JP2012015762A - Agent implementation scheme in remote monitoring system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a scheme which avoids lack of processing performance of an agent caused by an increase of monitoring target devices, and communication shielding by a router or firewall existing between the agent and the monitoring target devices in a remote monitoring system which installs an agent in a system such as a customer system.SOLUTION: Installing a plurality of agents 52 in a customer system 2 allows a communication amount between monitoring target devices such as a server 53 or a PC 54 and the agents 52 to be kept minimum, and also prevents each of the agents 52 from being loaded more than necessary by restraining the number of monitoring target devices each of the agents 52 takes charge of. In addition, when the agent 52 accesses a monitoring target device, a possibility to be obstructed by devices such as routers 51, 8 and a firewall 7 is lowered.

Description

  The present invention relates to a remote monitoring system that monitors a failure occurring in an electronic device or the like at a remote location.

  Conventionally, in a service for remotely monitoring system operation status and failure information, a method of installing an agent for collecting information on the system side is generally used. As for the installation method of this agent, two methods are generally employed: a method of placing an appliance-type dedicated box and a method of installing a software agent on a server in the system. However, the appliance-type dedicated box requires a high introduction cost because a box unnecessary for the system side is installed in the system and the cost is high. In the method of installing a software agent, the agent is installed as software on a server in the system, so there is no need to introduce hardware, but resources may compete with other applications already installed on the server. Assumed and easy installation can cause harmful effects on the system. Furthermore, when monitoring is performed in a large-scale system, the number of devices to be monitored increases, so the processing performance of the agent may become a bottleneck. There are various problems such as inaccessibility.

  For example, in Patent Document 1, an agent that monitors a plurality of sites and servers via a global IP area such as the Internet is installed or an agent program is introduced to the site or server. When an abnormality is detected, there is a failure monitoring method in which an abnormality is reported from the agent or agent program to the monitoring system, and the abnormality is reported from the monitoring system to the terminal registered in association with the site or server. It is disclosed.

  Also, in Patent Document 2, anomaly information is collected from sub-managers and agents triggered by polling from the manager side, and if there are multiple failures that occur in the agent, the anomaly that is collected by the manager according to its priority A failure monitoring system that enables efficient monitoring of the entire system by changing information is disclosed.

  7 to 9 show implementation examples of a remote monitoring system conventionally used. FIG. 7 shows an example in which an appliance type agent that is a dedicated monitoring terminal is installed. Since the monitoring operation is performed by the appliance type agent 520 that is a dedicated terminal, the influence on other electronic devices and the like can be reduced. FIG. 8 shows an implementation example using a software agent in which an agent program 520 is installed in a server of an existing system. Compared to an implementation example using an appliance-type agent, new hardware is not required and the implementation can be performed relatively inexpensively. FIG. 9 shows an implementation example when a remote monitoring system is introduced into a large-scale customer system. When the agent 1000 is introduced into an existing customer system, the agent 1000 is installed and used at a location that can be easily accessed from each department so that the monitoring is not easily affected.

JP 2002-41376 A JP 2001-337873 A

  However, in the method disclosed in Patent Document 1, abnormal information is directly transmitted to the monitoring system via the Internet from agents installed on a plurality of monitoring target sites and servers. For example, a plurality of servers are monitored. In this case, since the agents installed in each server individually transmit abnormality information to the monitoring system, the network traffic proportional to the number of servers to be monitored is excessively applied to the router as a connection port to the Internet.

  In the method of Patent Document 2, when the sub-manager and the terminal agent belong to different networks, polling cannot reach the agent due to the influence of a router, a firewall, or the like that will be installed between them, or Therefore, it is necessary to review the settings of routers and firewalls, and the cost and time cost for introduction become very large.

  Further, in the remote monitoring system implementation example of FIG. 7, boxes that are unnecessary for the customer are arranged in the system, and the hardware and the installation cost are high, so the side requesting the monitoring work It was very unacceptable from customers. In the implementation example of the remote monitoring system in FIG. 8, since the agent is installed as software on an existing server already used by the customer, no dedicated hardware is required as compared with the implementation example in FIG. Although it can be implemented, it is assumed that resources will compete with other applications used in the existing server, so it cannot be installed easily. Further, in the implementation example of the remote monitoring system of FIG. 9, it is necessary to monitor the monitoring target installed in many departments by one agent, and the processing performance of the agent may become a bottleneck. It is necessary to reset the access authority etc. of the routers installed in each department, and it may be difficult to apply the remote monitoring system depending on the settings of routers, firewalls, etc. on the route.

  The present invention has been made in view of the above-described problems, and an object thereof is to provide an agent mounting method in a remote monitoring system capable of suppressing the installation cost and time cost.

  The agent mounting method of the remote monitoring system according to claim 1 is a remote monitoring system in which agents are connected in a hierarchy of one to a plurality of layers and the status of the monitoring target detected by the end agent is transmitted to the data center. The status of the monitoring target detected by the end agent is sent to the higher order agent of the end agent, and then the monitoring target status is unilaterally transmitted from the lower order agent to the higher order agent. The top-level agent transmits the status of the monitoring target to the data center.

  According to the configuration of claim 1, in the communication from the terminal agent to the data center, the status of the monitoring target is unilaterally transmitted from the lower agent to the upper agent side, so only the upstream communication is performed. The amount of communication can be suppressed.

  The remote monitoring system agent mounting method according to claim 2 is the remote monitoring system agent mounting method according to claim 1, wherein the agent receives the initial status transmission from the lower agent, and It is registered as a new monitoring target.

  According to the configuration of claim 2, there is no need to reset the higher-level agent when newly installing an agent, such as when a new monitoring target is added.

  The remote monitoring system agent mounting method according to claim 3 is the remote monitoring system agent mounting method according to claim 1 or 2, wherein the agent is any one of NAS (Network Attached Storage), a printer, a copier, or a multifunction device. It is characterized by being mounted on the crab.

  According to the configuration of claim 3, NAS, printers, copiers, multifunction devices, etc., whose performance has been remarkable in recent years, without installing an agent program on a server dedicated to an agent or a server in which an existing system or the like is installed. An agent is installed in the OS and memory installed in the system.

  According to the agent mounting method of the remote monitoring system according to claim 1, the agent mounting method of the remote monitoring system according to claim 1 is configured such that agents are connected in a hierarchy of one to a plurality of layers and detected by a terminal agent. The remote monitoring system sends the monitored status to the data center, and the monitored status detected by the end agent is sent to the higher agent of the end agent, and then the lower agent to the higher agent. Since the status of the monitoring target is sequentially transmitted to the data center and the highest level agent transmits the status of the monitoring target to the data center, it is possible to suppress the occupation of the network bandwidth due to the mutual communication of the monitoring system agents. To other systems Communication failures and delays can be prevented, and agents can be placed close to the monitored devices, avoiding communication blocking by routers and firewalls and opening inbound ports, resulting in security benefits. Can be planned.

  The remote monitoring system agent mounting method according to claim 2 is the remote monitoring system agent mounting method according to claim 1, wherein the agent receives the initial status transmission from the lower agent, and Since it is registered as a new monitoring target, when adding a new agent, it is conventionally necessary to reconfigure all agents related to the new agent in agents installed in a hierarchy, but this is necessary. Instead, by setting only the specific information of the upper agent directly above as a new agent, the new agent can be added to the already installed monitoring system, and installation man-hours can be significantly reduced.

  The remote monitoring system agent mounting method according to claim 3 is the remote monitoring system agent mounting method according to claim 1 or 2, wherein the agent is mounted on any of a NAS, a printer, a copier, and a multifunction device. Therefore, companies often need to install agent programs on expensive dedicated terminals or installed servers by installing them in NAS, printers, copiers, or multifunction devices installed in departments or rooms. However, it can be provided as an inexpensive agent or an agent that does not compete with other systems, and it becomes easier to detect failures in these electronic devices and resource tightness. In addition, since NAS, printers, copiers, or multifunction peripherals have a lot of idle time, monitoring data processing (collection, analysis, transmission, etc.) can be performed during that time, and resources can be used effectively.

It is a block diagram which shows the structure of the remote monitoring system in the Example of this invention. It is a flowchart which shows the registration method of the lower agent by the upper agent of a remote monitoring system same as the above. It is a flowchart which shows the registration method of the monitoring object terminal of a remote monitoring system same as the above. It is a flowchart which shows the monitoring operation | movement of the monitoring object by a remote monitoring system same as the above. It is a flowchart which shows the detection operation of the failure which arose in the agent by a remote monitoring system same as the above. It is a block diagram which shows an example of installation of a remote monitoring system same as the above. It is a block diagram which shows the 1st example of the monitoring system used conventionally. It is a block diagram which shows the 2nd example of the monitoring system used conventionally. It is a block diagram which shows the 3rd example of the monitoring system used conventionally.

  DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments as the best mode for carrying out the present invention will be described below with reference to FIGS. Of course, it goes without saying that the present invention can be easily applied to other than those described in the embodiments without departing from the spirit of the invention.

  Based on FIG. 1, the structure of the remote monitoring system in a present Example is demonstrated. Reference numeral 1 denotes a remote monitoring system that monitors a plurality of monitoring targets installed in a customer system described later. Reference numeral 2 denotes a customer system having a plurality of monitoring targets by the remote monitoring system 1. Reference numeral 3 denotes a data center that manages the status of the monitoring target and performs monitoring work such as troubleshooting. The customer system 2 and the data center 3 are communicably connected via the Internet 4 which is a wide area network. A plurality of departments 5 are arranged in the customer system 2, and an electronic device such as a router 51, a server 53, and a PC 54 is installed for each department 5. An agent 52 monitors the status of these electronic devices. The customer system 2 further includes an upper agent 6, a firewall 7, a router 8, a server 9, and a top agent 10. The firewall 7 is responsible for preventing unauthorized access from an external network such as the Internet 4, and the router 8 mediates communication between different networks of the Internet 4 and the customer system 2. The upper agent 6 sends the monitoring status transmitted from the agent 52 installed in the department 5 to the highest agent 10 and monitors the server 9 and the like. The highest agent 10 transmits the monitoring status transmitted from the upper agent 6 to the data center 3 via the firewall 7, the router 8 and the Internet 4.

  As shown in FIG. 1, in the customer system 2, a plurality of terminal agents 52 for monitoring the monitoring target are installed, and are configured with a higher-level agent 6 and a highest-level agent 10 in a three-layer structure. Since a plurality of agents 52 can be installed, they can be arranged in the vicinity of the routers 51, servers 53, and PCs 54 to be monitored, and by distributing them, it is possible to prevent the load from being concentrated on a specific agent 52. it can. Further, since the routers 7 and 51 and the firewall 8 do not exist on the route between the monitoring target and the agent 52, access to the monitoring target is not hindered.

  Next, the operation when registering the lower agent in the upper agent will be described with reference to FIG. An ID indicating the upper agent to which the lower agent belongs (or information such as an IP address that can uniquely identify the agent) is set in the lower agent, and its own ID is transmitted to the upper agent. (S201). The upper agent that has received the ID determines whether the ID is included in the reception history (S202). If it is included, the process ends. If not, the upper agent registers the lower agent as a new agent under its own agent (S203). By such a registration operation, the higher agent side can add a new agent without any special setting. Further, the added agent is registered only with the upper agent immediately above, and the information of the newly added agent is not registered with the higher agent or the data center 3.

  The operation of the agent configuring the remote monitoring system 1 when a device to be monitored by the remote monitoring system 1 is newly added will be described with reference to FIG. The terminal to be monitored is registered in advance using, for example, an IP address or the like, in the terminal agent responsible for monitoring the newly added monitoring target terminal (for example, the server 53 or the PC 54), and the agent accesses the monitoring target terminal. This is performed (S301). The agent determines whether the terminal has accessed in the past from information unique to the monitoring target terminal (for example, a MAC address or a serial number held in the device) (S302). If it is determined that the terminal has been accessed in the past, the process is terminated assuming that the monitoring target terminal has been registered in the data center 3, and if not, it is determined that the terminal has not been registered. (S303), if there is an upper agent, the information is sent to the upper agent in sequence (S304), and if not, information specific to the monitored terminal is transmitted to the data center 3 (S305). Thereby, when a new monitoring target terminal is registered, the monitoring target terminal can be registered in the data center 3 and added as a monitoring target.

  Next, the monitoring operation of the monitoring target by the remote monitoring system 1 will be described with reference to FIG. The terminal agent accesses the monitoring target terminal (for example, the server 53 or the PC 54) and acquires the monitoring status. (S401). It is determined whether or not the content of the monitoring status needs to be transmitted to the data center 3 (S402). If transmission to the data center 3 is not necessary, the process is terminated, and if necessary, it is determined whether there is an upper agent set as the own agent (S403). When the upper agent is set, the monitoring status is transmitted to the upper agent (S404), and when the upper agent is not set, it is transmitted to the data center 3 (S405). By performing the above operations at regular intervals, the monitoring status can be transmitted to the data center 3 when a failure occurs in the monitoring target, and the monitoring status is transmitted to the data center 3 during other normal operations. There is nothing. As a result, excessive occupation of network traffic by the remote monitoring system 1 in the customer system 2 can be suppressed. In addition, the upper agent does not need to monitor the lower agent set under its own by polling or the like, and can further suppress the network traffic, and the introduction of the remote monitoring system 1 affects other systems. Can be significantly reduced.

  An example of determination of transmission of the monitoring status to the upper agent based on the contents of the monitoring status described in step 402 in FIG. 4 will be described. For example, if a lower level agent detects a failure of a monitored terminal and this is due to a temporary network failure, the network failure will usually recover after a few minutes. The device failure seems to have recovered. However, since the failure is detected once, the failure information is transmitted to the data center 3 through the upper agent as it is, and the data center tries to take a countermeasure based on the received failure information. Therefore, for example, only when the upper agent receives the same failure information three times in succession, it is a temporary failure by performing processing such as transmitting the failure information to a higher agent or the data center 3. It is possible to prevent the remote monitoring system 1 from reacting excessively to a case that can be recovered immediately.

  As another example, it is assumed that the agent that monitors the monitoring target terminal is changed by moving the monitoring target terminal. When a monitored terminal is physically moved from one network to another network, if the information (IP address, name, etc.) that identifies the monitored terminal is not deleted from the agent, the agent accesses the monitored terminal. It is not possible to recognize that the device has stopped, and this monitoring status is sent to the upper agent. However, since the corresponding monitoring target terminal has moved to another network, it will continue to be monitored by another agent, and the agent will move the monitored target terminal to a new one according to the above-described registration operation of the monitoring target terminal. Register as a monitoring target terminal. When moving, the monitoring status that the device has been stopped and the monitoring that a new terminal has been newly added, based on information unique to the monitored terminal, such as the name, serial number, and MAC address, are not normally changed, even if the IP address is changed By associating the new registration process of the target terminal with the higher-level agent or the data center 3, it is possible to recognize that the device has moved instead of being stopped, and that it is processed as unnecessary trouble information in the data center 3. It becomes possible to prevent.

  According to the above monitoring status transmission operation, the upper agent does not perform any active operation with respect to the lower agent. Therefore, the monitoring operation can be performed only by transmitting the monitoring status from the lower agent to the upper agent. it can. However, if a failure occurs in the agent itself and the monitoring status cannot be transmitted, failure monitoring cannot be performed by the above operation. Therefore, the status of the agent itself as shown in FIG. 5 is checked. First, the upper agent checks the reception history of the lower agent registered in itself (S501). It is determined for each registered lower agent whether the monitoring status is received from the lower agent at a predetermined time interval (S502). If the agent is regularly received, the agent has no failure. If the agent has not received the agent regularly, the data center 3 is notified that a failure has occurred in the subordinate agent (S503).

  According to the remote monitoring system mounting method as described above, a plurality of agents 52 are installed in the customer system 2 and are installed as close as possible to the devices to be monitored, so that monitored devices such as the server 53 and the PC 54 are installed. The amount of communication between the agent 52 and the agent 52 can be kept to a minimum, and the possibility that the agent 52 may be hindered by the router 8 or the firewall 7 when accessing the monitoring target device is reduced. A plurality of installed agents 52 operate in parallel or in series. When operating in parallel, each agent 52 transmits monitoring information directly to the data center 3, but the agent 52 filters the information in accordance with preset conditions and transmits the information to further increase the traffic. Suppression can be realized.

  On the other hand, when operating in series, each agent 52 transmits a monitoring status to a predetermined upper agent 6, and at that time, the agent may or may not have information that needs to be transmitted to the upper agent. Instead, at a predetermined interval, information indicating that the agent is active is transmitted to the upper agent as a monitoring status. Further, in the case where a higher agent (for example, the highest agent 10 in FIG. 1) is set for the higher agent (for example, the higher agent 6 in FIG. 1), the higher agent The monitoring status sent from the lower level agent is sent. In this way, the agent 52 repeats the operation of transmitting information to the upper agent 6, and after the information is transmitted to the highest agent 10 which is the highest agent in the customer system 2, the highest agent 10 transmits the information to the data center. To do. When installing an agent, the upper agent 6 and the highest agent 10 do not know all the agents from the beginning, the upper agent does not have information on the lower agent, and the lower agent has information on the upper agent (for example, ID And IP address) only. After the installation of the lower agent, as shown in FIG. 2, first, the upper agent can grasp the lower agent for the first time by accessing the upper agent registered in advance. The upper agent periodically confirms that information is sent from the lower agent, and in the unlikely event that information is not sent, notifies the data center 3 of the failure of the lower agent.

  By configuring the agent in this way, it eliminates the bottleneck of the traffic volume around the agent and the processing performance of the agent, and also eliminates the need to shield the device that the agent wants to monitor from being blocked by a router or firewall along the route. There is. In general, routers and firewalls have a higher degree of freedom in communication from the inside to the outside of the Internet than communication from the outside to the inside, and processing from lower agents to higher agents is triggered. Thus, information can be provided to the data center without changing the router and firewall settings.

  In the remote monitoring system as described above, an example of an implementation method of the agent 52, the upper agent 6, and the highest agent 10 will be described with reference to FIG. Description of the same reference numerals as those in FIG. 1 is omitted. The printer 11 is installed in each of the departments 5 (A to C), and the NAS 12 is installed in each of the departments 5 (X to Z). The printer 11 and NAS 12 are often provided with an OS, a memory, and a non-volatile storage device in many cases. It does not use a memory or a non-volatile storage device, and has sufficient resources. Therefore, by installing the agent 52 of the remote monitoring system 1 on these, an appliance type agent 520 as shown in FIG. And the agent 520 implemented in the form of borrowing from the business server 550 as shown in FIG. 8 or the method of covering the entire customer system as one agent 1000 as shown in FIG. , Inexpensive, unaffected by existing systems, and distributed agents It can be stably operating the remote monitoring system 1.

1 Remote monitoring system 2 Customer system 3 Data center 4 Internet 5 Department 51, 8 Router 52 Agent (subordinate agent)
53, 9 server (monitoring target terminal)
54 PC (monitoring target terminal)
6 Upper agent 7 Firewall 10 Top agent (upper agent)
11 Printer 12 NAS

Claims (3)

Connect agents in a hierarchy of one or more layers,
A remote monitoring system that transmits the status of a monitoring target detected by a terminal agent to a data center,
The status of the monitoring target detected by the end agent is sent to the upper agent of the end agent, and then the monitoring target status is unilaterally sent from the lower agent to the upper agent, The agent implementation method of the remote monitoring system, characterized in that the agent sends the status of the monitoring target to the data center.   The upper agent receives the initial status transmission from the lower agent, registers the lower agent as a new lower agent, and an arbitrary agent manages only the lower agent registered in itself. The remote monitoring system agent mounting method according to claim 1. 3. The agent mounting method for a remote monitoring system according to claim 1, wherein the agent is mounted on any one of NAS (Network Attached Storage), a printer, a copier, and a multifunction peripheral.

Images