server in the blacklist meets the aging condition, the DHCP server is deleted from the blacklist.

In this embodiment, the aging condition may be: setting a timer for the DHCP server corresponding to each blacklist entry, and deleting the DHCP server whose timer expires from the blacklist. Other aging conditions can also be set as needed, which is not limited by the present invention.

Step 2: After receiving the DHCP OFFER packet, check whether the DHCP server responding to the packet is in the blacklist and update the blacklist. That is, traverse the blacklist and check whether the timer has timed out. The blacklist entry is deleted. If there is no timeout, it is checked whether it is the DHCP server currently responding to the DHCP OFFER message; then the next blacklist entry is processed until the traversal is completed;

Step 3: According to the result returned by the blacklist search operation, if the DHCP server is in the blacklist, the DHCP OFFER packet received by the DHCP server is ignored, and the DHCP OFFER packet responded by other DHCP server is processed. If no other DHCP OFFER response packet is received, the DHCP DISCOVER packet is resent and waits for the response from the DHCP server on the network. If the DHCP server is not in the blacklist, DHCP REQUEST is sent to the DHCP server. Request the network parameter configuration;

Step 4: After receiving the DHCP ACK message from the server and obtaining the network configuration parameters, the DHCP client starts a process (connection status monitoring process), and periodically sends an ICMP (Internet Control Message Protocol) Echo (ping) to the assigned gateway server. The request message, if the ICMP Echo (ping) reply message of the gateway server is not received for a certain number of times, the gateway server is considered to be faulty, and a DHCP RELEASE message is sent to the DHCP server that has obtained the lease to release the current network configuration. , add this DHCP server to the blacklist, and add the remaining entries in the blacklist. If the aging update is performed, the aging update is not performed at this time. When the DHCP OFFER is received, the aging update is performed. At this time, the connection status monitoring process task is completed and exits. The DHCP client returns to the initialization state and resends the DHCP DISCOVER message.

The method of updating the blacklist list can also automatically delete the corresponding DHCP server from the blacklist when the timer expires. Or learn how to update regularly. The invention is not limited thereto.

Compared with the prior art, the network connection monitoring module and the network configuration adjustment module are mainly added, so that the client can react and adjust the real-time status of the current network environment, thereby improving the network configuration. Robustness and self-healing capabilities for connection failures saves network administrators the time to find faults and solve problems and the time users wait for network connections to be repaired.

Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general-purpose computing device, which can be concentrated on a single computing device or distributed over a network composed of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device, such that they may be stored in the storage device by the computing device, or they may be separately fabricated into individual integrated circuit modules, or they may be Multiple modules or steps are made into a single integrated circuit module. Thus, the invention is not limited to any particular combination of hardware and software.

Industrial applicability

Compared with the prior art, the embodiment of the present invention mainly adds a network connection monitoring module and a network configuration adjustment module, so that the client can react and adjust the real-time status of the current network environment, thereby improving the robustness of the network configuration and The self-healing capability of the connection failure saves the network administrator's ability to find faults and solve the problem and the time the user waits for the network connection to be repaired.

Claims

Claim

A network connection keep-alive method in a dynamic host configuration protocol (DHCP) dynamic address configuration, including:

2. The method of claim 1, wherein the monitoring the connection of the current network comprises: periodically monitoring the connection of the current network.

3. The method according to claim 1, wherein the method further comprises: the client further maintaining a blacklist, and rejecting the black when the client performs network access through a dynamic host configuration protocol. The services provided by the Dynamic Host Configuration Protocol server on the list.

The method of claim 3, wherein the method further comprises: adding, by the client, the currently associated dynamic host configuration protocol server to the blacklist when the network connection is faulty.

The method of claim 3, wherein the method further comprises: when the dynamic host configuration protocol server on the blacklist meets a preset aging condition, the dynamic host configuration protocol server from the Removed from the blacklist.

6. A network connection keep-alive device under dynamic host configuration protocol (DHCP) dynamic address configuration, including:

A dynamic host configuration protocol module, which is configured to: perform network access through a dynamic host configuration protocol;

7. The apparatus according to claim 6, wherein the network connection monitoring module is configured to: periodically monitor a connection of a current network.

The device of claim 6, wherein the device further includes a blacklist management module, the blacklist management module is configured to: maintain a blacklist; and the dynamic host configuration protocol module is further configured to: When the dynamic host configuration protocol performs network access, it refuses to accept the services provided by the dynamic host configuration protocol server on the blacklist.

The device of claim 8, wherein the blacklist management module is further configured to: when the network connection monitoring module detects a network connection failure, add the currently associated dynamic host configuration protocol server to the black List.

The device of claim 8, wherein the blacklist management module is further configured to: when the dynamic host configuration protocol server on the blacklist meets a preset aging condition, the dynamic host configuration protocol server Remove from the blacklist.