Generation device and generation method

Abstract

In order to automatically handle failures on a network with reference to the same type of failures in the past, this generation device comprises: a management unit that manages a meta-graph, in which the topology of the network is represented by nodes with labels and a link indicating the relationship between the nodes, and correspondence information indicating the correspondence relationship between the label and an IP address; an extraction unit that extracts, from the meta-graph on the basis of a first handling command corresponding to a first failure on the network, path regulations obtained by abstracting a path from the node as the cause of the first failure to the node at the destination of handling; a search unit that, when a second failure has occurred on the network, searches the meta-graph for a path that is from the node as the cause of the second failure to the node at the destination of handling and matches the path regulations; and a generation unit that generates a second handling command corresponding to the second failure from the first handling command on the basis of the path searched by the search unit and the correspondence information.

Description

Generation device and generation method

[0001] The present invention relates to a generating device and a generating method.

[0002] In network maintenance work, there are devices that support responses to malfunctions that occur in the network. For example, a fault isolation procedure generation device is known that generates a fault isolation procedure that minimizes the expected value of the total work cost estimated to be required to identify a faulty device in the network (see, for example, Patent Document 1).

[0003] JP 2013-141122 A

[0004] There is a demand for automatic countermeasures against a fault occurring on a network by referring to similar faults that have occurred in the past. However, conventional techniques such as those disclosed in Patent Document 1, for example, are unable to automatically counter a fault occurring on a network by referring to similar faults that have occurred in the past.

[0005] The embodiment of the present invention has been made in view of the above-mentioned problems, and makes it possible to automatically deal with a problem that occurs in a network by referring to similar problems that have occurred in the past.

[0006] In order to solve the above problems, a generation device according to an embodiment of the present invention includes a management unit that manages a metagraph representing the topology of a network using labeled nodes and links indicating the relationships between the nodes, and correspondence information indicating the correspondence between the labels and IP addresses; an extraction unit that extracts, based on a first troubleshooting command corresponding to a first fault that has occurred in the network, a path rule from the metagraph that abstracts a path from the node causing the first fault to the node at which the fault is to be dealt with; a search unit that, when a second fault occurs in the network, searches the metagraph for a path from the node causing the second fault to the node at which the path rule applies; and a generation unit that generates a second troubleshooting command corresponding to the second fault from the first troubleshooting command based on the path searched by the search unit and the correspondence information.

[0007] According to the embodiment of the present invention, it becomes possible to automatically deal with a problem that occurs in a network by referring to similar problems that have occurred in the past.

[0008] 1 is a diagram illustrating an example of the configuration of a generation system according to the present embodiment; FIG. 2 is a flowchart illustrating an overall processing flow according to the present embodiment; FIG. 3 is a diagram illustrating a metagraph according to the present embodiment; FIG. 4 is a diagram illustrating an example of correspondence information according to the present embodiment; FIG. 5 is a diagram illustrating an example of a response command according to Example 1; FIG. 6 is a diagram illustrating a processing portion and a variable portion according to Example 1; FIG. 7 is a diagram illustrating extraction of a path rule according to Example 1; FIG. 8 is a diagram illustrating an image when a second malfunction occurs according to Example 1; FIG. 9 is a diagram illustrating a search based on a path rule according to Example 1; FIG. 10 is a diagram illustrating creation of a response command according to Example 1; FIG. 11 is a diagram illustrating creation of a response command according to Example 1; FIG. 12 is a diagram illustrating an example of a response command according to Example 2; FIG. 13 is a diagram illustrating a processing portion and a variable portion according to Example 2; FIG. 14 is a diagram illustrating extraction of a path rule according to Example 2; FIG. 15 is a diagram illustrating a search based on a path rule according to Example 2; FIG. 16 is a diagram illustrating creation of a response command according to Example 2;

[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The embodiment described below is merely an example, and the embodiment to which the present invention is applied is not limited to the following embodiment.

[0010] (Summary) This embodiment relates to the automation of countermeasures against defects in network devices.

[0011] 18 is a diagram showing an example of a network assumed in this embodiment. In the network, a DC (Data Center) serving as a base is connected at L3 (Layer 3) as shown in FIG.

[0012] 18, L3 switches DC1 and DC2 of the DC are connected to L3 switches A1 and A2 of site A, L3 switches B1 and B2 of site B, and L3 switches C1 and C2 of site C via an IF (Interface) 10. Consider a scenario in which measures are required due to degradation of the L3 switches in section 2 between this DC and each site.

[0013] (Issue) Fig. 19 is a diagram for explaining the problem. Because there are a large number of L3 switches, in order to make the response to a failure more efficient, it is conceivable that when a failure similar to a previous failure occurs, the response content from that time can be reused.

[0014] However, failures do not always occur in the same device, but can occur in different devices. Therefore, for example, there may be cases where it is desired to use the countermeasures for the first failure that occurs in L3 switch A1 in Figure 19 for the second failure that occurs in L3 switch B1.

[0015] However, for example, the IP address and IF (Interface) of the troubleshooting command issued when a fault occurs in L3 switch A1 are different from the IP address and IF of the troubleshooting command to be issued when a fault occurs in L3 switch B1. Therefore, for example, in FIG. 19, the setting of IF 10a, which was set as a detour when the first fault occurred, must be carefully replaced with IF 10b, which is set as a detour when the second fault occurs. However, this process requires setting the same type of process (detour) for many IFs, and requires many rewrites. Furthermore, the method of this rewrite (mapping) differs depending on the L3 switch where the first fault occurred and the L3 switch where the second fault occurred, making it difficult for manual implementation.

[0016] Therefore, the generation device according to this embodiment has a configuration, for example, as shown in FIG. 1, so that when a second malfunction occurs, a second handling command corresponding to the second malfunction can be automatically generated based on the handling command entered when a first malfunction occurs.

[0017] <Configuration of Generation System> Fig. 1 is a diagram showing an example of the configuration of a generation system according to this embodiment. In the example of Fig. 1, the generation system 100 includes a generation device 110. However, this is just an example, and the generation system 100 may be configured by multiple devices.

[0018] The generating device 110 is, for example, an information processing device having a computer configuration, or a system including multiple computers. The generating device 110 realizes, for example, each functional configuration as shown in FIG. 1 by the computer included in the generating device 110 executing a predetermined program stored in a storage medium. In the example of FIG. 1, the generating device 110 has each functional configuration, such as a management unit 111, a command reception unit 112, an analysis unit 113, an extraction unit 114, an event reception unit 115, a search unit 116, a generating unit 117, a storage unit 118, and a response command database (DB) 119. Note that at least a portion of each of the above functional configurations may be realized by hardware.

[0019] The management unit 111 executes a management process for storing and managing the metagraph 102 generated based on the network topology information 101 and the correspondence information 103 in the storage unit 118 or the like. The metagraph 102 is a labeled graph that represents the network topology with labeled nodes and rings that indicate the relationships between the nodes. The correspondence information 103 is information (e.g., a correspondence table) that indicates the correspondence between the above-mentioned labels and IP addresses.

[0020] Preferably, management unit 111 generates metagraph 102 and correspondence information 103 from network topology information 101, and stores and manages the created metagraph 102 and correspondence information 103 in storage unit 118, etc. However, without being limited to this, management unit 111 may also acquire, from an external system, metagraph 102 and correspondence information 103 that the external system has created based on the network topology information, and store and manage the acquired metagraph 102 and correspondence information 103 in storage unit 118, etc.

[0021] When a first malfunction occurs in the network, the command reception unit 112 receives a first troubleshooting command corresponding to the first malfunction from an administrator, maintenance technician, etc., and executes a command reception process to store the received troubleshooting command in the troubleshooting command DB 119, etc.

[0022] The analysis unit 113 analyzes the first response command corresponding to the first malfunction that occurred in the network, which was received by the command receiving unit 112, and performs an analysis process that divides the command into a processing part (processing block) and a variable part (IP address, IF, etc.).

[0023] The extraction unit 114 executes an extraction process to extract, from the metagraph 102, a path rule that abstracts a path from the node that causes the first malfunction to the node that is the target of the troubleshooting, based on the first troubleshooting command, and to store the extracted path rule 104 in the storage unit 118, etc. Note that the path rule will be described later.

[0024] The event reception unit 115 executes an event reception process for receiving event information indicating that a second failure has occurred in the network.

[0025] When a second fault occurs in the network, the search unit 116 executes a search process to search the metagraph 102 for a path from the node causing the second fault to the mode to which the path rule 104 extracted by the extraction unit 114 applies, the path from the node causing the second fault to the mode to which the fault is addressed. Here, the following description will be given assuming that the network in which the second fault occurred is the same network as the network in which the first fault occurred. However, this is not limited to this, and the present embodiment can also be applied to a case in which the second fault occurs in another network having a configuration similar to that of the network in which the first fault occurred.

[0026] The generation unit 117 executes a generation process to generate a second response command corresponding to the second malfunction from the first response command, based on the path searched by the search unit 116 and the correspondence information 103. For example, the generation unit 117 determines the variable part of the processing part and variable part analyzed by the analysis unit 113, based on the path searched by the search unit 116 and the correspondence information 103, and generates a second response command (command 105) including the processing part and the determined variable part.

[0027] The storage unit 118 stores various information, data, programs, etc., including, for example, the metagraph 102 and correspondence information 103 managed by the management unit 111, and the path rules 104 extracted by the extraction unit 114, etc.

[0028] <Processing Flow> Next, the processing flow of the generation method according to this embodiment will be described.

[0029] (Overall Processing) Fig. 2 is a flowchart showing the flow of the overall processing according to this embodiment. This processing shows an overview of the overall processing executed by the generating device 110 described with reference to Fig. 1 .

[0030] In step S201, as a pre-processing, the management unit 111 creates the metagraph 102 and the correspondence information from the network topology information 101. Note that the specific processing by the management unit 111 will be described later.

[0031] In step S202, the analysis unit 113 separates the first troubleshooting command corresponding to the first malfunction into a processing portion and a variable portion, and the extraction unit 114 extracts a path rule that abstracts the path from the node that causes the first malfunction to the troubleshooting node. Specific processing details by the analysis unit 113 will be described later using several examples.

[0032] In step S203, when the second malfunction occurs, the search unit 116 searches the metagraph 102 created by the management unit 111 for a path that satisfies the path rule extracted by the extraction unit 114. Note that the specific processing content by the search unit 116 will be described later using multiple examples as examples.

[0033] In step d204, the generation unit 117 determines the value of the variable part analyzed by the analysis unit 113 based on the searched path and the correspondence information 103, and generates a second troubleshooting command that includes a processing part and a variable part and corresponds to the second malfunction. Note that the specific processing content by the generation unit 117 will be described later using several examples as examples.

[0034] By the processing of Figure 2, the generation device 110 can automatically generate a second response command corresponding to a second malfunction when a second malfunction occurs, based on the response command entered when a first malfunction occurs.

[0035] (Pre-processing) Next, a specific example of the pre-processing executed by the management unit 111 will be described.

[0036] 3 is a diagram illustrating a metagraph according to this embodiment. The management unit 111 creates a metagraph 102 that summarizes network topology information 310, for example, as shown in FIG. 3. The example in FIG. 3 shows an example of a metagraph 320 within a dashed line 311 in the topology information 310.

[0037] 3, metagraph 320 includes nodes 321 representing the device itself, such as an L3 switch, nodes 322 representing the interfaces of the device, rings 323 representing the logical connection relationships between the nodes, and labeled links 324 representing the redundant relationships between the nodes. The links are also called edges.

[0038] A metagraph is a labeled graph. In a metagraph, each node is labeled with a "device name" or an "interface name of the device." By using a naming convention that adds a predetermined prefix to labels, such as "N: device name" or "P: Exx," it is possible to easily distinguish whether a node represents the device itself or an interface.

[0039] Links (or edges) between devices that are configured for redundancy are labeled r, and other links do not require labels (no marking). However, unmarked links may also be labeled, for example, with Null, and all links may be managed as labeled links.

[0040] In addition, the management unit 111 creates correspondence information 103, separate from the metagraph 102, which indicates the correspondence between the device name and the management IP address, as shown in Figure 4, for example, and stores and manages the created correspondence information 103 in the memory unit 118, etc.

[0041] Preferably, the management unit 111 automatically generates the metagraph 102 as a pre-processing step. The metagraph 102 uses the IP addresses, interfaces, and redundancy relationships of devices such as L3 switches. This information can be easily obtained, for example, by a command equivalent to "show" obtained from a device using NETCONF (Network Configuration Protocol). For example, IP addresses have a format such as xxx.xxx.xxx.xxx, and interfaces have a predetermined format such as FastEthernet x / xx, and can be easily obtained using regular expressions, etc. Furthermore, redundancy relationships can be easily obtained by analyzing the show results to determine whether VRRP (Virtual Router Redundancy Protocol) or the like is running.

[0042] [Example 1] (Example of a Troubleshooting Command) Fig. 5 is a diagram showing an example of a troubleshooting command according to Example 1. For example, in a network 510 as shown in Fig. 5, a malfunction due to deterioration occurs in the L3 switch A1, and an administrator or maintenance person manually executes a troubleshooting command 520 as shown in Fig. 5. Note that the troubleshooting command 520 is abbreviated here by referring to an actual command but omitting parts that do not affect the system.

[0043] 5, the contents of the countermeasures performed by the countermeasure command 520 are as follows: first, in order to bypass the faulty L3 switch A1, the Open Shortest Path First (OSPF) costs of the IFs connected to the L3 switch A1 are reduced in the L3 switches DC1, DC2, A11, A12, and A2; second, the L3 switch A1 is restarted, and the OSPF costs of the IFs that were bypassed are increased.

[0044] (Analysis Process) Fig. 6 is a diagram for explaining a processing part and a variable part according to the first embodiment. The analysis unit 113 executes an analysis process that analyzes the troubleshooting command 520 and divides it into a processing part (processing block) and a variable part (IP address, IF, etc.). This IP address or IF is the node to be dealt with. In the first embodiment, the processing block is defined as a "device unit."

[0045] 6 shows the "detouring" portion of the troubleshooting command 520 in Fig. 5. In the troubleshooting command 600, the processing blocks 601 and 602 have the same hash value obtained by hashing the portions other than the IP address and IF name, so the analysis unit 113 can determine that DC1 and DC2 are symmetrical on the metagraph.

[0046] Similarly, in the troubleshooting command 600, since the processing blocks 603 and 604 have the same hash value obtained by hashing the parts other than the IP address and IF name, the analysis unit 113 can determine that A11 and A12 are symmetrical on the metagraph.

[0047] 7 is a diagram illustrating extraction of a path rule according to Example 1. The extraction unit 114 refers to the metagraph 102 and extracts a path rule from a node causing a malfunction to a node to be addressed.

[0048] As an illustrative example, assume that the node causing the malfunction is A1 in a metagraph 700 such as that shown in FIG. 7. In this case, the extraction unit 114 identifies the nodes DC1, DC2, A11, and A12 to be dealt with from, for example, the management IP addresses described in the troubleshooting command 600 such as that shown in FIG. 6. The extraction unit 114 also lists paths 701 from node A1 causing the malfunction to the nodes DC1, DC2, A11, and A12 to be dealt with. Note that the extraction unit 114 distinguishes between unmarked links and labeled links, and therefore does not extract a path from node A1 to node A2.

[0049] The extraction unit 114 also abstracts (commonizes) the listed paths 701 to extract path rules 104. As shown in Fig. 7 , the path rules 104 are paths described by node types (device nodes, IF nodes, etc.) without including individual names such as device names and IF names. In the example of Fig. 7 , the start point 703 of the path rule 104 indicates the node that causes the malfunction, and the end point 704 indicates the node where the malfunction is to be addressed. The extraction unit 114 extracts such path rules 104 and stores the extracted path rules 104 in the storage unit 118, etc.

[0050] (Second Fault) FIG. 8 is a diagram illustrating an example of the occurrence of a second fault according to the first embodiment. The second fault occurs after the first fault occurs. The second fault may be the same as the first fault, or may be a different fault. In the following description, the second fault occurs when degradation occurs in the L3 switch B1, requiring an action (restart).

[0051] 9 is a diagram illustrating a search based on a path rule according to Example 1. When a second malfunction occurs in the network, the search unit 116 executes a search process to search for a path from the metagraph 102, to which the path rule 104 applies, from a node that is the cause of the second malfunction, to a mode to which the malfunction is to be dealt with.

[0052] For example, the search unit 116 performs pattern matching using a query in which node B1, which is the cause of the second fault, is substituted for the start point 702 of the path rule 104 in a metagraph 910 (an excerpt from the metagraph 102) as shown in FIG. 9, and lists applicable paths. In the example of FIG. 9, five paths 920 are extracted. This allows the search unit 116 to identify the device to be addressed and its IF. The search unit 116 can also identify the device to be addressed and the IP address of its IF from the correspondence information 930 (an excerpt from the correspondence information 103).

[0053] 10 and 11 are diagrams for explaining the creation of a second troubleshooting command according to Example 1. The generator 117 executes a generation process to generate a second troubleshooting command corresponding to a second malfunction from the first troubleshooting command based on the path searched by the searcher 116 and the correspondence information 103.

[0054] Here, it is assumed that a second troubleshooting command 1020 corresponding to a second malfunction is generated from a first troubleshooting command 1010 as shown in Fig. 10. It is also assumed that the node causing the malfunction is B1, and that the troubleshooting destination nodes are DC1, DC2, B11, B12, and B13, based on the path 920 searched in Fig. 9.

[0055] First, the generation unit 117 searches for a processing block with the same management IP address in the first troubleshooting command 1010 among the troubleshooting destination nodes DC1, DC2, B11, B12, and B13. In the example of Fig. 10, processing blocks 1011 and 1012 are found to be relevant, so the generation unit 117 uses processing blocks 1011 and 1012 to create processing blocks 1021 and 1022 that correspond to the second malfunction. At this time, the generation unit 117 rewrites the troubleshooting destination IFs of processing blocks 1021 and 1022 by referring to the path 920 extracted in Fig. 9.

[0056] 11 , the generation unit 117 generates processing blocks 1023, 1024, and 1025 corresponding to the second fault using processing blocks 1013 and 1014 whose variable parts are not the same. For example, the generation unit 117 generates processing blocks 1023, 1024, and 1025 corresponding to the second fault by substituting the variable parts (IP address, IF) of nodes B11, B12, and B13 to be dealt with into the variable parts of processing blocks 1013 and 1014. Note that although the number of processing blocks 1013 and 1014 when the first fault occurred is different from the number of processing blocks 1023, 1024, and 1025 corresponding to the second fault, they can be duplicated because they are the same processing.

[0057] In this way, the generation unit 117 can generate a second troubleshooting command 1020 corresponding to the second defect from the first troubleshooting command 1010 based on the path searched by the search unit 116 and the correspondence information 103.

[0058] [Example 2] (Example of Troubleshooting Command) Fig. 12 is a diagram showing an example of a troubleshooting command according to Example 2. For example, in a network 1210 as shown in Fig. 12, a malfunction due to deterioration occurs in the L3 switch A1, and an administrator or maintenance person manually executes a troubleshooting command 1220 as shown in Fig. 12. Note that the troubleshooting command 1220 is abbreviated here by referring to an actual command but omitting parts that do not affect the system.

[0059] 12, the contents of the countermeasures performed by the countermeasure command 1220 are as follows: first, the cost of each IF of the L3 switch A2 is reduced to bypass the faulty L3 switch A1, and then the L3 switch A1 is restarted and the OSPF cost of each IF of the bypassed L3 switch A2 is increased.

[0060] (Analysis Process) FIG. 13 is a diagram for explaining a processing part and a variable part according to the second embodiment. The analysis unit 113 executes an analysis process that analyzes the handling command 1220 and divides it into a processing part (processing block) and a variable part (IP address, IF, etc.). This IP address or IF is the node to be handled. In the second embodiment, the processing block is defined as an "IF unit." Also, similar to the first embodiment, processing blocks that are the same are determined as such.

[0061] The troubleshooting command 1300 in Fig. 13 shows the "detouring" portion of the troubleshooting command 1220 in Fig. 12. In this example, the analysis unit 113 separates the command into four processing portions, one IP address, and four variable portions of IF.

[0062] 14 is a diagram illustrating extraction of a path rule according to Example 2. The extraction unit 114 refers to the metagraph 102 and extracts a path rule from a node causing a malfunction to a node to be addressed.

[0063] As an example for explanation, assume that the node causing the problem is A1 in the metagraph 1400 shown in Fig. 14. In this case, the extraction unit 114 lists paths 1401 to E18, E19, E20, and E21, which are interfaces of the management IP address described in the troubleshooting command 1300 shown in Fig. 13.

[0064] The listing conditions for cases including Examples 1 and 2 are that "one path contains only one treatment target device" and "one path contains one treatment target device and one IF."

[0065] The extraction unit 114 also abstracts (commonizes) the listed paths 1401 to extract path rules 104. As shown in FIG. 14 , the path rules 104 are paths described by node types (device nodes, IF nodes, etc.) without including individual names such as device names and IF names. The path rules 104 indicate that the start point 1402 is the cause and the end point 1403 is the solution. In the example of FIG. 14 , the path rule 104 indicates that "the solution is the third hop from the start point 1402." The extraction unit 114 extracts such path rules 104 and stores the extracted path rules 104 in the storage unit 118, etc.

[0066] 15 is a diagram for explaining a search based on a path rule according to Example 2. Here, the following description will be given assuming that the second malfunction occurs in the L3 switch B1, and that an action (restart) is required.

[0067] When a second fault occurs in the network, the search unit 116 executes a search process to search for a path from the metagraph 102, to which the path rule 104 applies, from the node that is the cause of the second fault, to the mode to be dealt with.

[0068] For example, the search unit 116 performs pattern matching using a query in which the node B1 that is the cause of the second fault is substituted for the starting point 702 of the path rule 104 in the metagraph 1400 (an excerpt of the metagraph 102) as shown in Fig. 15, and lists applicable paths. In the example of Fig. 15, five paths 1501 are extracted. The device node closest to the end of each path 1501 corresponds to the current problem-target device, and the end of each path 1501 becomes the IF of the problem-target device.

[0069] 15, it can be seen from five paths 1501 that the device to be handled is B2 and its interfaces are E22 to E26. The searching unit 116 can also identify the IP address of the device to be handled, B2, from the correspondence information 1502 (an excerpt from the correspondence information 103).

[0070] 16 is a diagram illustrating the creation of a second troubleshooting command according to Example 2. The generator 117 determines the variable part from the processing part and variable part analyzed by the analyzer 113 based on the path searched by the searcher 116 and the correspondence information 103, and generates a second troubleshooting command 1620 corresponding to the second malfunction, which includes the processing part and the determined variable part.

[0071] 16 , a second troubleshooting command 1620 corresponding to a second fault is generated from a first troubleshooting command 1610. In the second embodiment, unlike the first embodiment, the troubleshooting node B2 is not included in the first troubleshooting command 1610. In this case, the generation unit 117 generates the second troubleshooting command 1620 by substituting the IP addresses of the troubleshooting node B2 and the IFs of the troubleshooting node B2 for the five paths 1501 searched for by the search unit 116 into the variable parts (IP addresses, IFs) of the first troubleshooting command 1610.

[0072] In Example 2, unlike Example 1, the first troubleshooting command 1610 and the second troubleshooting command 1620 have different numbers of IFs, and the generation unit 117 generates (duplicates) processing blocks on an IF basis and rewrites the IF names.

[0073] In this way, the generation unit 117 can generate a second troubleshooting command 1020 corresponding to the second defect from the first troubleshooting command 1010 based on the path searched by the search unit 116 and the correspondence information 103.

[0074] 12 sets a preset setting value (e.g., OSPF cost) when returning from a detour. However, this is just one example, and the troubleshooting command 1200 may read and save the current setting value of the L3 switch before executing the troubleshooting command 1200, and use the saved setting value when returning from a detour.

[0075] <Hardware Configuration> The generating device 110 has, for example, the hardware configuration of a computer 1700 as shown in Fig. 17. Alternatively, the generating device 110 is realized by a plurality of computers 1700.

[0076] 17 is a diagram showing an example of the hardware configuration of a computer 1700. The computer 1700 includes, for example, a processor 1701, a memory 1702, a storage device 1703, a communication device 1704, an input device 1705, an output device 1706, and a bus B.

[0077] The processor 1701 is, for example, an arithmetic unit such as a CPU (Central Processing Unit) that executes predetermined programs to realize various functions. The memory 1702 is a storage medium readable by the computer 1700, and includes, for example, a RAM (Random Access Memory) and a ROM (Read Only Memory). The storage device 1703 is a computer-readable storage medium, and may include, for example, a HDD (Hard Disk Drive), an SSD (Solid State Drive), various optical disks, and magneto-optical disks.

[0078] The communication device 1704 includes one or more pieces of hardware (communication devices) for communicating with other devices via a wireless or wired network. The input device 1705 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that accepts input from the outside. The output device 1706 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that outputs to the outside.

[0079] The bus B is commonly connected to the above components and transmits, for example, address signals, data signals, and various control signals. The processor 1701 is not limited to a CPU, and may be, for example, a DSP (Digital Signal Processor), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array).

[0080] (Supplementary Note) The generation device 110 in this embodiment is not limited to being realized by a dedicated device, but may also be realized by a general-purpose computer. In this case, a program for realizing this function may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be read into a computer system and executed. Note that the term "computer system" here includes hardware such as an OS and peripheral devices.

[0081] Furthermore, "computer-readable recording media" includes various storage devices such as portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, as well as hard disks built into computer systems. Furthermore, "computer-readable recording media" may also include devices that dynamically store programs for a short period of time, such as communication lines when transmitting programs via networks such as the Internet or communication lines such as telephone lines, and devices that store programs for a certain period of time, such as volatile memory within computer systems that serve as servers or clients in such cases.

[0082] Furthermore, the above program may be one that realizes part of the above-mentioned functions, or may be one that can realize the above-mentioned functions in combination with a program already recorded in a computer system, or may be one that is realized using hardware such as a PLD (Programmable Logic Device) or FPGA (Field Programmable Gate Array).

[0083] <Effects of the embodiment> According to the present embodiment, it becomes possible to automatically deal with a problem that occurs in a network by referring to similar problems that have occurred in the past.

[0084] For example, in this embodiment, if a similar event has occurred in the past on a device of the same type (e.g., an L3 switch on the base side), it is possible to reuse that to automatically generate and execute a response command. In particular, when there are a large number of devices or interfaces, adapting (adapting) requires modifying many variable parts, which is cumbersome and complicated. However, this embodiment makes it possible to automatically change these parts, which shortens the time required for failure recovery processing, reduces errors, and reduces operation.

[0085] In this embodiment, the only information that needs to be stored as knowledge is a metagraph and a small table, which requires little storage capacity and is easy to maintain. Both can also be generated automatically. For example, raw topology information contains a large amount of information.

[0086] Furthermore, this embodiment does not require a large amount of case data, and can be applied, for example, even in the early stages of network operation when a large amount of case data is not collected and machine learning cannot be applied.

[0087] (Regarding scalability) In the above-described embodiment, OSPF is used for the DC and the L3 switch on the base side, but this embodiment is not limited to the L3 switch and can be applied to various devices that make up a network, such as a router or a server. This embodiment can also be applied to routing protocols other than OSPF.

[0088] Furthermore, in the above-described embodiment, the cause device and the detouring device are adjacent to each other, but the present invention can also be applied to use cases in which the cause location and the handling location are separated by several hops or more.

[0089] Furthermore, in the above-described embodiment, the IP addresses are separated from the graph and managed separately as a table, which makes maintenance easier.

[0090] Furthermore, by abstracting the network and the relationship between causes and solutions as a graph and inferring from it, it becomes possible to generate it for various networks not limited to this embodiment, and it has high scalability to accommodate other use cases in the future.

[0091] Summary of Embodiments This specification discloses at least the following generating device, generating method, generating system, and program: (Item 1) A generating device comprising: a management unit that manages a metagraph representing a network topology with labeled nodes and links indicating the relationships between the nodes, and correspondence information indicating the correspondence between the labels and IP addresses; an extraction unit that extracts, based on a first troubleshooting command corresponding to a first fault that has occurred in the network, from the metagraph a path rule that abstracts a path from a node that causes the first fault to a troubleshooting node; a search unit that, when a second fault has occurred in the network, searches from the metagraph for a path from the node that causes the second fault to a troubleshooting node, to which the path rule applies; and a generation unit that generates, from the first troubleshooting command, a second troubleshooting command corresponding to the second fault, based on the path searched by the search unit and the correspondence information. (2) The generating device according to paragraph 1, further comprising an analysis unit that analyzes a first troubleshooting command corresponding to the first fault and separates it into a processing portion and a variable portion, and the generation unit determines the variable portion based on the path searched by the search unit and the correspondence information, and generates a second troubleshooting command corresponding to the second fault, the second troubleshooting command including the processing portion and the determined variable portion. (3) The generating device according to paragraph 1 or 2, further comprising: the management unit that creates the metagraph and the correspondence information from topology information of the network.(Clause 4) A generation method in which a computer executes the following processes: a process for managing a metagraph representing a network topology with labeled nodes and links indicating the relationships between the nodes, and correspondence information indicating the correspondence between the labels and IP addresses, an extraction process for extracting, based on a first troubleshooting command corresponding to a first fault that has occurred in the network, from the metagraph a path rule that abstracts a path from the node that causes the first fault to a troubleshooting node, a search process for, when a second fault has occurred in the network, searching from the metagraph for a path from the node that causes the second fault to a troubleshooting node to which the path rule applies, and a generation unit for generating, from the first troubleshooting command, a second troubleshooting command corresponding to the second fault, based on the path searched in the search process and the correspondence information. (Clause 5) A program, or a storage medium having a program stored thereon, that causes a computer to execute the generation method described in Clause 4. (Clause 6) A generation system comprising: a management unit that manages a metagraph that represents the topology of a network with labeled nodes and links that indicate the relationships between the nodes, and correspondence information that indicates the correspondence between the labels and IP addresses; an extraction unit that extracts, based on a first troubleshooting command corresponding to a first fault that has occurred in the network, a path rule that abstracts a path from the node that causes the first fault to the node that is to be dealt with; a search unit that, when a second fault has occurred in the network, searches from the metagraph for a path from the node that causes the second fault to the node that is to be dealt with, to which the path rule applies; and a generation unit that generates a second troubleshooting command corresponding to the second fault from the first troubleshooting command based on the path searched by the search unit and the correspondence information.

[0092] Although the present embodiment has been described above, the present invention is not limited to such a specific embodiment, and various modifications and changes are possible within the scope of the gist of the present invention described in the claims.

[0093] REFERENCE SIGNS LIST 100 Generation system 102 Metagraph 103 Correspondence information 104 Path rule 110 Generation device 111 Management unit 113 Analysis unit 114 Extraction unit 116 Search unit 117 Generation unit 118 Storage unit 1700 Computer

Claims

1. A generation device comprising: a management unit that manages a metagraph that represents the topology of a network using labeled nodes and links that indicate the relationships between the nodes, and correspondence information that indicates the correspondence between the labels and IP addresses; an extraction unit that extracts, based on a first troubleshooting command corresponding to a first fault that has occurred in the network, a path rule that abstracts a path from the node that causes the first fault to the node that is to be dealt with; a search unit that, when a second fault occurs in the network, searches from the metagraph for a path from the node that causes the second fault to the node that is to be dealt with, to which the path rule applies; and a generation unit that generates a second troubleshooting command corresponding to the second fault from the first troubleshooting command based on the path searched by the search unit and the correspondence information.

2. A generating device as described in claim 1, having an analysis unit that analyzes a first troubleshooting command corresponding to the first defect and separates it into a processing part and a variable part, and the generation unit determines the variable part based on the path searched by the search unit and the correspondence information, and generates a second troubleshooting command corresponding to the second defect, which includes the processing part and the determined variable part.

3. The generating device according to claim 1 or 2, wherein the management unit creates the metagraph and the correspondence information from topology information of the network.

4. A generation method in which a computer executes the following steps: a process for managing a metagraph that represents the topology of a network with labeled nodes and links that indicate the relationships between the nodes, and correspondence information that indicates the correspondence between the labels and IP addresses; an extraction process for extracting, from the metagraph, a path rule that abstracts a path from the node that causes a first fault to a node that is to be dealt with, based on a first troubleshooting command that corresponds to a first fault that has occurred in the network; a search process for searching, when a second fault has occurred in the network, for a path from the node that causes the second fault to a node that is to be dealt with, to which the path rule applies, from the metagraph; and a generation unit that generates, from the first troubleshooting command, a second troubleshooting command that corresponds to the second fault, based on the path searched in the search process and the correspondence information.

Images