KR20030052564A - Operation system and method for high availability using agent in gateway - Google Patents

Abstract

PURPOSE: An HA(High Availability) operating system using an agent in a gateway is provided to locate the agent for monitoring a system on the gateway operating as a primary system, and to communicate with a backup system, thereby converting to an IP address of the primary system in the backup system if the primary system is abnormal, and performing a function of the primary system in the backup system. CONSTITUTION: The first hub(102) is connected with the Internet(101). The second hub(104) is connected with a client(103) and a server(103a). A primary system(105) performs a gateway function of reciprocally connecting more than two communication networks and transceiving information between the communication networks. A backup system(106) reciprocally connects more than two communication networks to transceive information, and re-drives a network with an IP address of the primary system(105) if the primary system(105) is abnormal, by transceiving a message with an agent of the primary system(105).

Description

Operation system and method for high availability using agent in gateway}

The present invention relates to a gateway, in particular to provide an agent to monitor the system on the gateway operating as the primary system to provide high availability so that the backup system communicates with the backup system, so that the backup system in case of an abnormal detection of the primary system. It is related to a high availability operating system and an operation method using an agent that converts to an IP address, and the original system suspends a network function so that a session connected to the existing system is maintained and the backup system performs a function previously performed by the original system.

In general, in order to implement high availability for equipment that performs a main function, it is implemented using load balancing composed of L4 switches.

In this case, the switching layer classification in the network is determined according to which of the seven layers defined in the OSI Referenc Model (RM). In other words, when switching function is performed in Ethernet, Frame Relay, ATM, etc., which is a protocol of Layer 2, it is an L2 switch, and when switching is performed in IP, IPX, etc., which is a protocol of Layer 3, it is classified as an L3 switch. do. In addition, if it is possible to switch TCP, UDP, etc., a protocol of Layer 4 (L4: Transport Layer 4), it is a general method to divide the L4 switch. Switch 4's ability to directly connect a client to a server with the desired resources without multicasting is useful for communication service providers with a variety of server and information resources.

For example, L4 switches installed inside and outside of the equipment are normally operated so as not to be loaded only on one specific device by distributing input packets, and when one or more devices occur, the communication path is directed only to another device. It acts as a backup.

In other words, a plurality of equipments that perform important functions are configured between the external L4 switch and the internal L4 switch. Normally, two switches improve performance by properly distributing incoming packets to devices. Health check using ping test between external L4 switch and internal L4 switch is used to provide service continuously by sending packets to other devices when network connection is lost due to equipment or cable problem. In the internal and external L4 switches, packet routing is determined by using a hashing algorithm combining a source IP and a destination IP to determine a device to transmit a packet.

As described above, in the case of using the conventional load balancing equipment, there is an advantage that the service can be continuously provided by distributing the network load in the usual time, thereby improving the performance, and changing the path of the packet in case of an abnormality of the equipment.

However, only the load balance itself is expensive and there is a network fault in the equipment between the two L4 switches. As a result, the network of the equipment is intact and the L4 switch cannot detect it when an important process is running in the equipment. Therefore, there is a need for a method in which a backup system performs a function in case an abnormality occurs in an original system while monitoring a process or network state running in the equipment.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned conventional problems, and has an agent for monitoring the system on a gateway operating as the original system to communicate with the backup system, so that the backup system in case of an abnormality of the original system is detected. The purpose is to provide a high availability system using an agent and a method of operating the agent in a gateway so that the backup system can perform the function that the original system has performed.

1 is a block diagram showing a high availability operating system using the agent in the gateway according to an embodiment of the present invention.

2 is a flowchart of an agent in a high availability management method using an agent in a gateway according to an embodiment of the present invention.

3 is a flow chart of an HA server in a high availability management method using an agent in a gateway according to an embodiment of the present invention.

<Explanation of symbols for the main parts of the drawings>

101..Internet 102,104 ... hub

103 ... Client 103a ... Server

105.One system 106 ... Backup system

115 Agent ... High Availability Servers

In order to achieve the above object, the operating system using the agent of the gateway according to the present invention,

A first hub connected to the Internet,

A second hub connected to the client and the server,

An original system located between the first and second hubs and interconnecting two or more communication networks to function as a gateway to exchange information with each other;

Located between the first and the second hub to interconnect two or more communication networks to exchange information with each other, but to exchange messages with the agent of the original system, the IP address of the original system when an error occurs in the original system And a backup system for restarting the network.

The high availability operating method using the agent of the gateway according to the present invention,

Checking whether there are any abnormalities in the network interface card, the various daemons, and each of the important processes in operation;

Transmitting a down message to the backup system and terminating the network of the original system when the check result is abnormal;

If normal, transmitting a normal message to the backup system and receiving an ACK message from the backup system.

Preferably, after the backup system operates the timer to confirm whether to receive a message from the original system, when receiving a normal message from the original system, sending a response message to the original system and restarting the timer;

And if the received message from the original system is an abnormal message, reading the environment file, changing the network environment of the backup system, and restarting the network of the backup system.

Preferably, if there is no message received during the timer termination from the original system, and after changing the network environment of the backup system to read the environment file, characterized in that it comprises the step of restarting the network to the backup system.

Hereinafter, with reference to the accompanying drawings as follows.

1 is a block diagram illustrating a high availability management system using an agent in the gateway of the present invention, FIG. 2 is a flowchart of the original system of the present invention, and FIG. 3 is a flowchart of a backup system of the present invention.

Referring to FIG. 1, between a first hub 102 connected to the Internet 101, a second hub 104 connected to a client and a server 103, 103a, and the first and second hubs 102 and 104. It is located between the original system 105 and the first system and the second hub (102,104) to function as a gateway to interconnect two or more communication networks to exchange information between the communication network and two or more communication networks Backup to exchange information between the communication network and to exchange information with the agent of the original system 105, and to restart the network to the IP address of the original system 105 when an error occurs in the original system 105. And a system 106.

A high availability management system using an agent and a method of operating the same in the gateway according to an embodiment of the present invention as described above will be described with reference to the accompanying drawings.

The first hub 102 is connected to the Internet 101 and the second hub 104 is connected to the clients / servers 103 and 103a and between this first hub 102 and the second hub 104. An original system 105 functioning as a gateway is located in the network, and two or more communication networks are connected to each other so that information can be exchanged between the communication networks.

This internal agent 115 of the original system 105 periodically notifies the backup system 106 while monitoring important processes or status on the system. The backup system 106 which receives the result from the original system 105 gives an acknowledgment (ACK) to the original system 105, where the original system does not receive the response message or the original system 105 has an error. When this occurs and the backup system 106 is notified, the original system 105 stops the network to block communication.

The backup system 106 then restarts the network with the IP address of the original system. As such, sessions previously connected to the original system 105 are maintained by the backup system 106 and the backup system 106 performs the functions that the original system 105 was performing.

That is, by placing an agent 115 to monitor the system on the original system 105 functioning as a gateway, through the High Availability (HA) server 115 and the first hub 102 of the backup system 106 When an abnormality of the original system 105 is detected, the backup system 106 translates to the IP address of the original system 105, and the original system 105 stops the function of the network to maintain the existing session. This allows the backup system 106 to perform the function that the original system was performing.

2 and 3 are flowcharts illustrating a method for operating a high availability using an agent in a gateway according to an exemplary embodiment of the present invention. FIG. 2 is an operational state flowchart of an original agent, and FIG. 3 is an operational state flowchart of a backup system.

2, the agent of the original system checks the network interface card (S101), checks various daemons (Daemon), and checks important processes (S103) in order to determine the current state of the system (S103). .

In step S101, S102, and S103, it is checked whether there is an error (S104). If the result is normal, the system transmits a normal message (OK) to the backup system (S105), and then returns a response message to the normal message. It is checked whether or not received (S106).

When the response message to the normal message is received from the backup system, the above procedure is repeated.

However, if one of the network interface card, the various daemons, and the critical process check results in abnormal operation in S104, a down message is transmitted to the backup system (S107). Down (stop) the network of the system (S108).

In addition, when the response message is not received from the backup system as a result of step S106, it is determined that there is a cable problem on the network, and the network of the original system is down and terminated (S108).

In the message exchange between the agent of the original system and the home agent server of the backup system, the user datagram protocol (UDP) is used to minimize the load on the network by utilizing the characteristic that the message length is short and close to the physical location. I use it.

The HA (High Availability) server of the backup system starts the timer (S111), checks whether a message has been received from the original system (S112), and if so, whether the message is OK or Down. Check (S113).

If the original system is a normal message, a response message is sent to the original system (S114), the timer is reset (S115), and the procedure is repeated.

However, if there is no message received from the original system, check whether the timer has expired (S116). If the timer is terminated, read the pre-stored environment file (S117), and go to the network environment such as the IP address of the original system stored in the environment file. After changing the backup system (S118), the network is restarted and terminated with the backup system (S119).

In addition, if the message received from the source system is abnormal as a result of the reception of step S114, the backup system reads the environment file (S117), changes the backup system network environment (S118), and restarts the backup system network. It is made (S119). In this way, the backup system performs the function of the original system that was previously performed.

As described above, according to the present invention, a high availability management system using an agent and a method of operating the same in the gateway according to the present invention are simpler by using communication between the agent and the high availability server at a lower cost than expensive load balancing equipment on the gateway. Not only does it provide high availability, but it also uses the results of agent monitoring of the state of the original system, thereby providing more detailed high availability.

Claims (4)

A first hub connected to the Internet, A second hub connected to the client and the server, An original system located between the first and second hubs and interconnecting two or more communication networks to function as a gateway to exchange information with each other; Located between the first and the second hub to interconnect two or more communication networks to exchange information with each other, but to exchange messages with the agent of the original system, the IP address of the original system when an error occurs in the original system High availability management system using the agent of the gateway, characterized in that it comprises a backup system for restarting the network. Checking whether there are any abnormalities in the network interface card, the various daemons, and each of the important processes in operation; Transmitting a down message to the backup system and terminating the network of the original system when the check result is abnormal; And transmitting a normal message to the backup system and receiving an ACK message from the backup system in case of normal operation. The method of claim 2, Checking whether the backup system receives a message from the original system by operating a timer, and when receiving a normal message from the original system, sending a response message to the original system and restarting the timer; If the received message from the original system is an error message, read the environment file, change the network environment of the backup system, and then restart the network of the backup system. Way. The method of claim 2, If there is no message received during the timer termination from the original system, read the preset environment file, change the network environment of the backup system, and then restart the network to the backup system. High availability operation method.