JP2014045437A - Network topology generation system, method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a network topology generation system capable of generating a network topology which includes a switch, a physical machine, and a virtual machine and discriminatively displays the physical machine and the virtual machine.SOLUTION: A network topology compensation means comprises the steps of: specifying a first network topology element to be a connection destination of an interface of a switch; specifying a second network topology element to be a connection destination of an interface of the first network topology element on the basis of a link using the first network topology element as a starting point; specifying a third network topology element to be a connection destination of an interface of the second network topology element; and deleting the link between the third network topology element and the switch on the condition that the third network topology element and the switch are matched.

Description

  The present invention relates to a network topology generation system, a network topology generation method, and a network topology generation program. About.

  It is preferable to manage a system used for a service so that it can be understood what service is affected when a failure occurs in a physical machine. In order to perform such management, system configuration information used for services may be held as a CMDB (Configuration Management Database). Here, the service means a business realized by a plurality of machines.

  In order to realize the above management, it is conceivable to create a network topology of a system used for a service. Various techniques related to the network topology have been proposed (see, for example, Patent Documents 1 to 4).

  In the method for creating a network topology described in Patent Document 1, a filtering database collected and stored in a switching hub with a management function is referred to, and terminals belonging to each port of the switching hub with a management function based on the value of the filtering database Get the pair of the physical address and its port number. If a physical address belonging to multiple ports of a lower management switching hub belongs to a single port of the above management hub, the lower management switching hub Judged to be connected to a switching hub with management function.

  Patent Document 2 describes that a guest operating system is informed about the topology of a subset of host resources currently allocated to the guest operating system.

  Patent Document 3 describes that a network management device operates a router so as to transmit a broadcast ping to a network to which a business LAN is connected with reference to a routing table. It is described that when the ping passes through the switch, the MAC address of the transmission source router is associated with the port number of the connection port connected from the switch to the router.

  In Patent Document 4, the physical addresses of PCs (personal computers) that have accessed terminals through hub ports are counted for each hub port. If there is a single physical address for a port, the physical address is It is described that it is determined as the physical address of the PC of the hub port.

  Patent Document 5 describes a method of deriving a port topology using TTL (Time To Live) for a layer 2 network device managed only by a MAC address.

JP 2002-190819 A (first page) JP 2006-178933 A (paragraph 0040) JP 2007-228382 A (paragraphs 0030 and 0031) JP-A-8-191326 (first page) JP 2008-172449 A (first page)

  When a system used for a service is managed by the CMDB, as time elapses, there may be a discrepancy between the contents of the CMDB and the actual system configuration. It may not be possible to fully grasp whether there is an impact on

  Therefore, it is preferable to create a network topology of the system used for the service. As described above, in the technique described in Patent Document 4, when the number of physical addresses counted for a certain port is singular, the physical address is determined as the physical address of the PC connected to the port. However, the technique described in Patent Document 4 does not consider a virtual machine. Therefore, with the technique described in Patent Document 4, a network topology relating to a PC and a hub can be created, but a network topology including a virtual machine cannot be created. A virtual machine running on a physical machine also has an address. Therefore, when a physical machine is connected to a port and a virtual machine is operating on the physical machine, a plurality of addresses are aggregated for the port. Cannot determine if connected.

  In general, a MAC address (Media Access Control Address) is used in creating a network topology. The switch can communicate with the virtual machine via the physical machine. For each port, a list of MAC addresses of physical machines and virtual machines connected to the ports is held. However, it is not possible to distinguish between a physical machine and a virtual machine based only on MAC address information. Therefore, it has been difficult to create a network topology including switches, physical machines, and virtual machines.

  Therefore, the present invention provides a network topology including a switch, a physical machine, and a virtual machine, and a network topology generation system, a network topology generation method, and a network topology generation method capable of generating a network topology that distinguishes between a physical machine and a virtual machine. An object is to provide a network topology generation program.

  The network topology generation system according to the present invention assigns the type of node or switch, the identification information of the interface of the node or switch, and the interface itself from the nodes and switches that are physical machines or virtual machines. For each interface of the switch, the address of the node to which the interface is connected or the interface of the switch is obtained from the switch, and the interface of the node is obtained from the node corresponding to the physical machine. Information acquisition means for acquiring the interface address of the node corresponding to the virtual machine to which the interface is connected, and the node or switch interface to which the switch interface is connected Create a link between the switch and another node or switch by referring to the address, refer to the address of the interface of the node that is the connection destination of the node interface, and link between the nodes. A network topology generation means for generating a network topology including nodes and switches as elements by creating a link having a node as an end point, and a network topology element that is a connection destination of a switch interface. A second network topology element which is an element of the network topology to which the interface of the first network topology element is connected based on a link starting from the first network topology element Identify and further A third network topology element that is an element of a network topology to which the interface of the second network topology element is connected is specified, and the third network topology element and the switch are matched on condition that the third network topology element matches the switch. It is characterized by comprising a network topology correcting means for deleting a link between a switch and a node corresponding to a virtual machine by deleting a link between the network topology element and the switch.

  In addition, the network topology generation method according to the present invention includes a node and a switch that are physical machines or virtual machines, a type of node or switch, identification information of the interface of the node or switch, and the interface itself. For each interface of the switch, the node to which the interface is connected or the address of the interface of the switch is acquired from the switch, and the node corresponding to the physical machine For each interface, obtain the interface address of the node corresponding to the virtual machine that is the connection destination of the interface, and refer to the address of the node or switch interface that is the connection destination of the switch interface Create a link between the switch and another node or switch, refer to the interface address of the node that is the connection destination of the node interface, and link between the nodes and become the connection destination A network topology including nodes and switches as elements is created by creating a link having a node as an end point, and a first network topology element that is an element of the network topology to which a switch interface is connected is identified. Based on the link starting from the first network topology element, a second network topology element that is an element of the network topology to which the interface of the first network topology element is connected is specified. Network topology element interface A third network topology element, which is an element of the network topology to which the network is connected, and between the third network topology element and the switch on condition that the third network topology element matches the switch The link between the switch and the node corresponding to the virtual machine is deleted by deleting the link.

  In addition, the network topology generation program according to the present invention includes, from a node that is a physical machine or a virtual machine, and a switch, a type that is a node or a switch, identification information of an interface of the node or switch, In addition to obtaining the address assigned to the interface itself, for each interface of the switch, obtain the address of the node to which the interface is connected or the interface address of the switch from the node corresponding to the physical machine. For each interface of the node, information acquisition processing for acquiring the interface address of the node corresponding to the virtual machine to which the interface is connected, and the node or switch to which the switch interface is connected Create a link between the switch and another node or switch by referring to the interface address of the node, and refer to the interface address of the node to which the node interface is connected. Network topology generation processing for generating a network topology including nodes and switches as elements by creating a link whose destination is the node that is the connection destination, and the network topology that is the connection destination of the switch interface A first network topology element that is an element is specified, and a second network topology element that is a connection destination of an interface of the first network topology element based on a link starting from the first network topology element Network A log element is specified, and a third network topology element, which is a network topology element to which the interface of the second network topology element is connected, is specified, and the third network topology element matches the switch. A network topology correction process for deleting a link between a switch and a node corresponding to a virtual machine is performed by deleting a link between the third network topology element and the switch on condition that To do.

  According to the present invention, it is possible to generate a network topology including a switch, a physical machine, and a virtual machine, in which a physical machine and a virtual machine are distinguished from each other.

It is a block diagram which shows the structural example of the network topology production | generation system of this invention. It is explanatory drawing which shows the example of the IP address which a connection information storage part memorize | stores in an initial state. It is explanatory drawing which shows the example of the information which an authentication information storage part memorize | stores. It is explanatory drawing which shows the example of the state by which the protocol and the user name were added to the IP address. It is a schematic diagram which shows the example of the information acquired from one network topology element. It is explanatory drawing which shows the specific example of the information shown in FIG. It is explanatory drawing which shows the specific example of the information shown in FIG. It is explanatory drawing which shows typically the example of the information which a service information storage part memorize | stores. It is explanatory drawing which shows the example of the network topology to which the information was added by the service information addition part. It is a flowchart which shows the example of process progress of this invention. It is a flowchart which shows the example of the process progress of step S3. It is a flowchart which shows the example of the process progress of step S3. It is explanatory drawing which shows the example of the link obtained as a result of step S14. It is explanatory drawing which shows the example of the link between the nodes added by the loop process of step S18-S21. It is a flowchart which shows the example of the process progress of step S4. It is a flowchart which shows the example of the process progress of step S4. It is explanatory drawing which shows S, X, Y, Z and the link deleted. It is explanatory drawing which shows the example of the network topology after step S4. It is a block diagram which shows the example of the minimum structure of the network topology production | generation system of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

First, terms in the present embodiment will be described.
“Network topology element” means a switch and a node that are elements of a network topology.

  The “node” is a physical machine or a virtual machine connected to the switch. Therefore, there are nodes corresponding to physical machines and nodes corresponding to virtual machines.

  Each network topology element has an interface with another network topology element, and each interface has an address. For example, when the network topology element is a switch, the switch has a port as an interface and an address for each port. When the network topology element is a node, the node has a network device as an interface, and has an address for each network device. In the present embodiment, a case where an address assigned to each interface (a port in a switch, a network device in a node) of a network topology element is a MAC address will be described as an example. However, an address assigned to each interface is limited to a MAC address. Not. For example, a WWPN (World Wide Port Name) address may be used.

  In addition to the MAC addresses assigned to the ports and network devices, IP addresses are assigned to the switches and the nodes themselves. In this embodiment, a case where an IP address is used as identification information of individual network topology elements (switches, nodes) will be described as an example.

  The switch has a plurality of ports, and holds the MAC address of the interface of the node to which the port is connected for each port.

  A node corresponding to a physical machine and a node corresponding to a virtual machine include a plurality of network devices.

  The node corresponding to the physical machine holds the MAC address of the network device of the node corresponding to the virtual machine, which is the connection destination of its own network device.

  The physical machine is a machine that can start a plurality of virtual machines. As an example of a physical machine, for example, a machine on which an OS (Operating System) such as Windows (registered trademark) or Linux (registered trademark) operates can be cited. Hereinafter, the physical machine may be referred to as PM (Physical Machine).

  The virtual machine is a guest OS that operates on a physical machine. Hereinafter, a virtual machine may be referred to as a VM (Virtual Machine).

  Next, a configuration example of the network topology generation system of the present invention will be described. FIG. 1 is a block diagram showing a configuration example of a network topology generation system according to the present invention. In FIG. 1, the switch 31, physical machines (PM) 41 to 4n, virtual machines (VM) 51 to 5n, and virtual machines (VM) 61 to 6n are also illustrated along with the network topology generation system. These switches, physical machines, and virtual machines are network topology elements included in the system used to provide services. The number of switches is not limited to one. Further, the number of virtual machines and the number of physical machines are not limited respectively.

  The network topology generation system 1 of the present embodiment includes an element detection unit 2, an information acquisition unit 3, a network topology generation unit 4, a network topology correction unit 5, a service information addition unit 6, and an authentication information storage unit 7. A connection information storage unit 8, an element information storage unit 9, a service information storage unit 10, and an output unit 11.

  The connection information storage unit 8 stores in advance all IP addresses that can be identification information of network topology elements. For example, in the case of the configuration illustrated in FIG. 1, the IP address of the network (subnet) having the switch 31 as a route is stored. These IP addresses may include unused IP addresses (that is, IP addresses for which no corresponding network topology element exists). FIG. 2 shows an example of the IP address that the connection information storage unit 8 stores in the initial state. These IP addresses may be stored in advance in the connection information storage unit 8 by the administrator. Of the IP addresses stored in the connection information storage unit 8, the IP address assigned to the network topology element is added by the element detection unit 2 with the protocol, user name, etc. for accessing the network topology element. The

  The authentication information storage unit 7 holds in advance all information necessary for accessing any network topology element. FIG. 3 is an explanatory diagram illustrating an example of information stored in the authentication information storage unit 7. In the example shown in FIG. 3, a plurality of combinations (denoted as records) of protocols, user names, and passwords are stored. Such information may be stored in advance in the authentication information storage unit 7 by the administrator. Further, the network topology element corresponding to the combination of the protocol, the user name, and the password need not be stored in the authentication information storage unit 7.

  The element detection unit 2 detects accessible network topology elements. Specifically, the element detection unit 2 selects an IP address stored in the connection information storage unit 8. Then, the element detection unit 2 pings the selected IP address, and if there is a response, logs in with the first record (combination of protocol, user name and password) stored in the authentication information storage unit 7 Determine if you can. If the login is not possible, the element detection unit 2 performs the same determination using the next record stored in the authentication information storage unit 7. Thereafter, when the same processing is repeated and login is possible, the connection information storage unit 8 stores the protocol and the user name used for login in association with the selected IP address. An example of a state in which a protocol and a user name are added to the IP address is shown in FIG. The element detection unit 2 also associates the IP address with the password used for login. The element detection unit 2 may store this password in the connection information storage unit 8 in association with the IP address. This means that the network topology element (switch or node) identified by the selected IP address has been detected. Thereafter, it is possible to log in to the network topology element of the IP address using the protocol, user name, and password associated with the IP address, and acquire the network topology element.

  The element detection unit 2 sequentially selects other IP addresses and performs the same processing. For an IP address that has not responded to a ping, there is no network topology element with that IP address.

  The information acquisition unit 3 logs into the network topology element detected by the element detection unit 2 and acquires information on the type of whether the network topology element is a node or a switch. For example, the information acquisition unit 3 transmits a command that normally responds only to a node to the network topology element, and if there is a normal response, determines that the network topology element is a node, and if there is no normal response, The network topology element may be determined as a switch. Further, for example, the information acquisition unit 3 transmits a command that normally responds only to the switch to the network topology element, and if there is a normal response, determines that the network topology element is a switch, and if there is no normal response, The network topology element may be determined as a node.

  Further, the information acquisition unit 3 requests the network topology element detected by the element detection unit 2 for a list of a set of identification information of an interface included in the network topology element and a MAC address assigned to the interface. And the information acquisition part 3 acquires the list | wrist of the group of the identification information and MAC address of the interface which the network topology element has from a network topology element. When the network topology element is a switch, the information acquisition unit 3 acquires a list of port identification information and a set of MAC addresses of the port. When the network topology element is a node, the information acquisition unit 3 acquires a list of a set of network device identification information and the MAC address of the network device.

  Further, for the network topology element detected by the element detection unit 2, the information acquisition unit 3 has, for each interface included in the network topology element, the interface identification information and the network topology that is the connection destination of the interface. Requests a list of MAC addresses for the element's interface. Then, the information acquisition unit 3 acquires, for each interface of the network topology element, a list of interface identification information and the MAC address of the interface of the network topology element to which the interface is connected.

  For each port, the switch maintains a list of pairs of port identification information and the MAC address of the node or switch interface to which the port is connected. Therefore, in response to a request from the information acquisition unit 3, the switch sends a list of pairs of port identification information and the MAC address of the connection destination node or switch interface to the information acquisition unit 3 for each port, and acquires information. The unit 3 may acquire the information.

  In addition, for each network device, the node corresponding to the physical machine holds a list of pairs of the network device identification information and the MAC address of the network device of the node corresponding to the virtual machine to which the network device is connected. ing. Therefore, the node corresponding to the physical machine is a list of pairs of the identification information of the network device and the MAC address of the network device of the connection destination node (node corresponding to the virtual machine) in response to a request from the information acquisition unit 3. Is sent to the information acquisition unit 3 for each network device, and the information acquisition unit 3 may acquire the information.

  Note that the node corresponding to the virtual machine does not return to the information acquisition unit 3 the set of the identification information of the network device and the MAC address of the connection destination of the network device. Therefore, for the node corresponding to the virtual machine, the information acquisition unit 3 does not acquire a list of sets of the interface identification information and the MAC address of the interface of the network topology element to which the interface is connected. Absent.

  That is, the information acquisition unit 3 acquires, from the switch, the MAC address of the node to which the port is connected or the interface of the switch for each port, and from the node corresponding to the physical machine, for each network device, The MAC address of the network device of the node corresponding to the virtual machine that is the connection destination is acquired.

  The information acquisition unit 3 stores the information acquired from the network topology element in the element information storage unit 9 for each network topology element detected by the element detection unit 2. FIG. 5 is a schematic diagram illustrating an example of information acquired from one network topology element. The first line shown in FIG. 5 is the type of network topology element. The second line shown in FIG. 5 is the IP address of the network topology element itself. The third row shown in FIG. 5 is a list of sets of interface identification information and MAC addresses assigned to the interfaces included in the network topology element. The fourth and subsequent lines shown in FIG. 5 are a list of a set of interface identification information included in the network topology element and a MAC address to which the interface is connected. Specific examples of the information shown in FIG. 5 are shown in FIGS.

  FIG. 6 shows an example of information acquired from the switch. The first line shown in FIG. 6 indicates that the type of network topology element is a switch. The second line indicates that the IP address of the switch is “192.168.0.1”. The third line shown in FIG. 6 shows, for example, the identification information of the port “GigabitEather0 / 0/1” and the MAC address “100000000001” of the port itself in the switch. For other ports, a set of port identification information and the MAC address of the port itself is shown in the same format. Further, in the fourth and subsequent lines shown in FIG. 6, for example, a set of port identification information “GigabitEather0 / 0/1” and the MAC address “200000000001” of the connection destination of the port is shown. Another set of port identification information and the MAC address to which the port is connected is shown in the same format.

  FIG. 7 shows an example of information acquired from a node. The first line shown in FIG. 7 indicates that the type of network topology element is a node. The second line indicates that the IP address of the node is “192.168.0.2”. The third line shown in FIG. 7 shows, for example, the identification information of the network device “eth0” and the MAC address “200000000001” of the network device itself in the node. For other network devices, a set of network device identification information and the MAC address of the network device itself is shown in the same format. Furthermore, the fourth line shown in FIG. 7 shows, for example, a set of network device identification information “vnet0” and the MAC address “300000000001” of the connection destination of the network device. If there are other such sets, they may be shown in the same format as the fourth row. In the information acquired from the node corresponding to the physical machine, the information corresponding to the fourth and subsequent lines shown in FIG. 7 indicates the MAC address of the network device of the node corresponding to the virtual machine that is the connection destination of the network device of the physical machine. Show.

  Further, information obtained from the node corresponding to the virtual machine does not include information corresponding to the fourth and subsequent lines shown in FIG. This is because, as described above, the node corresponding to the virtual machine does not return to the information acquisition unit 3 the list of the combination of the identification information of the network device and the MAC address of the connection destination of the network device.

  The information acquisition unit 3 stores information acquired from network topology elements (information illustrated in FIGS. 6 and 7) in the element information storage unit 9, respectively.

  The network topology generation unit 4 refers to the information stored in the element information storage unit 9 (information illustrated in FIGS. 6 and 7) for each detected network topology element, and creates a link between the network topology elements. To generate a network topology. The network topology generation operation will be described later.

  The switch can communicate with the virtual machine via the physical machine. Therefore, when the virtual machine is activated on the physical machine, the MAC address of the network device of the physical machine and the MAC address of the network device of the virtual machine exist as connection destinations of the switch ports. In this case, the network topology generated by the network topology generation unit 4 is represented as if both a node corresponding to the physical machine and a node corresponding to the virtual machine are connected to one port of the switch.

  If the network topology correction unit 5 is represented in the network topology such that both a node corresponding to the physical machine and a node corresponding to the virtual machine are connected to one port of the switch, the switch, The network topology is corrected by deleting the link with the node corresponding to the virtual machine. As a result, in the network topology, the physical machine has a link with the switch, but the virtual machine has no link with the switch. Therefore, a network topology in which a node corresponding to a physical machine and a node corresponding to a virtual machine are distinguished depending on whether or not a link exists with the switch. The link deletion operation by the network topology correction unit 5 will be described later.

  The service information storage unit 10 stores a correspondence relationship among service providers, services, clusters, and addresses (IP addresses) for identifying virtual machines. FIG. 8 is an explanatory diagram schematically illustrating an example of information stored in the service information storage unit 10.

  As shown in FIG. 8, a service provider is associated with a service provided by the service provider. For example, FIG. 8 shows that the service provider 1 provides the services 11 and 12.

  Each service is associated with a cluster used in the service. A cluster is a set of functions included in a service. Examples of the cluster include, for example, a Web layer function, an application layer function, a set of DB layer functions (Web three layers), and the like. In FIG. 8, for example, the clusters 11 to 13 are associated with the service 11.

  In addition, for each cluster, a set of addresses (in this example, IP addresses) that are identification information of each virtual machine used in the cluster is associated. In FIG. 8, for example, address 11, address 12,.

  The service information storage unit 10 only needs to store at least information indicating a correspondence relationship between a service and identification information (IP address) of a virtual machine used to provide the service.

  The service information adding unit 6 refers to the information stored in the service information storage unit 10 (for example, the information illustrated in FIG. 8), and adds the virtual machine to the virtual machine in the network topology. Performs processing to add the corresponding service information. The service information adding unit 6 may create an icon for identifying a service and associate the icon with a node corresponding to the virtual machine.

  The service information adding unit 6 may add not only service information but also cluster and service provider information. For example, the service information adding unit 6 may create an icon for identifying a cluster and associate the icon with a node corresponding to the virtual machine. A corresponding service icon may be associated with the icon representing the cluster. Furthermore, an icon for identifying service provision may be created, and the icon representing the service may be associated with the icon of the corresponding service provider.

  An example of the network topology to which information is added by the service information adding unit 6 is shown in FIG. Part A shown in FIG. 9 represents the network topology obtained by the processing up to the network topology correction unit 5. Part B shown in FIG. 9 represents information added by the service information adding unit 6.

  The output unit 11 is, for example, a display device, and displays a network topology (network topology illustrated in FIG. 9) to which information such as a service is added by the service information adding unit 6.

  The element detection unit 2, the information acquisition unit 3, the network topology generation unit 4, the network topology correction unit 5, and the service information addition unit 6 are realized by a CPU of a computer that operates according to a network topology generation program, for example. In this case, for example, the CPU reads a network topology generation program from a computer-readable recording medium, and in accordance with the program, the element detection unit 2, the information acquisition unit 3, the network topology generation unit 4, the network topology correction unit 5, and the service information What is necessary is just to operate | move as the addition part 6. FIG. Further, the element detection unit 2, the information acquisition unit 3, the network topology generation unit 4, the network topology correction unit 5, and the service information addition unit 6 may be realized by separate hardware.

  Further, the authentication information storage unit 7, the connection information storage unit 8, the element information storage unit 9, and the service information storage unit 10 are realized by a storage device, for example. Alternatively, each storage unit 7 to 10 may be realized by a database server or the like.

  Next, the operation will be described. FIG. 10 is a flowchart showing an example of processing progress of the present invention.

  The element detection unit 2 detects accessible network topology elements (switches, nodes) (step S1). In step S1, the element detection unit 2 associates the IP address of the detected network topology element with the protocol, user name, and password used at the time of access. Since the operation of step S1 (the operation of the element detection unit 2) has already been described, the description thereof is omitted here. As a result of step S1, it is possible to log in to the network topology element of the IP address using the protocol, user name, and password associated with the IP address, and acquire information on the network topology element.

  Next, the information acquisition unit 3 logs in each network topology element detected in step S1, and acquires information from the network topology element (step S2).

  Specifically, the information acquisition unit 3 acquires information on the type of whether the logged-in network topology element is a node or a switch.

  Further, the information acquisition unit 3 requests the logged-in network topology element for a list of a set of interface identification information and MAC addresses assigned to the interface of the network topology element, and acquires the list.

  Further, for each interface that the network topology element has for the logged-in network topology element, the information acquisition unit 3 identifies the interface identification information and the MAC address of the interface of the network topology element that is the connection destination of the interface Request a list of and get the list.

  When the network topology element is a switch, a list of identification information of the switch interface (port) and the MAC address of the node or switch interface to which the port is connected is obtained.

  When the network topology element is a node corresponding to a physical machine, the identification information of the interface (network device) of the node and the MAC address of the interface of the node corresponding to the virtual machine to which the network device is connected A list of pairs is obtained.

  When the network topology element is a node corresponding to a virtual machine, a list of a set of interface identification information and a MAC address of a connection destination cannot be obtained for the node.

  Hereinafter, a group of information acquired from one network is referred to as element information.

  The information acquisition unit 3 causes the element information storage unit 9 to store element information for each network topology element. In the present embodiment, a case where the information acquisition unit 3 stores element information in the format illustrated in FIG. 5 (more specifically, FIG. 6 and FIG. 7) will be described as an example. Further, in the example shown in FIGS. 5 to 7, information (list of a set of network topology element interface identification information and its connection destination MAC address) shown in the fourth and subsequent lines is hereinafter referred to as linkage information. However, the link information is not included in the element information of the node corresponding to the virtual machine. Moreover, FIG.6 and FIG.7 is an illustration, and the information acquisition part 3 may memorize | store element information in the element information storage part 9 in another format.

  After step S2, the network topology generation unit 4 generates a network topology by creating a link between network topology elements based on the information acquired in step S2 (step S3).

  FIG. 11 and FIG. 12 are flowcharts showing an example of the processing progress of step S3. In step S3, the network topology generation unit 4 selects element information whose type is a switch (step S11). The switch corresponding to this element information is one end of the link created in step S14.

  Next, the network topology generation unit 4 selects identification information of a port that has not been selected from the element information (switch element information) (step S12). For example, when the element information illustrated in FIG. 6 is selected, unselected identification information is selected from the port identification information in the third row.

  Subsequently, the network topology generation unit 4 extracts all the MAC addresses paired with the identification information of the port selected in Step S12 from the link information in the selected element information (Step S13). For example, it is assumed that the element information illustrated in FIG. 6 is selected, and the port identification information “GigabitEather0 / 0/1” is selected in step S12. In this case, MAC addresses “200000000001”, “200000000002”, and “200000000003” that are paired with “GigabitEather0 / 0/1” are extracted from the link information in the element information (here, the fourth and subsequent lines in FIG. 6). .

  Subsequently, the network topology generation unit 4 specifies all the element information of the network topology element (node or switch) having the interface to which the MAC address extracted in step S13 is assigned, and the port indicated by the identification information selected in step S12 And a link with the interface to which the MAC address extracted in step S13 is assigned (step S14). In the above example, the element information (for example, the element information illustrated in FIG. 7) of the network topology element having the interface to which “200000000001” is assigned as the MAC address is specified, and “GigabitEather0 / 0/1 selected in step S12” A link is created between the port indicated by "" and the interface assigned with "200000000001" among the interfaces of the node represented by the element information shown in FIG. Similar processing is performed for the other MAC addresses extracted in step S13. By identifying the network topology element having the interface to which the MAC address extracted in step S13 is assigned, the network topology element serving as one end of the link is determined.

  The network topology generation unit 4 does not have to determine the link direction for the link created in step S14.

  FIG. 13 is an explanatory diagram illustrating an example of a link obtained as a result of step S14. In FIG. 13, the names of the network topology elements are indicated as “switch01”, “ND01”, “ND02”, and “ND03” for convenience, but the network topology elements are identified by IP addresses.

  In the example shown in FIG. 13, link 1 is created between “Port1” of “switch01” and “Dev1” of “ND01”, and “Port1” of “switch01” and “Dev2” of “ND02” A link 71 is created between “Port01” of “switch01” and “Dev2” of “ND03”. Since it cannot be determined whether the node is a virtual machine or a physical machine from the element information itself of the node, it cannot be determined at this time whether “ND01”, “ND02”, and “ND03” are virtual machines or physical machines.

  After step S14, the network topology generation unit 4 determines whether there is identification information of an unselected port in the element information of the switch selected in step S11 (step S15). If there is identification information of an unselected port (Yes in step S15), the network topology generation unit 4 repeats the processing after step S12. By repeating the loop processing of steps S12 to S15, a link is created for each port of the switch. For example, in the example illustrated in FIG. 13, “Port2” and “Port3” of “switch01” are also created in the same manner as “Port1”.

  Further, when there is no identification information of an unselected port (No in Step S15), the network topology generation unit 4 determines whether there is element information of an unselected switch (Step S16). If there is element information of an unselected switch (Yes in step S16), the network topology generation unit 4 repeats the processing after step S11. When there is no element information of an unselected switch (No in step S16), the process proceeds to step S17. At this point, all the links between the ports of the switch and the interfaces of other nodes have been created. The direction of the link may not be determined.

  In step S17, the network topology generation unit 4 selects element information whose type is a node (step S17). The node corresponding to this element information is one end of the link created in step S20.

  Next, the network topology generation unit 4 selects the identification information of the network device that has not been selected from the element information (node element information) (step S18). For example, when the element information illustrated in FIG. 7 is selected, unselected identification information is selected from the identification information of the network device in the third row.

  Subsequently, the network topology generation unit 4 extracts all the MAC addresses paired with the identification information of the network device selected in Step S18 from the link information in the selected element information (Step S19). For example, it is assumed that the element information illustrated in FIG. 7 is selected, and the network device identification information “vnet0” is selected in step S18. In this case, the MAC address “300000000001” paired with “vnet0” is extracted from the link information in the element information (here, the fourth and subsequent lines in FIG. 7).

  Subsequently, the network topology generation unit 4 identifies the element information of the node having the interface to which the MAC address extracted in step S19 is assigned, and the network device indicated by the identification information selected in step S18 and extracted in step S19. A link is created with the interface to which the MAC address is assigned (step S20). In the above example, the element information of the node having the interface to which “300000000001” is assigned as the MAC address is specified, and “300000000001” is selected from the network device indicated by “vnet0” selected in step S18 and the interface of the node. Create a link with the interface to which is assigned. Similar processing is performed for the other MAC addresses extracted in step S19. By identifying a node having an interface to which the MAC address extracted in step S19 is assigned, a node serving as one end of the link is determined.

  The network topology generation unit 4 determines the direction of the link for the link created in step S20. Specifically, the network topology generation unit 4 sets a node having an interface to which the MAC address extracted in step S19 is assigned as a link end point. Accordingly, the network topology generation unit 4 sets the node corresponding to the element information selected in step S17 as the link start point.

  Note that if the link information does not exist in the element information selected in step S17, the network topology generation unit 4 does not perform the processes in steps S19 and S20. The node corresponding to the virtual machine does not include linkage information. Accordingly, in step S20, a link starting from the node corresponding to the virtual machine is not created. In step S20, a link starting from a node corresponding to the physical machine is created.

  After step S20, the network topology generation unit 4 determines whether or not there is identification information of an unselected network device in the element information of the node selected in step S17 (step S21). If there is identification information of an unselected network device (Yes in step S21), the network topology generation unit 4 repeats the processing after step S18. When there is no identification information of the unselected network device (No in Step S21), the network topology generation unit 4 determines whether there is element information of the unselected node (Step S22). If there is element information of an unselected node (Yes in Step S22), the network topology generation unit 4 repeats the processes after Step S17. When there is no element information of an unselected node (No in step S22), the process of step S3 is terminated.

  By repeating the loop processing of steps S17 to S22, a link between nodes starting from the node corresponding to the physical machine is created. FIG. 14 shows an example of a link between nodes added by the loop processing of steps S18 to S21 after the state shown in FIG. In the example shown in FIG. 14, a link 4 is created between “Dev2” of “ND01” and “Dev1” of “ND02”, and between “Dev3” of “ND01” and “Dev1” of “ND03”. A link 5 is created. The links 4 and 5 shown in FIG. 14 are links starting from “ND01”.

  After step S3, the network topology correction unit 5 is represented in the network topology such that both a node corresponding to the physical machine and a node corresponding to the virtual machine are connected to one port of the switch. In this case, the network topology is corrected by deleting the link between the switch and the node corresponding to the virtual machine (see step S4, FIG. 10).

  15 and 16 are flowcharts showing an example of the progress of the process in step S4. In step S4, the network topology correction unit 5 determines whether there is a switch that has not yet been selected in step S32 described later among the switches included in the network topology generated in step S3 (step S31).

  When there is an unselected switch (Yes in step S31), the network topology correction unit 5 selects an unselected switch (step S32). Let S be the switch selected in step S32.

  Next, the network topology correction unit 5 determines whether there is a link that has not yet been selected in step S34 among the links between the switch S and other network topology elements (step S33). If there is no unselected link (No in step S33), the processes after step S31 are repeated.

  If there is an unselected link (Yes in step S33), the network topology correction unit 5 selects an unselected link among the links between the switch S and other network topology elements, and further, the network topology element of the link destination Is selected (step S34). Let X be the network topology element selected in step S34. Since the direction between the switch and the other network topology element does not need to be determined, the direction of the link does not need to be considered in steps S33 and S34.

  Next, the network topology correction unit 5 determines the type of the network topology element X (step S35). When the type of the network topology element X is a switch, the processes after step S33 are repeated.

  When the type of the network topology element X is a node, the network topology correction unit 5 is a link other than the link between the network topology element X (here, the node X) and the switch S, and the node X is the starting point. It is determined whether there is a link that has not yet been selected in step S37 among the links having the network topology element as an end point (step S36). If there is no unselected link (No in step S36), the processes in and after step S33 are repeated.

  If there is an unselected link (Yes in step S36), the network topology correction unit 5 is a link other than the link between the node X and the switch S, and the node X is a start point and other network topology elements are end points. The unselected link is selected from the links to be. Then, the network topology correction unit 5 selects the network topology element of the link destination (that is, the end point of the selected link) (step S37). Let Y be the network topology element selected in step S37.

  Next, the network topology correction unit 5 determines the type of the network topology element Y (step S38). When the type of the network topology element Y is a switch, the processes after step S36 are repeated.

  When the type of the network topology element Y is a node, the network topology correction unit 5 is a link between the network topology element Y (here, the node Y) and another network topology element, and the node Y, the node X, It is determined whether or not there is a link that has not yet been selected in step S40 among links other than those between (step S39). If there is no unselected link (No in step S39), the processes in and after step S36 are repeated.

  If there is an unselected link (Yes in step S39), the network topology correction unit 5 is a link between the node Y and another network topology element other than the link between the node Y and the node X. Select an unselected link from among the links. Then, the network topology correction unit 5 selects the network topology element of the link destination (step S40). Let the network topology element selected in step S40 be Z. In steps S39 and S40, the link direction need not be considered.

  Next, the network topology correction unit 5 determines whether or not the switch S and the network topology element Z match (step S41). If they do not match (No in step S41), the processes in and after step S39 are repeated.

  If the switch S and the network topology element Z match (Yes in step S41), the network topology correction unit 5 deletes the link between the node Y and the network topology element Z (that is, the switch S) ( Step S42). After step S42, the processes after step S39 are repeated.

  The network topology correction unit 5 repeats the processes in steps S31 to S42, and when it is determined in step S31 that there are no unselected switches (No in step S31), the process in step S4 ends.

  FIG. 17 is an explanatory diagram showing S, X, Y, Z and the link to be deleted. S, X, Y, and Z shown in FIG. 17 have the following relationship. Both the nodes X and Y are linked to the switch S. Further, a link is created between the nodes X and Y, starting from the node X and ending at the node Y. The link destination from the node Y (however, other than X) is Z. In the processing of steps S31 to S42 (see FIG. 15 and FIG. 16), when Z matches with the switch S, processing for deleting the link between the switch S and the node Y is realized. Also in the example shown in FIG. 17, since Z matches S, the link between S and Y is deleted.

  When a plurality of nodes exist as link destinations of the same port of the switch, one node corresponds to a physical machine and the other nodes correspond to virtual machines. A node corresponding to a physical machine is a starting point of a link between nodes, but a node corresponding to a virtual machine is not a starting point of a link between nodes, but is an end point. Therefore, when the relationship shown in FIG. 17 is established, the node X is a node corresponding to a physical machine, and the node Y is a node corresponding to a virtual machine. Therefore, deleting the link in the process of step S42 means deleting the link between the switch and the virtual machine. Therefore, by performing the processing of step S4 (steps S31 to S42), each link created between the switch and the virtual machine can be deleted.

  Referring to FIG. 14 as an example, “switch01” corresponds to the switch S and the network topology element Z, and “ND01” corresponds to the node X. “ND02” and “ND03” both correspond to the node Y. Accordingly, the link 71 between “switch01” and “ND02” and the link 72 between “switch01” and “ND03” are deleted in the process of step S4.

  In FIG. 14, the links related to “Port2” and “Port3” of “switch01” are omitted. However, the links between “Port2” and “Port3” with the nodes corresponding to the virtual machines are not shown. Deleted. An example of the network topology after step S4 is shown in FIG. In the network topology shown in FIG. 18, nodes “ND01”, “ND02”, and “ND06” having links with switches are nodes corresponding to physical machines, and have links with switches. Nodes “ND02”, “ND03”, “ND05”, and “ND07” that are not included are nodes corresponding to the virtual machine. In the network topology after step S4 of the present invention, a physical machine and a virtual machine can be distinguished from each other. In FIG. 18, illustration of the direction of the link between the nodes is omitted.

  After step S4, the service information adding unit 6 uses the information (information illustrated in FIG. 10) stored in the service information storage unit 10 to correspond to a virtual machine in the network topology after the correction in step S4. And the service information are associated with each other and the result is displayed on the output unit 11 (step S5). The service information storage unit 10 stores correspondence relationships among service providers, services, clusters, and virtual machine IP addresses (identification information). The element information of the node corresponding to the virtual machine also includes an IP address as identification information of the node corresponding to the virtual machine. Therefore, for example, a node corresponding to a virtual machine having a certain IP address may be associated with a service icon associated with the IP address.

  Further, as already described, the service information addition unit 6 may create an icon for identifying a cluster and associate the icon with a node corresponding to the virtual machine. A corresponding service icon may be associated with the icon representing the cluster. Furthermore, an icon for identifying service provision may be created, and the icon representing the service may be associated with the icon of the corresponding service provider.

  As a result, the network topology illustrated in FIG. 9 is obtained. In this way, the service information adding unit 6 causes the output unit 11 to display the network topology to which information such as service has been added in step S5. In FIG. 9, switches, physical machines, and virtual machines are shown in a simplified manner, but a network topology including information such as a MAC address may be displayed as illustrated in FIG.

  As described above, according to the present embodiment, it is possible to generate a network topology that distinguishes between a physical machine and a virtual machine. By using such a network topology, it is possible to easily grasp which service is affected when a failure occurs in a physical machine.

  In the present embodiment, in step S5, service information defined in the service information storage unit 10 is associated with a node corresponding to the virtual machine. Therefore, when a failure occurs in a physical machine, it can be further easily understood which service is affected.

  In addition, you may complete | finish a process by step S4 of said embodiment. In this case, the network topology correction unit 5 may display the network topology obtained as a result of step S4 on the output unit 11. In this case, a network topology with no service information added is displayed. In this case, the network topology generation system 1 may not include the service information adding unit 6 and the service information storage unit 10. And step S5 may be omitted.

  Further, when information for accessing the network topology element (for example, protocol, user name, password) is known for each network topology element, the process of step S1 may be omitted. In this case, the network topology generation system 1 may not include the element detection unit 2.

  Next, an example of the minimum configuration of the present invention will be described. FIG. 19 is a block diagram showing an example of the minimum configuration of the network topology generation system of the present invention. The network topology generation system of the present invention includes an information acquisition unit 3, a network topology generation unit 4, and a network topology correction unit 5.

  The information acquisition unit 3 determines whether a node is a node or a switch, identification information about the interface of the node or switch, and an address assigned to the interface itself, from a node or switch that is a physical machine or a virtual machine For each interface of the switch, the address of the node to which the interface is connected or the interface of the switch is acquired from the switch, and from the node corresponding to the physical machine, for each interface of the node, Gets the interface address of the node corresponding to the virtual machine that is the connection destination of the interface.

  The network topology generation unit 4 creates a link between the switch and another node or switch with reference to the node or switch interface address that is the connection destination of the switch interface, and A network topology that includes nodes and switches as elements is created by creating links that are links between nodes, with the destination node as the end point, by referring to the address of the node's interface To do.

  The network topology correction unit 5 identifies a first network topology element (for example, X) that is an element of the network topology to which the switch interface is connected, and is based on the link starting from the first network topology element. The second network topology element (for example, Y) that is the element of the network topology to which the interface of the first network topology element is connected is specified, and the interface of the interface of the second network topology element is further specified. A third network topology element (for example, Z) that is an element of the network topology to be determined, and the third network topology element and the switch are matched on condition that the third network topology element and the switch match. By deleting the link between Te, to remove a switch, a link with the node corresponding to the virtual machine.

  With such a configuration, a link between a switch and a node and a link between nodes are created, and then a link between the node corresponding to the virtual machine and the switch is deleted. Therefore, in the network topology, a node corresponding to a physical machine having a link with a switch can be distinguished from a node corresponding to a virtual machine having no link with a switch.

  Further, a service information adding unit that associates a service corresponding to the virtual machine with a node corresponding to the virtual machine in the network topology based on information indicating a correspondence relationship between the virtual machine and the service may be provided.

  Moreover, you may provide the output part which displays the network topology after a link is deleted by the network topology correction | amendment part 5. FIG.

  The present invention is preferably applied to the generation of a network topology including switches, physical machines, and virtual machines.

DESCRIPTION OF SYMBOLS 1 Network topology generation system 2 Element detection part 3 Information acquisition part 4 Network topology generation part 5 Network topology correction | amendment part 6 Service information addition part 7 Authentication information storage part 8 Connection information storage part 9 Element information storage part 10 Service information storage part 11 Output Part

Claims (7)

Obtaining the type of node or switch, the identification information of the interface of the node or switch, and the address assigned to the interface itself from the nodes and switches that are physical machines or virtual machines, For each interface of the switch, obtain the address of the node to which the interface is connected or the interface of the switch from the switch. From the node corresponding to the physical machine, the interface is connected to the interface of the node. Information acquisition means for acquiring the interface address of the node corresponding to the virtual machine,
Create a link between the switch and another node or switch by referring to the address of the node or switch interface to which the switch interface is connected, and change the interface of the node to which the node interface is connected. Network topology generation means for generating a network topology including nodes and switches as elements by creating a link between the nodes with reference to the address and having the connection destination node as an end point ,
A first network topology element, which is an element of a network topology to which a switch interface is connected, is identified, and the interface of the first network topology element is connected based on a link starting from the first network topology element A second network topology element that is an element of the previous network topology is specified, and further, a third network topology element that is an element of the network topology to which the interface of the second network topology element is connected is specified. And deleting the link between the third network topology element and the switch on condition that the third network topology element matches the switch, and a node corresponding to the virtual machine, Remove link for Network topology generation system characterized by comprising a network topology correction means. The network topology according to claim 1, further comprising: service information adding means for associating a service corresponding to the virtual machine with a node corresponding to the virtual machine in the network topology based on information indicating a correspondence relationship between the virtual machine and the service. Generation system. The network topology generation system according to claim 1, further comprising: an output unit that displays a network topology after the link is deleted by the network topology correction unit. Obtaining the type of node or switch, the identification information of the interface of the node or switch, and the address assigned to the interface itself from the nodes and switches that are physical machines or virtual machines, For each interface of the switch, obtain the address of the node to which the interface is connected or the interface of the switch from the switch. From the node corresponding to the physical machine, the interface is connected to the interface of the node. Get the interface address of the node corresponding to the virtual machine
Create a link between the switch and another node or switch by referring to the address of the node or switch interface to which the switch interface is connected, and change the interface of the node to which the node interface is connected. A network topology including nodes and switches as elements is generated by creating a link between nodes, which is a link between nodes, with the node being the connection destination as an end point,
A first network topology element, which is an element of a network topology to which a switch interface is connected, is identified, and the interface of the first network topology element is connected based on a link starting from the first network topology element A second network topology element that is an element of the previous network topology is specified, and further, a third network topology element that is an element of the network topology to which the interface of the second network topology element is connected is specified. And deleting the link between the third network topology element and the switch on condition that the third network topology element matches the switch, and a node corresponding to the virtual machine, Remove link for Network topology generation method characterized by. The network topology generation method according to claim 4, wherein a service corresponding to the virtual machine is associated with a node corresponding to the virtual machine in the network topology based on information indicating a correspondence relationship between the virtual machine and the service. On the computer,
Obtaining the type of node or switch, the identification information of the interface of the node or switch, and the address assigned to the interface itself from the nodes and switches that are physical machines or virtual machines, For each interface of the switch, obtain the address of the node to which the interface is connected or the interface of the switch from the switch. From the node corresponding to the physical machine, the interface is connected to the interface of the node. Information acquisition processing to acquire the interface address of the node corresponding to the virtual machine
Create a link between the switch and another node or switch by referring to the address of the node or switch interface to which the switch interface is connected, and change the interface of the node to which the node interface is connected. A network topology generation process for generating a network topology including nodes and switches as elements by creating a link between nodes, which is a link between the nodes with reference to the address and having the connection destination node as an end point; and,
A first network topology element, which is an element of a network topology to which a switch interface is connected, is identified, and the interface of the first network topology element is connected based on a link starting from the first network topology element A second network topology element that is an element of the previous network topology is specified, and further, a third network topology element that is an element of the network topology to which the interface of the second network topology element is connected is specified. And deleting the link between the third network topology element and the switch on condition that the third network topology element matches the switch, and a node corresponding to the virtual machine, Remove link for Network topology generation program for executing the network topology correction process. On the computer,
The network according to claim 6, wherein a service information addition process for associating a service corresponding to the virtual machine is executed on a node corresponding to the virtual machine in the network topology based on information indicating a correspondence relationship between the virtual machine and the service. Topology generation program.

Images