JP2011090429A - Integrated monitoring system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology for achieving early detection of a problem, specification of a problem range when the number of data centers (DCs) of an objective system is two or more in a technology for operation management (including monitoring) of a system. <P>SOLUTION: An integrated monitoring system includes a first processing unit (monitoring server 3 in DC and integrated monitoring server 5) for monitoring the inside state of each of the plurality of data centers 2 constituting the objective system such as a site, a second processing unit (service monitoring server 4) for monitoring the access state to the service of the site from a user terminal 1, and a third processing unit (integrated monitoring server 5) for specifying the problem part or problem range of the plurality of DCs 2 of the objective system and for output of the information by comparing first monitoring information by the first processing unit and second monitoring information by the second processing unit and performing decision. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to technologies such as cloud computing and the Internet (Web), and more particularly to a technology for operation management and monitoring of a data center (hereinafter abbreviated as “DC” where appropriate).

  On the Internet, there are many information processing systems that provide services, contents, and the like to websites of companies and the like, that is, computer terminals of users (customers, etc.), but these sites are often composed of data centers. ing. The number of data center operators is also increasing. In the data center, various services and functions such as a high-speed communication connection environment, various security / disaster countermeasures, operation management, and the like are provided, and these differ depending on individual data centers.

  It is related to the above-mentioned data center and system operation management (including monitoring) technology. Currently, the monitoring technology in one data center is generalized. In this technique, examples of monitored devices (hardware and software) include servers and networks (network devices). Monitored items include logs, resources, errors / failures, turnaround time (TAT), and the like. In this technique, the monitoring target item of the monitoring target device can be monitored from the monitoring unit (such as a monitoring server) in the data center, and monitoring information can be recorded and output to the data center manager or the like. .

  On the other hand, the data center that is managed or configured by one company or one site now or in the future due to the appearance and development of technologies such as SaaS (Software as a Service) and cloud computing (hereinafter abbreviated as “cloud” as appropriate). There are many. For example, as a data center that constitutes one company site in a cloud environment, there is a system (FIG. 1, site α (101)) including a plurality of data centers including its own data center and another company's data center. In other words, in order to realize site services and contents, service processing and data of a plurality of data centers are used in parallel or in cooperation.

  As described above, when there are a plurality of data centers to be managed or configured in the system of one company or one site, the conventional technology such as monitoring in units of one data center performs operation management (monitoring) for the entire system. Including) is insufficient. In particular, it is difficult to quickly and quickly find a problem such as an error or failure in the entire target system and specify (isolate) the location and range (for example, data center unit) of the problem. The above becomes more difficult as the number of data centers constituting the target system increases with the cloud environment.

  In view of the above, the main object of the present invention relates to the operation management (including monitoring) of a system such as a Web site or a data center. When there are a plurality of data centers of the target system, early detection of problems and problems It is to provide a technology that can realize the range specification.

  Of the inventions disclosed in the present application, the outline of typical ones will be briefly described as follows.

  (1) In the system (integrated monitoring system) of the present invention, as a first means, a system monitoring unit that performs processing for monitoring each state of a plurality of data centers constituting a target system such as a site (service), As a second means, there is provided a service monitoring unit that performs processing for monitoring the state of access to a target system such as a site (service) by a user terminal (group) via a communication network, the state of a service, and the like.

  The first means, for example, monitors the status of each of the plurality of data centers by receiving or acquiring the DC monitoring information from each data center (for example, the DC monitoring unit). Further, the first means compares the monitoring information of each data center or the status indicated by it (for example, the presence or absence of an error or failure (NG / OK)) to determine the presence or absence of the problem in the target system and its location or range. (For example, data center unit) may be determined and the result information may be output. As a result, the manager of the data center corresponding to the location or range of the problem can recognize the problem and take action.

  (2) Further, the second means, for example, manages and aggregates the access status and service status in units of services (units such as associated URL units or routes to data centers). Monitor. Then, the second means transmits the monitoring information or the state indicated by it (presence / absence of error or failure (NG / OK)) to the first means (the first means is from the second means). That information may be obtained).

  Then, the first means uses the monitoring information / status of each of the plurality of data centers by the first means and the monitoring information / status by the second means, and compares them to determine the problem in the target system. The presence or absence, the location or range (for example, data center unit) is determined, and the result information is output. As a result, the manager of the data center corresponding to the location or range of the problem can recognize the problem and take action.

  The form of this system is, for example, an information processing system that monitors a target system having a plurality of data centers that constitute a site that provides services or contents to user terminals in a cloud environment or the like, and each of the plurality of data centers A first processing unit that monitors the internal state of the first processing unit, a second processing unit that monitors the status of access to the site service from the user terminal, first monitoring information by the first processing unit, By using the second monitoring information by the processing unit and comparing them, it is possible to identify a problematic part or a range including the data center among a plurality of data centers of the target system and output the information And a third processing unit.

  The effects obtained by typical ones of the inventions disclosed in the present application will be briefly described as follows. According to the present invention, it is related to the operation management (including monitoring) of a system such as a website or a data center, and when a target system has a plurality of data centers, early detection of a problem, specification of a problem range, and the like are realized. it can.

As a premise, it is a figure which shows the structural example of a correspondence relationship, such as a company, a site, and a data center. 1 is a diagram illustrating a configuration of a system (an integrated monitoring system and an information processing system including the system) according to an embodiment of the present invention. In the system of this Embodiment, it is a figure shown about the case where the form of a target system (other company DC) is HaS, and the case of provision of monitoring information as a 1st specific example. In the system of this Embodiment, it is a figure shown about the case where the form of an object system (other company DC) is SaaS and the case where monitoring information is not provided as a 2nd specific example. It is a figure shown about the case where communication between the data centers of an object system exists as a 3rd specific example in the system of this Embodiment. It is a figure for demonstrating the other specific example in the system of this Embodiment.

  Hereinafter, an embodiment (integrated monitoring system) of the present invention will be described in detail based on the drawings (FIGS. 1 to 6). Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted.

<Target system>
In FIG. 1, in order to confirm the premise of the present embodiment, a configuration example of a correspondence relationship between target systems by companies, sites, DCs, and the like will be briefly described. Examples of a plurality of DCs include DC-A1, B1, C1,..., Etc. as DCs operated and managed for each company (DC operator) X, Y, Z,. There may be a plurality of DCs operated and managed by one company. For example, DC-A1, A2,... The configuration of each DC may be individually different or similar. In addition, the presence / absence and details of the DC monitoring technique for each DC may be different or similar.

  And it has the site (Web site) which a company etc. provide using the above-mentioned DC (its service and function). For example, as a first case, the site α (101) is configured by a plurality of DCs including three of the company X, DC-A1, the other company Y, DC-B1, and the other company Z, DC-C1. By using technologies such as SaaS and cloud among a plurality of DCs, site services and contents are realized. In addition, for example, the main DC (for example, the backbone system) constituting the site α by the company X can be the DC-A1 of the company X, and the DC that is parallel or linked to it can be the DC-B1 of the other company Y. is there.

  As a second case, the site β (102) is constituted by DC-A1, A2,... As a third case, there is of course a case where a site is configured by one DC. For example, the site γ (103a) by the company X is configured by the DC-A1 of the company X. Alternatively, the site γ (103b) by own company X is constituted by DC-B1 of other company Y.

  In FIG. 1, for example, the site α is illustrated so as to include a plurality of DCs. However, the site α may be configured by using all the services and functions of each DC, and the services of each DC. Or a combination of some of the functions and the like.

  In the case of operation management for systems such as sites α and β (when multiple DCs make up one site or when multiple DCs are managed by one company), monitoring within individual DCs However, integrated monitoring of the entire system (a plurality of DCs) has not been realized conventionally.

  For example, when an error or the like occurs in providing the service of the site α (service by cloud), the target system as a whole is not conventionally monitored, so which DC (company) in particular is related to each related DC (company). It was insufficient to identify whether there was a problem.

  In addition, for example, there are DCs that do not provide (disclose) the information to the outside even if monitoring in the DC according to individual DCs. In this case, even if a DC problem (for example, server failure) related to the service error or the like can be detected in the corresponding DC, the other (external) related DC (for example, a DC linked with another company) As long as there is no provision of information or notification of a problem from the corresponding DC, the problem cannot be detected.

<Overview>
The outline and features of the present embodiment will be described below with reference to FIG. As a characteristic element, the integrated monitoring system includes an integrated monitoring server 5 (first means) and a service monitoring server 4 (second means). The service monitoring server 4 and the integrated monitoring server 5 are provided on the Internet 90, for example. As a target system, for example, as a plurality of DCs 2 constituting the site α (corresponding to 101 in FIG. 1), DC-A of own company X that is the first DC and DC-B of other company Y that is the second DC , Is included. The user terminal 1 (group) accesses the service of the site α (DC that realizes it) via the Internet 90. In FIG. 2, examples of flows and routes such as solid line arrows for normal access, broken line arrows for monitoring (K0 (K1) and K2), two-dot chain line arrows for monitoring result notification (K1 information, K2 information), etc. Indicates. As types of monitoring, K0 is DC internal monitoring, K1 is system monitoring (monitoring a plurality of DCs) using K0, and K2 is service monitoring.

  The integrated monitoring server 5 monitors (K1) the target system using a plurality of DC2. In this monitoring (K1), the integrated monitoring server 5 receives monitoring result information by the in-DC monitoring (K0) by each in-DC monitoring server 3 (3A, 3B) (example: K1A, K1B). Each in-DC monitoring server 3 monitors (K0) the status of the network device 21 and server 22 in the DC, records the monitoring information in a log, etc., and transmits it to the external integrated monitoring server 5.

  On the other hand, the service monitoring server 4 monitors (K2) the state of the user terminal 1 (group) due to access to the service of the site α (K2A, K2B). The service monitoring server 4 notifies (transmits) the K2 result information to the integrated monitoring server 5 (K2S).

  The integrated monitoring server 5 receives information (for example, K1A, K1B, K2S) based on each of the monitoring (K1, K2) by the corresponding monitoring result receiving unit 51 (51A, 51B, 51S). The monitoring result receiving unit 51 can also be called a system monitoring unit.

  Then, the monitoring result comparison unit 52 compares the information of (K1, K2) and makes a comprehensive judgment, so that the location of the problem related to the target system and the service (for example, DC failure related to service error, etc.) or The range including it is detected and specified, and the information is output. For example, the problem part is specified in server units (hardware unit / software unit) in DC, or the problem range including the problem part is specified in DC unit. The monitoring result comparison unit 52 can be rephrased as a comprehensive determination unit or the like.

  Further, the integrated monitoring server 5 may include a not-shown notification unit. Based on the comprehensive determination result, the notification unit notifies the DC manager corresponding to the problem range, the manager of the related DC, and the like. May be. As a result, the manager or the like of the corresponding DC can immediately cope with the problem.

<Integrated monitoring system>
Details will be described below. In FIG. 2, a user terminal 1 is a PC or a mobile phone equipped with a Web browser or the like, and accesses a site α (service) on the Internet 90 based on an operation by a user. For example, URLs (one or more) of access destinations are associated with each other according to the service. At this time, DC2 to be accessed is determined according to the target service (URL) and the like. For example, there is a case of DC-A (for example, route RA) and a case of DC-B (for example, route RB).

  Each DC 2 includes a plurality of network devices 21 and servers 22. The DC 2 has a monitoring server 3 (DC monitoring server) which is a DC monitoring unit, depending on the individual DC 2, and has predetermined monitoring target devices (21, 22, etc.) and monitoring target items (log, Predetermined monitoring (K0) processing by resources, errors / failures / security, TAT, etc.) is performed.

  In addition to the Internet 90, there may be a communication means such as a dedicated line / VPN 91 for connecting the DCs 2 between the plurality of DCs 2. For example, between DC-A and DC-B, communication is performed via a dedicated line / VPN 91 (for example, route RC) at the time of cooperation of service processing by the cloud.

  The site α provides, for example, cloud services and content. The service by cloud is composed of, for example, a combination of service processing by DC-A, service processing by DC-B, and service processing by other DCs.

<DC form>
As a form of DC2, a form such as Haas (Hardware as a Service), PaaS (Platform as a Service), and SaaS can be used according to a known concept. For example, in the case of a HaAS form (hardware rental), hardware such as the server 22 in the DC is provided to the site operator as a unit. In the case of the SaaS form (software lending), software operating on hardware such as the server 22 in the DC and its service processing are provided to the site operator as a unit.

  In this system, according to the DC form, detailed forms such as in-DC monitoring (K0), system monitoring (K1), service monitoring (K2), etc. (for example, detailed contents and presence / absence of monitoring information, The status and information that can be grasped and judged are different. The level of information provided from the DC 2 to the integrated monitoring server 5 is also different. For example, in the case of HaS, problems such as errors can be grasped for each server 22 device.

<System monitoring (K1)>
The integrated monitoring server 5 performs system monitoring (K1) processing by receiving the K0 result information from the monitoring server 3 (3A, 3B) in each DC2 by the monitoring result receiving unit 51 (51A, 51B) (example) : K1A, K1B). In the integrated monitoring server 5, even if the monitoring server 3 does not provide (disclose) K0 result information according to individual DC2, the information is not provided (the state of the DC is unknown). to manage.

  The monitoring result comparison unit 52 compares the information of the reception result by each monitoring result receiving unit 51 (51A, 51B, 51S), that is, the state of each DC2 (DC-A, DC-B) and the service state. Make a comprehensive decision, identify the problem area, and output the information. The information indicating the state (monitoring result) has at least two types, for example, OK (no problem) / NG (problem). The method of comprehensive judgment will be described later.

<Service monitoring (K2)>
The service monitoring server 4 performs service monitoring (K2) processing by, for example, periodically accessing the target site α (its configuration DC2) (eg, K2A, K2B).

  In the service monitoring (K2) process in the service monitoring server 4, at least one or more DCs 2 that may be access destinations related to the service are grasped. If it is known, it is possible to specify a problem range or the like at the time of the comprehensive judgment processing in the integrated monitoring server 5 even if the actual access destination DC2 cannot be specified as one.

  In addition, the service monitoring server 4 manages the association information between the URL of access to the service of the site α from the user terminal 1 (group) and the DC 2 (the network device 21 or the like) as the access destination. May be. In particular, information regarding the access route R (eg, RA, RB) or the address of the access destination DC (eg, DC-A, DC-B) may be managed. This management information can be grasped by, for example, registering in the service monitoring server 4 in advance using information defined on the site (target system) side. When information such as URLs and routes is managed in this way, it is possible to specify the problem range in more detail during the comprehensive determination process in the integrated monitoring server 5.

  In addition, when a plurality of DCs 2 as access destination selection candidates are associated with one service (URL) (dynamic access destination form), for example, information on the plurality of selection candidates is managed as it is. Can be grasped. Further, when the management information is not limited to DC units but can be grasped in more detail in units of the server 22 or the like in the DC, the information can be managed in the same manner as described above, thereby enabling more detailed monitoring and judgment. Become.

  Based on the above, the service monitoring server 4 performs service monitoring (K2) by the service monitoring unit 41. For example, the number of errors related to service responses is tabulated for each service URL. Then, the service monitoring server 4 notifies the integrated monitoring server 5 (monitoring result receiving unit 51S) of the K2 result information by the monitoring result notification unit 42, for example, when an error has occurred or at a regular timing. (K2S).

<Specific example 1>
An example of integrated monitoring in the case where the form of DC2 is HaAS will be described as a first specific example with reference to FIG. DC-B in FIG. 3 is in the form of HaS, and is a DC (cloud) that performs cooperation and the like for services by the cloud with respect to DC-A. In DC-B, HaaS is provided in units of servers 22 (eg, servers a, b, c, etc.). Further, monitoring information for each server 22 device corresponding to HaaS is obtained as in-DC monitoring (K0) and service monitoring (K2).

  In addition, both DC-A and DC-B provide (open) K0 result information to the outside. That is, both monitoring servers 3 (3A, 3B) transmit each K0 result information to the corresponding processing units (51A, 51B) of the integrated monitoring server 5 (K1A, K1B). For this reason, in this example, it is possible to identify a problem location such as an error or a failure for each server 22 device in the DC.

  The DC monitoring server 3 uses, as K0 result information, information indicating the state (NG) when externally detecting a negative state (NG) such as an error or failure in the DC or at a regular timing. To the integrated monitoring server 5. In addition, even when the state in the DC is normal (no problem), the monitoring server 3 desirably transmits information indicating the state (OK) to the external integrated monitoring server 5 as the K0 information. In this embodiment, the OK information is also transmitted. For example, if there is no transmission of OK information from the in-DC monitoring server 3 to the integrated monitoring server 5 (for example, periodic transmission), the integrated monitoring server 5 is likely to have a problem (NG) in the DC. I can judge.

  By accessing from the user terminal 1, the service of the site α by the HaaS (for example, the service SA, the corresponding URL is URL1) is used. At that time, for example, when an error or failure occurs in the server a in the DC-B (when the response of the service SA results in an error or the like due to that), the following is an example.

  (0) The user terminal 1 (group) accesses the service SA through the URL 1.

  (1) The service monitoring server 4 performs service monitoring (K2) of the service SA (URL1) by the service monitoring unit 41, and for example, DC-A (path RA) as the access destination DC2 of the service SA (URL1) Or DC-B (path RB). Furthermore, the actual access destination DC2 may be specified as one (for example, DC-B (path RB)). In this example, it is assumed that K2 cannot identify one access destination DC2. As an example of K2 monitoring item information, in URL1 (RA / RB), TAT: [500 ms], error presence / absence: NG (error present) [5 cases (all 1000 cases)], security: [0/1000 pattern] ( Problem pattern hit rate), etc.

  (2) Since the service monitoring server 4 has detected an error in the service SA (URL1) from the result of (1), the monitoring result notification unit 42 sends the K2 result information to the integrated monitoring server 5 (monitoring result receiving unit 51S). Is notified (K2S). This K2S information includes, for example, URL1, error presence / absence (NG), etc., as in (1), as an example of monitoring item information.

  (3) On the other hand, the integrated monitoring server 5 performs system monitoring (K1). That is, the in-DC monitoring result information is received from the in-DC monitoring server 3 (3A, 3B) of each DC 2 to the corresponding monitoring result receiving unit 51 (51A, 51B) (K1A, K1B). As this information, for example, in the information of K1B, based on the log in DC-B, “a problem such as a failure has occurred in server a” (DC-B, server a, NG, etc.) is notified. . At that time, the information of K1A is “no problem with DC-A” (DC-A, OK, etc.).

  (4) The integrated monitoring server 5 uses the monitoring result comparison unit 52 based on the reception information at each monitoring result receiving unit 51, that is, the K2 result information of (2) and the K1 result information of (3). By comparing and comprehensively determining, a process for identifying a problem location / range in the target system is performed. As a result, each monitoring information is NG in K2S, OK in K1A, and NG in K1B. Therefore, for example, in URL1 (RB), server a (problem point 301) of DC-B (problem range 302) is obtained as result information. ) Is detected and identified as being NG (error, failure, etc.) (eg, NG of DC-B (server a) is related to NG of service SA).

  (5) The integrated monitoring server 5 outputs the result information. For example, the information is notified to the manager of the DC-B corresponding to the problem range or the manager of another DC-A that does not correspond. This information is, for example, that there is a problem (NG) in DC-B (server a) regarding the provision of service (SA). As a result, each DC 2 can confirm the status, cause, responsibility, etc. of the problem.

  As described above, in the first specific example, among the various monitoring information (K1, K2), the problem location 301 and the problem range 302 such as an error are determined based on the DC2 provision information (particularly K1B) in the K1 information. This is the case when it can be detected and identified. Further, in this example, as a result, only the K1 (K0) information (no K2 information) can identify the problem location / range, etc. (when the relationship with the service status is not considered). However, as in the example described later, it may not be possible to specify unless both the K1 and K2 information are referenced. Further, the relationship between the service state (service response error or the like) and the DC state (server failure or the like) can be specified or estimated by determining the K1 information plus the K2 information.

<Specific example 2>
An example of integrated monitoring in the case where the form of DC2 is SaaS will be described as a second specific example with reference to FIG. DC-B in FIG. 4 is in the form of SaaS, and is a DC that performs cooperation for the service by the cloud with respect to DC-A. In DC-B, SaaS is provided in units of software, service processing, etc. of the server 22 (eg, servers a, b, c, etc.). Further, as DC internal monitoring (K0) and service monitoring (K2), monitoring information in the service processing unit of the server 22 corresponding to SaaS is obtained.

  DC-A provides K0 result information to the outside, but DC-B does not provide K0 result information to the outside (no K1B information). That is, in the integrated monitoring server 5 (51B), the state in DC-B is basically unknown.

  Therefore, in this example, it is impossible to specify a detailed problem location (server unit) in DC-B, but the problem range in DC unit including the problem location is specified by comprehensive determination using K1 information and K2 information. Is possible.

  By accessing from the user terminal 1, the service of the site α by the SaaS (for example, service SB, URL2) is used. At that time, for example, when an error or failure occurs in the server a in the DC-B (when the response of the service SB results in an error or the like due to that), the following is an example.

  (0) The user terminal 1 (group) accesses the service SB through the URL 2.

  (1) The service monitoring server 4 performs service monitoring (K2) of the service SB (URL2) by the service monitoring unit 41, and the access destination DC2 of the service SB (URL2) is, for example, DC-A or DC-B. I know that. The contents of the monitoring item information of K2 are the same as in the first specific example, for example (URL2 (RA / RB), NG, etc.).

  (2) Since the service monitoring server 4 generates an error in the service SB (URL2) from the result of (1), the monitoring result notification unit 42 sends the K2 result information to the integrated monitoring server 5 (monitoring result receiving unit 51S). Is notified (K2S). This K2S information is the same as (1) (URL2, NG, etc.).

  (3) On the other hand, the integrated monitoring server 5 performs system monitoring (K1). That is, the K0 result information is transmitted from the in-DC monitoring server 3 (only 3A here) of each DC2 to the corresponding monitoring result receiving unit 51 (51A) (K1A). For example, the information of K1A from DC-A notifies that it is in a normal state (DC-A, OK, etc.). Further, since there is no K1B information from DC-B, the state in DC-B is unknown. For example, the monitoring result receiving unit 51B records that the state of DC-B is unknown.

  (4) The integrated monitoring server 5 makes a comprehensive judgment by comparing the results of K1 (K1A, K1B) and K2 (K2S). As a result, each monitoring information is NG in K2S, OK in K1A, and unknown in K1B, and as result information, for example, in URL2 (RB), it is NG in DC-B (problem range 402) (service) It is detected and specified that there is a high possibility that DC-B NG is related to SB NG.

  (5) The integrated monitoring server 5 outputs the result information as in the first specific example.

  As described above, in the second specific example, when there is no provision information (K1B) of some DC2 (DC-B) in the K1 information among the various monitoring information (K1, K2), In this case, the problem range 402 can be specified in DC units by comprehensive judgment.

<Specific example 3>
With reference to FIG. 5, as a third specific example, a case will be described in which communication between DC2s (such as cooperative communication between DC2 for service processing by the cloud) exists as the form of the target system. Further, as an example, both DC-A and DC-B provide K0 result information to the outside.

  By accessing from the user terminal 1, the service of the site α (for example, service SC, URL3) is used. At that time, for example, when an error or failure occurs in the dedicated line / VPN 91 between DC-A and DC-B (when the response of the service SC causes an error or the like as a result), the following is an example. .

  (0) The user terminal 1 (group) accesses the service SC through URL3. The URL 3 is associated with, for example, the route RA from the user terminal 1 to the DC-A first, and further associated with the route RC from the DC-A to the DC-B for cooperation of processing of the service SC. To do. When providing the service SC (URL 3), the dedicated line / VPN 91 is connected to the DC-B from the DC-A network device 21 that has received access (request) from the user terminal 1 via the route RA, for example. Then, the network device 21 of DC-B is accessed by the route RC. At that time, the DC-A network device 21 cannot obtain a normal response from DC-B due to a failure in the dedicated line / VPN 91, for example (error detection: URL3 (RC), NG, etc.). As a result, it is assumed that the response of the service SC to the user terminal 1 is an error or the like.

  (1) The service monitoring server 4 performs service monitoring (K2) of the service SC (URL3) by the service monitoring unit 41, and the access destination DC2 of the service SC (URL3) is, for example, DC-A (RA) and It is known to be DC-B (RB). Furthermore, it is desirable to know that there is an access (RC) from DC-A to DC-B, but comprehensive judgment in the integrated monitoring server 5 is possible even if it is not known. The contents of the monitoring item information of K2 are the same as in the first specific example, for example (URL3 (RA / RB / RC), NG, etc.).

  (2) Since an error in the service SC (URL 3) has occurred from the result of (1), the service monitoring server 4 sends the K2 result information to the integrated monitoring server 5 (monitoring result receiving unit 51S) by the monitoring result notification unit 42. Is notified (K2S).

  (3) On the other hand, the integrated monitoring server 5 performs system monitoring (K1). That is, the K0 result information is transmitted from the in-DC monitoring server 3 (3A, 3B) of each DC2 to the corresponding monitoring result receiving unit 51 (51A, 51B) (K1A, K1B). For example, OK or the like is notified in the K1A information from DC-A and the K1B information from DC-B, respectively. This OK information means that there is no problem as DC. Further, for example, the OK information of K1A may include detection information (URL3 (RC), NG, etc.) such as an error in communication via the dedicated line / VPN91 in the route RC between DC-A and B ( DC-A, OK (A-B NG, etc.), etc. can be comprehensively judged even if they are not included.

  (4) The integrated monitoring server 5 makes a comprehensive judgment by comparing the results of K1 (K1A, K1B) and K2 (K2S). Accordingly, in this example, K2S is NG, K1A is OK, and K1B is OK. As a result information, for example, in URL3 (RC), the dedicated line / VPN91 (problem range) between DC-A and B is It is detected and identified as being NG (eg, there is a high possibility that NG between DC-A and B is related to NG of service SC).

  (5) The integrated monitoring server 5 outputs the result information as in the first specific example.

  As described above, in the third specific example, when there is no problem in the K1 information (the state in each DC) among the various types of monitoring information (K1, K2), the communication between DCs is a problem by comparison. This is a case where identification or estimation is possible.

<Other specific examples>
Other examples related to comprehensive judgment will be described with reference to FIG. As described above, the K1 information (DC state) includes OK, NG, unknown, and the like. Further, there are OK, NG, etc. as K2 information (service status). In K2, there are cases where the access destination DC can be specified for the service (URL) and where it cannot be specified. In addition, as a form of providing (disclosing) K0 result information from the DC 2 to the integrated monitoring server 5, there are provided (◯), non-provided (×), and the like. In the case of ×, in detail, in the corresponding DC, it is monitored by the monitoring unit in the DC and grasps the state (problem) in the DC, but since the information is not disclosed to the outside, the external There are cases where the state (problem) in the DC cannot be grasped, or the case where the DC does not have a DC monitoring unit for providing to the outside.

  There are various combinations of the above states depending on the form of the target system. An example of the overall judgment output according to them (OK: no problem, NG: problem exists) is as follows. For example, as shown in FIG. 6, it is assumed that there are four DC-A (own company X), DC-B (other company Y), DC-C (other company Z), and DC-D (other company). For providing the K0 result information, for example, DC-A and B are ◯, and DC-C and D are ×. Basically, it is assumed that one access destination DC cannot be specified for a service (URL). Further, for example, it is assumed that a failure in DC has actually occurred in DC-B and DC-D when accessing the service of the site.

  (1) {K2 = NG, K1A = OK, K1B = NG} → {DC-A: OK, DC-B: NG, DC-C, D: Unknown (possible NG possibility)}. DC-A has no problem and DC-B has a problem. For example, a service response error occurs due to a server failure in DC-B. Further, when the access destination DC or the like of the service can be specified by K2 (service monitoring server 4) or the like, it can be specified in more detail. For example, when the access destination is DC-B, the above-mentioned DC-C and D have a high possibility of OK. For example, when the access destination is DC-B and D, the above-mentioned DC-C has a high possibility of being OK, and the DC-D has a high possibility of being NG.

  (2) {K2 = NG, K1A = OK, K1B = OK} → {DC-A, B: OK, DC-C, D: NG possible} Since there is no problem in both DC-A and B, it can be determined that there is a high possibility that there is a problem in at least one of DC-C and D. For example, the problem range is DC-C, D. Further, if there is communication between DC-A and B as in the third specific example described above, it is also a candidate for the problem range.

  (3) {K2 = NG, K1A = NG, K1B = NG} → {DC-A: NG, DC-B: NG, DC-C, D: unknown (possible NG possibility)}. Both DC-A and B have problems. Further, when the access destination of the service can be identified as DC-B by K2 or the like, it can be determined that there is a high possibility that there is the most problem in DC-B with respect to a service error or the like. Further, for example, when it is known that the access destination is DC-A and the linkage from DC-A to DC-B, it can be determined that DC-A is most likely to have a problem with respect to a service error or the like.

  (4) {K2 = OK, K1A = OK, K1B = NG} → {DC-A: OK, DC-B: NG, DC-C, D: unknown (possible NG possibility)}. This is a case where there is no error or the like as a service and a normal response can be made, but there is a case where an internal problem has occurred on the DC2 side (DC-B). That is, for example, it can be determined that DC-B has a problem such as a server failure that is not related to the processing of the service.

<Effects of the embodiment>
According to the present embodiment, when there are a plurality of data centers that constitute a target system such as a Web site, it is possible to realize early detection of a problem (error, failure, security, etc.) and specification of a problem range. Then, it is possible to proceed to dealing with a problem in the corresponding DC by notification or the like. For example, when an error or the like of a site service occurs, the corresponding DC can immediately recognize that there is a problem with its own DC and the necessity of handling by identifying the problem range and notifying. Further, it is possible to immediately recognize that there is no problem with at least the own DC in the DC other than the corresponding DC. That is, it is possible to clarify and isolate the necessity and responsibility for dealing with problems in a plurality of DCs.

  Even when only a plurality of DCs are monitored by the integrated monitoring server 5 (first means) (when there is no monitoring information by the service monitoring server 4 (second means)), there is a corresponding effect. That is, the problem range may be determined and specified by comparing the monitoring information of a plurality of DCs. Further, by making a comprehensive judgment by combining monitoring information from the service monitoring server 4, more detailed judgment and identification can be made.

  In addition, for example, even when a predetermined service level agreement (SLA) is set for providing a cloud service by the target system, according to the present system, the state related to the SLA is monitored and whether or not the SLA is satisfied. Etc. can be confirmed.

<Modification>
In the present embodiment (FIG. 2), the integrated monitoring server 5 performs the comprehensive judgment (problem range specification) processing based on the monitoring (K1, K2). Not only this but the form which performs the said process with another apparatus (for example, server which cooperates), the form which integrated the service monitoring server 4 and the integrated monitoring server 5, etc. are all possible.

  The service monitoring server 4 has, as a detailed form, for example, a form implemented as a software module in the user terminal 1, a form such as a server provided on the site (DC 2) side, a form like a relay server provided on the Internet 90, etc. Various types are possible.

  Further, the service monitoring server 4 is not limited to service monitoring (K2) for one service (URL) -1 site, but manages service information by managing correspondence information such as a plurality of services (URL) -multiple sites. (K2) may be performed.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

  The present invention can be used for a data center, a Web site, a cloud computing system, and the like.

  DESCRIPTION OF SYMBOLS 1 ... User terminal, 2 ... DC (data center), 3 (3A, 3B) ... Monitoring server in DC, 4 ... Service monitoring server, 5 ... Integrated monitoring server, 21 ... Network equipment, 22 ... Server, 41 ... Service monitoring , 42 ... Monitoring result notification unit, 51 (51A, 51B, 51S) ... Monitoring result receiving unit, 52 ... Monitoring result comparison unit, 90 ... Internet, 91 ... Dedicated line / VPN, 100 ... Integrated monitoring system, 101, 102 103, site (system), 301, problem location, 302, 402, 502, problem range.

Claims (8)

An information processing system for monitoring a target system having a plurality of data centers constituting a website that provides services or contents to user terminals via a communication network,
A first processing unit for monitoring an internal state of each of the plurality of data centers;
A second processing unit that monitors the status of access to the service or content of the website from the user terminal;
By using the first monitoring information by the first processing unit and the second monitoring information by the second processing unit and comparing them comprehensively, a plurality of data centers of the target system are obtained. An integrated monitoring system, characterized by comprising a third processing unit that identifies a problematic part or a range including the problem and outputs the information. The integrated monitoring system according to claim 1,
A first monitoring server comprising the first processing unit and the third processing unit;
An integrated monitoring system comprising: a second monitoring server including the second processing unit. The integrated monitoring system according to claim 1,
Integrated monitoring characterized by having a fourth processing unit for notifying the manager of the corresponding data center of the information on the part having the problem or the range including the same obtained by the third processing unit system. The integrated monitoring system according to claim 1,
The second processing unit manages URL information or a route in accessing the service or content of the Web site from the user terminal or information on the address of the data center of the access destination, An integrated monitoring system characterized by monitoring whether a problem has occurred in a route or an access destination data center. The integrated monitoring system according to claim 1,
In the case where the second monitoring information is NG, the third processing unit includes a first data center in which the state is OK in the first monitoring information, and a second in which the state is NG. An integrated monitoring system, characterized in that if there is a data center, it is determined that there is a problem in the second data center. The integrated monitoring system according to claim 1,
The first processing unit is provided with monitoring information on a state inside the data center from each of the plurality of data centers,
The plurality of data centers include a data center that provides the monitoring information to the first processing unit and a data center that does not provide the monitoring information,
When the second monitoring information is NG and the status of the data center to be provided is OK among the first monitoring information, the third processing unit has a problem with the data center that is not provided. An integrated monitoring system characterized by being judged to be present. The integrated monitoring system according to claim 1,
The third processing unit is when the second monitoring information is NG and there is a first data center and a second data center whose status is OK among the first monitoring information. And determining that there is a problem in communication between the first data center and the second data center. The integrated monitoring system according to claim 1,
In the case where the second monitoring information is OK and the first monitoring information includes the first data center in which the state is NG, the third processing unit And determining that there is a problem that is not related to the state of access to the service or content.

Images