JP5686027B2 - Network failure detection method, apparatus, and program in virtual machine environment - Google Patents

Description

  The present invention relates to a network failure detection method, apparatus, and program in a virtual machine environment for detecting a network communication failure between virtual servers in a large-scale data center.

  2. Description of the Related Art In recent years, with the development of cloud computing, a large number of servers are arranged in a data center, and a service is provided by using a plurality of servers. In such a form, a server virtualization technique for operating a plurality of virtual servers on one physical server is known. Thousands or tens of thousands of virtual servers are deployed in one data center, and a huge amount of processing is realized by communication between virtual servers. The virtual server can be generated and deleted in a short time as necessary, and can be moved to another physical server.

  In a network system including a virtual server, since the network configuration frequently changes, there is an increased risk of falling into an unstable state such as communication being impossible. For this reason, there is a need for a network communication reachability confirmation technique that periodically shortens the communication normality of the inter-server network and detects an abnormality early so as to reduce the service stop time.

  The following technique is known as a conventional technique for checking network reachability (for example, the technique described in Patent Document 1). In a method for managing computer and relay node abnormalities in communication between adjacent segments of a computer network in which a plurality of segments are relayed by a plurality of relay nodes, management of whether or not all computers and relay nodes constituting the network are abnormal A network management device, and each relay node polls the computer belonging to the segment to which the relay node belongs and other relay nodes belonging to the segment to monitor whether the computer and other relay nodes are abnormal, This is a technique for notifying a network management device when an abnormality is found.

  Further, as another conventional technique, the following technique is known (for example, the technique described in Patent Document 2). A management console and a plurality of computers are connected to each other via a communication path, the plurality of computers constitute one or a plurality of computer groups, and the one or a plurality of computer groups are trees having a management console as a root. A computer monitoring method logically related by a structure, wherein each of the plurality of computers communicates with a computer in its own computer group based on a computer address of a computer registered in an address list. This is a technique characterized by comprising monitoring means for mutual monitoring, fault detection means for detecting faults, and fault notification means for notifying all computers belonging to the computer group corresponding to the parent node of fault detection results. .

JP 11-88399 A JP 2003-271471 A

  Any of the above-described conventional techniques is a technique for confirming communication normality between servers, and the number of combinations for confirming communication normality can be reduced. However, these conventional techniques have a problem in that it is impossible to confirm the normality of communication for all destinations because the number of destinations is also reduced.

  Therefore, an object of the present invention is to implement a virtual server as a destination and reduce the number of combinations.

  In an example of the aspect, in each physical server in which a plurality of virtual servers are operated and connected to each other via a network, in the physical server that acquires and manages information on communication reachability regarding each virtual server that operates on the own physical server A measurement result acquisition step, and in each physical server, information related to communication reachability for each virtual server operating on another physical server, requesting and acquiring the other physical server to acquire measurement results between physical servers, and In each physical server, in accordance with a request from another physical server, acquiring information on communication reachability for each virtual server operating on the own physical server, and transmitting / receiving measurement results between physical servers responding to the other physical server; Run.

It becomes possible to acquire communication reachability information regarding all virtual servers included in the own and other physical servers.
As a result, the number of combinations for transmitting and receiving measurement packets can be reduced, and the network load and server processing load can be reduced.

It is a network block diagram of the communication reachability confirmation method generally considered. It is a network block diagram of an embodiment. It is a functional block diagram of an embodiment. It is a figure which shows the data structural example of the virtual server information table in embodiment. It is a flowchart of the measurement process in a physical server in embodiment. It is a flowchart by the side of the transmission of the measurement process between physical servers in embodiment. It is a flowchart by the side of the reception of the measurement process between physical servers in embodiment. It is a figure which shows the example of the measurement packet format between the physical servers in embodiment. It is a figure which shows the example of the measurement response packet format between the physical servers in embodiment. It is a figure which shows the hardware structural example of the physical server in embodiment.

Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings.
First, a generally conceivable technique for checking network reachability will be described, and then an embodiment will be described.

As a technique for confirming the network reachability of a network, a technique for detecting a communication failure in an IP network using a probe device is generally known.
FIG. 1 is a network configuration diagram of a communication reachability confirmation method that can be generally considered. In FIG. 1, a server 101, which is a plurality of physical computers, is divided into several network groups # 1 to # 4 called sub-networks 104. Each sub-network 104 is configured to route IP (Internet Protocol) packets. Are mutually connected by a router 102 which is a device for performing the above.

  As a general communication reachability confirmation method in such a network system, a configuration in which one probe dedicated device 103 (hereinafter simply referred to as “probe 103”) is arranged in each sub-network 104 can be considered. The communication reachability between the sub-networks 104 is confirmed by performing full-mesh communication between the probes 103 arranged for each sub-network 104 through the router 102 and all the combinations of the probes 103. To do.

However, as a problem 1, this method cannot confirm the communication reachability between the servers 101 even if the communication reachability between the sub-networks 104 can be confirmed.
Furthermore, in recent years, server virtualization technology that operates a plurality of virtual servers on one physical server 101 has become widespread. In this case, not only the physical server 101 but also a technology for confirming communication reachability between virtual servers is required. In such a background, it is also conceivable to incorporate a probe function in a virtual server and perform full mesh communication between the virtual servers. However, as a problem 2, when such a method is adopted, as the number of physical servers 101 and virtual servers operating on the physical server 101 increases, the number of communication combinations for reachability confirmation becomes enormous and the amount of communication increases. It becomes. Such a situation can induce network congestion.

FIG. 2 is a diagram illustrating a network configuration in an embodiment for solving the above-described problem.
The physical server 201 is connected by a router 207. On the physical server 201, a plurality of virtual servers 202, which are software systems that run under a predetermined operating system, run.

  The hypervisor 203 comprehensively controls a plurality of virtual servers 202 operating on the physical server 201. Specifically, the hypervisor 203 can operate each virtual server 202 in cooperation with the operating system of the physical server 201 by sharing resources such as a CPU (Central Processing Unit) and a memory of the physical server 201. To control. The hypervisor 203 has the function of the virtual switch 205. The virtual switch 205 distributes packet data input / output to / from a network interface card (NIC) 206 mounted on the physical server 201 to each virtual server 202 on the physical server 201. With this function, each virtual server 202 on the physical server 201 can communicate with the virtual server 202 on the other physical server 201, the router 207, and other client computers on the Internet via the NIC 206.

  In the present embodiment, the hypervisor 203 on one physical server 201 has the function of the probe 204 corresponding to the physical server 201. The probe 204 transmits a communication confirmation packet to each virtual server 202 in the own physical server 201, and stores a reply result. The probe 204 also transmits a communication confirmation packet to each virtual server 202 in the other physical server 201. When the probe 204 receives a communication confirmation packet addressed to the virtual server 202 in the own physical server 201 from the other physical server 201, the probe 204 searches the communication status of the virtual server 202, and if OK, OK, NG, NG, and other physical server 201. Reply to

  As described above, in this embodiment, the function of the probe 204 is not arranged in each virtual server 202 but is arranged in the hypervisor 203 in correspondence with the physical server 201. With the function of this probe 204, it is possible to check the normality of communication addressed to all virtual servers 202 in the own and other physical servers 201. With this configuration, the number of combinations for transmitting and receiving measurement packets can be reduced, and the network load and server processing load can be reduced.

FIG. 3 is a functional block diagram of the present embodiment for realizing the above effects, and is a diagram showing a functional block configuration of the probe 204 of FIG.
In FIG. 3, the in-physical server measurement result acquisition unit 300-1 acquires and manages information related to communication reachability regarding each virtual server 202 operating on the own physical server 201. More specifically, the in-physical server measurement result acquisition unit 300-1 acquires a measurement result and response time related to communication reachability with respect to the virtual server 202 operating on the own physical server 201, and accumulates it as a measurement result 314. At this time, the intra-physical server measurement result acquisition unit 300-1 refers to the database table of the virtual server information 312 and identifies the virtual server 202 operating on the own physical server 201.

  The inter-physical server measurement result acquisition unit 300-2 requests and acquires information on communication reachability regarding each of the virtual servers operating on the other physical server 201 from the other physical server. More specifically, the inter-physical server measurement result acquisition unit 300-2 relates to the reachability of the virtual server 202 included in the physical server 201 with respect to the probes 204 arranged one by one on the other physical server 201. The measurement result and response time are requested as a measurement packet. At this time, the inter-physical server measurement result acquisition unit 300-2 refers to the intra-network server information 313 and identifies the virtual server 202 operating on the other physical server 201. The inter-physical server measurement result acquisition unit 300-2 measures the measurement result and response time related to communication reachability related to the virtual server 202 included in the physical server 201 from the probes 204 arranged one by one on the other physical server 201. Received as a response packet. Then, the inter-physical server measurement result acquisition unit 300-2 accumulates the received measurement result and response time as the measurement result 314.

  The inter-physical server measurement result transmission unit 300-3 receives the measurement result request by the measurement packet from the probe 204 arranged one by one in the other physical server 201, and reaches the communication of the virtual server 202 included in the own physical server 201. Get measurement results and response time for sex. At this time, the inter-physical server measurement result transmission unit 300-3 refers to the database table of the measurement result 314 and acquires the measurement result of the target virtual server 202. Then, the inter-physical server measurement result transmitting unit 300-3 returns the acquired measurement result and response time as a measurement response packet to the physical server 201 that transmitted the measurement packet.

  As described above, in this embodiment, the measurement packet and the measurement response packet are transmitted and received only between the physical servers 201 by the probe 204 having the configuration illustrated in FIG. As a result, it is possible to acquire the communication reachability measurement results and response times for all virtual servers 202 included in the own and other physical servers 201.

  In FIG. 3, the physical server internal measurement result acquisition unit 300-1 includes a virtual server information acquisition unit 301, a virtual server measurement packet transmission unit 302, a virtual server response packet reception unit 303, and a virtual server measurement result accumulation. Part 304 is provided. The physical server internal measurement result acquisition unit 300-1 refers to the database table of the virtual server information 312 and identifies the virtual server 202 operating on the own physical server 201. The virtual server measurement packet transmission unit 302 transmits a ping request to the virtual server 202 specified by the virtual server information acquisition unit 301. Then, the virtual server response packet receiving unit 303 acquires a measurement result and a response time for the virtual server 202 operating on the own physical server 201 by monitoring a ping response corresponding to the ping request. The virtual server measurement result accumulation unit 304 accumulates the measurement result and response time for the virtual server 202 operating on the acquired physical server 201 as the measurement result 314.

  As described above, the communication reachability measurement result and response can be transmitted to the virtual server 202 operating on the own physical server 201 by simply transmitting a ping request to the virtual server 202 obtained from the virtual server information 312. It becomes possible to get time.

  In FIG. 3, the inter-physical server measurement result acquisition unit 300-2 includes an intra-network server information acquisition unit 305, a counter server measurement packet transmission unit 306, a counter server response packet reception unit 307, and a counter server measurement result storage unit 308. Is provided. The intra-network server information acquisition unit 305 refers to the intra-network server information 313 and identifies the virtual server 202 operating on the other physical server 201. The opposing server measurement packet transmission unit 306 transmits a measurement packet requesting a measurement result of communication reachability to the virtual server 202 operating on the identified other physical server 201. The opposite server response packet receiving unit 307 monitors and receives a response of the measurement response packet storing the communication reachability measurement result transmitted by the probe 204 in the target physical server 201 with respect to the measurement packet. The opposite server measurement result accumulation unit 308 obtains the measurement result from the received measurement response packet and calculates a response time from when the measurement packet is transmitted until the measurement response packet is received. Then, the opposing server measurement result storage unit 308 stores the measurement result and response time corresponding to the virtual server 202 operating on the other physical server 201 acquired as described above as the measurement result 314.

  As described above, the probes 204 arranged one by one on the physical server 201 communicate with the probes 204 of one physical server 201 including them even if there are a plurality of virtual servers 202 in the other physical server 201. It is possible to obtain communication reachability measurement results and response times for the plurality of virtual servers 202 simply by performing the above. As a result, it is possible to efficiently collect information of each virtual server 202 in the other physical server 201.

  3, the inter-physical server measurement result transmission unit 300-3 includes a counter server measurement packet reception unit 309, a counter server measurement result acquisition unit 310, and a counter server response packet transmission unit 311. The opposing server measurement packet receiving unit 309 receives a measurement result request addressed to the virtual server 202 operating on the own physical server 201 by a measurement packet from the probe 204 arranged one by one on the other physical server 201. The opposing server measurement result acquisition unit 310 acquires the measurement result regarding the virtual server 202 requested by the received measurement packet from the database table of the measurement result 314. The opposite server response packet transmission unit 311 returns a measurement response packet storing the acquired measurement result to the physical server 201 that transmitted the received measurement packet.

  As described above, the inter-physical server measurement result transmission unit 300-3 communicates with the virtual server 202 operating on the own physical server 201 in accordance with a request from the probe 204 disposed on the other physical server 201. It becomes possible to respond to the measurement result.

FIG. 4 is a diagram illustrating a data configuration example of each database table of the virtual server information 312, the in-network server information 313, and the measurement result 314 of FIG. 3.
The virtual server information 312 is information related to all virtual servers 202 in the own physical server 201. The virtual server information 312 table corresponds to each virtual server 202 in the own physical server 201 with a record composed of three pieces of information of “virtual server name”, “virtual server MAC address”, and “virtual server IP address”. Multiple records. No. 4 in FIG. 1 table structure. “Virtual server name” is the host name of the virtual server 202. The “virtual server MAC (Media Access Control) address” is a MAC address used by the virtual server 202 for communication with other servers. The “virtual server IP address” is an Internet protocol address assigned to the virtual server 202.

  The intra-network server information 313 is information related to all virtual servers 202 in the other physical server 201. The intra-network server information 313 table has the same record structure as the virtual server information 312 table. The intra-network server information 313 table includes each virtual server in the other physical server 201 that can grasp a record including three pieces of information “virtual server name”, “virtual server MAC address”, and “virtual server IP address”. A plurality of records are stored corresponding to the server 202. No. 4 in FIG. 2 table structure.

  The measurement result 314 is information regarding the communication reachability of each virtual server 202 in the own / other physical server 201. The measurement result 314 table confirms a record including “virtual server name”, “virtual server MAC address”, “virtual server IP address”, “measurement result” indicating presence / absence of response, and “delay time calculation result”. A plurality of records are stored corresponding to each virtual server 202 of the own / other physical server 201. No. 4 in FIG. 3 table structure.

  The operation of this embodiment having the functional block configuration of FIG. 3 and the table configuration example of FIG. 4 will be described below. This embodiment is roughly classified into three processes. The processing on the transmission side of the measurement processing within the physical server, the measurement processing between physical servers, and the processing on the reception side of the measurement processing between physical servers. Hereinafter, description will be made sequentially with reference to the flowcharts of FIGS. These processes are, for example, a hardware configuration of the physical server 201 in FIG. 10 to be described later, and the CPU 1001 reads the control program of the hypervisor 203 stored in the external storage device 1005 such as a hard disk into the memory 1002 and executes it. is there. In the following description of processing, the physical server 201, virtual server 202, hypervisor 203, probe 204, virtual switch 205, NIC 206, router 207, etc. in FIG. In addition, the respective processing units 301 to 311 in FIG. 3, the virtual server information 312, the in-network server information 313, and the measurement result 314 databases are also referred to.

FIG. 5 is a flowchart showing the flow of the physical server internal measurement process. This process corresponds to the function of the physical server internal measurement result acquisition unit 300-1 in FIG.
First, the virtual server information acquisition unit 301 acquires virtual server information 312 related to the virtual server group 202 existing in the own physical server 201 from the database table (step S501). The database table of the virtual server information 312 is held on the memory 1002 or the external storage device 1005 in the hardware configuration of the physical server 201 in FIG. The table of the virtual server information 312 is a database provided in advance for the function of the hypervisor 203 that controls the virtual server 202. No. 4 in FIG. As illustrated as 1, the virtual server information 312 table includes, for example, “virtual server name”, “virtual server MAC address”, “virtual server IP address”, and the like. Note that the virtual server information 312 may include management information regarding other virtual servers 202.

  Next, as a process of the virtual server measurement packet transmission unit 302, a measurement packet is transmitted to each “virtual server IP address” of the virtual server group 202 acquired in step S501 (step S503). This transmission process is an iterative process for each “virtual server IP address” (loop process in steps S502 and S504). Since the virtual server 202 is not provided with a special function, an ICMP echo request (hereinafter referred to as a “ping request”) by ping software, which is a packet that can be answered by a general server, is used as a measurement packet. Internet Control Notification Protocol (ICMP) is a protocol used for diagnosing a communication state and notifying an error in the Internet protocol. The ping software can confirm reachability by throwing a ping request, which is an ICMP echo request, to the destination server and returning an ICMP echo reply (hereinafter referred to as a “ping response”) from the destination server.

  After a series of ping requests are transmitted, a ping response waiting timer is started as processing of the virtual server response packet receiving unit 303 (step S505). As this timer, for example, a timer function provided in the physical server 201 is used.

Next, as the processing of the virtual server response packet receiving unit 303, it is determined whether or not the time measurement of the ping response wait timer has expired (step S506).
If the timing of the ping response wait timer has not expired and the determination in step S506 is NO (displayed as “[N]” in the figure, the same applies hereinafter), a response to the ping response is returned as the processing of the virtual server response packet receiving unit 303. Whether or not is detected is determined (step S507).

  If no reply of the ping response is detected and the determination in step S507 is NO, the determination processing in steps S506 and S507 is repeated as the processing of the virtual server response packet reception unit 303.

  If the virtual server 202 is operating normally and normal communication is possible via the virtual switch 205 in the physical server 201, a ping response is returned from the virtual server 202. As a process of the virtual server response packet receiving unit 303, when a reply of the ping response is detected and the determination in step S507 is YES (displayed as “[Y]” in the figure, the same applies hereinafter), the ping response is received. (Step S508).

Next, as a process of the virtual server measurement result accumulation unit 304, first, the virtual server IP address of the transmission source is extracted from the received ping response (step S509).
Subsequently, as the processing of the virtual server measurement result storage unit 304, the following information is written in the record corresponding to the virtual server IP address extracted in step S509 of the database table of the measurement result 314 (step S510). This record is stored in the NO. 3 table structure. If there is no corresponding record, a new record is created, and “virtual server name”, “virtual server MAC address”, and “virtual server IP address” are first written. In the record, “normal” (“OK”) is recorded as “measurement result”, and the ping response time, that is, the timer value of the current ping response wait timer is recorded as the delay time calculation result.

  Thereafter, as a process of the virtual server measurement result accumulation unit 304, it is determined whether or not a ping response has been returned from all the virtual servers 202 that have transmitted the ping request (step S511). The virtual server measurement result accumulation unit 304 receives and holds the virtual server IP address of the virtual server 202 that is the transmission destination of the ping request transmitted from the virtual server measurement packet transmission unit 302. Each time the virtual server IP address of the ping response is extracted in step S509, the virtual server IP address is deleted from the held virtual server IP address list. In step S511, it is determined whether all the virtual servers 202 having transmitted the ping request have received a ping response by determining whether the list is empty.

  If the reply of the ping response from all the virtual servers 202 that transmitted the ping request has not yet been completed and the determination in step S511 is NO, the process returns to step S506. As a result, as the process of the virtual server response packet receiving unit 303, the determination of the expiration of the ping response wait timer and the return of the next ping response is repeated.

  When the reply of the ping response from all the virtual servers 202 that transmitted the ping request is completed and the determination in step S511 is YES, the intra-physical server measurement process in FIG. 5 is terminated.

  On the other hand, if the ping response wait timer expires before the reply of the ping response from all the virtual servers 202 that transmitted the ping request is not yet completed and the determination in step S506 becomes YES, the following processing is executed. As a process of the virtual server measurement result storage unit 304, the virtual server IP address of the virtual server 202 that has not yet received the ping response is extracted from the list. Then, the following information is written in each record corresponding to the virtual server IP address. This record is stored in the NO. 3 table structure. If there is no corresponding record, a new record is created, and “virtual server name”, “virtual server MAC address”, and “virtual server IP address” are first written. In the record, “abnormal” (“NG”) is recorded as “measurement result”.

  As described above, in this embodiment, the following operation is realized by one probe 204 arranged as a function of the hypervisor 203 for one physical server 201. It becomes possible to acquire the measurement result and response time regarding the communication reachability of the virtual server 202 operating on the own physical server 201.

Next, FIG. 6 is a flowchart showing a flow on the transmitting side of the inter-physical server measurement process. This process is a process corresponding to the function of the inter-physical server measurement result acquisition unit 300-2 in FIG.
First, the intra-network server information acquisition unit 305 acquires intra-network server information 313 related to all virtual server groups 202 existing in the other physical server 201 from the database table (step S601). The database table of the server information 313 in the network is held in the memory 1002 in the hardware configuration of the physical server 201 in FIG. Alternatively, it is held on the external storage device 1005. No. 4 in FIG. As illustrated as 2, the intra-network server information 313 includes, for example, “virtual server name”, “virtual server MAC address”, “virtual server IP address”, and the like. The intra-network server information 313 may include management information regarding other virtual servers 202. The method for generating the intra-network server information 313 will not be described in this embodiment. It may be created statically by a maintenance person or dynamically generated by a device such as a management manager that manages the entire server group in the network. For example, the management manager makes an inquiry to each physical server 201, and each physical server 201 returns the information of all the virtual servers 202 within itself to the management manager.

  Next, as a process of the counter server measurement packet transmission unit 306, a measurement packet is transmitted to each “virtual server IP address” in the intra-network server information 313 acquired in step S601 (step S603). Here, a value indicating a request (that is, a measurement packet) is set in the field indicating the packet type in the measurement packet. Next, the physical server IP address of the own physical server 201 is set in the field indicating the source address in the measurement packet. A number corresponding to the processing of the measurement packet is set in the field indicating the destination port number in the measurement packet. Further, the packet includes transmission time information (up to millisecond units). Details of a packet format example of the measurement packet will be described later with reference to FIG. The above transmission process is a repetitive process for each “virtual server IP address” (loop process in steps S602 and S604).

  After the series of measurement packets are transmitted, a measurement response packet waiting timer is started as processing of the counter server response packet receiving unit 307 (step S605). As this timer, for example, a timer function provided in the physical server 201 is used.

Next, as processing of the counter server response packet receiving unit 307, it is determined whether or not the measurement response packet waiting timer has expired (step S606).
If the timing of the measurement response packet waiting timer has not expired and the determination in step S606 is NO, it is determined as a process of the opposite server response packet reception unit 307 whether or not a response to the measurement response packet has been detected (step S607). . Specifically, it is determined whether or not the destination port number of the received packet is a number corresponding to the processing of the measurement response packet.

  If no response of the measurement response packet is detected and the determination in step S607 is NO, the determination processing in steps S606 and S607 is repeated as the processing of the opposing server response packet reception unit 307.

  As processing of the counter server response packet receiving unit 307, when a response of the measurement response packet is detected and the determination in step S607 is YES, the measurement response packet is received (step S608).

  Next, as processing of the counter server measurement result storage unit 308, first, the source virtual server IP address is extracted from the received measurement response packet (step S609). The counter server measurement result accumulation unit 308 receives and holds the virtual server IP address of the virtual server 202 that is the transmission destination of the measurement packet transmitted from the counter server measurement packet transmission unit 306. Then, every time the virtual server IP address of the measurement response packet is extracted in step S609, the virtual server IP address is deleted from the held list.

  Subsequently, as processing of the virtual server measurement result storage unit 304, the following information is written in the record corresponding to the virtual server IP address extracted in step S609 of the database table of the measurement result 314 (step S610). This record corresponds to No. 1 in FIG. 3 table structure. If there is no corresponding record, a new record is created, and “virtual server name”, “virtual server MAC address”, and “virtual server IP address” are first written. In the record, the response result “normal” (“OK”) / “abnormal” (“NG”) stored in the measurement response packet received in step S608 is recorded as “measurement result”. . Further, in the record, the time of the difference between the reception time information of the measurement response packet in step S608 and the time information of the transmission source included in the measurement response packet is recorded as “response time”.

  Thereafter, the process returns to step S606. As a result, the process of the opposite server response packet receiving unit 307 repeats the determination of the expiration of the measurement response packet waiting timer and the return of the next measurement response packet again.

  Then, when the measurement response packet waiting timer expires and the determination in step S606 becomes YES, the following processing is executed. As processing of the counter server measurement result storage unit 308, the virtual server IP address of the virtual server 202 that has not yet received the measurement response packet is extracted from the list. Then, the following information is written in each record corresponding to the virtual server IP address. This record is stored in the NO. 3 table structure. If there is no corresponding record, a new record is created, and “virtual server name”, “virtual server MAC address”, and “virtual server IP address” are first written. Then, “not received” is recorded as “measurement result” in the record.

  As described above, in this embodiment, the following operation is realized by one probe 204 arranged as a function of the hypervisor 203 for one physical server 201. It is possible to request the measurement result and response time regarding the communication reachability of the virtual server 202 included in the physical server 201 to the probes 204 arranged one by one in the other physical server 201.

  Finally, FIG. 7 is a flowchart showing the flow on the receiving side of the inter-physical server measurement process. This process corresponds to the function of the inter-physical server measurement result transmission unit 300-3 in FIG.

  First, the process of the counter server measurement packet receiving unit 309 is in a waiting state for receiving an arbitrary packet via the NIC 206 that is a LAN interface mounted on the physical server 201 (repetition of determination in step S701). ).

  When an arbitrary packet is received and the determination in step S701 is YES, as the processing of the counter server measurement packet receiving unit 309, it is determined whether or not the destination address of the packet is addressed to the virtual server 202 operating on the own physical server 201. Determination is made (step S702).

  If the destination address of the received packet is not addressed to the virtual server 202 operating on the own physical server 201 and the determination in step S702 is NO, the process returns to the packet reception waiting state in step S701.

  If the determination in step S702 is YES, it is determined whether or not a measurement packet has been received (step S703) as processing of the opposing server measurement packet reception unit 309. Specifically, it is determined whether or not the destination port number of the received packet is a number corresponding to the processing of the measurement packet.

If the received packet does not indicate a measurement packet and the determination in step S703 is NO, the process returns to the packet reception waiting state in step S701.
If the received packet indicates a measurement packet and the determination in step S703 is YES, the measurement packet is received and processed as processing of the opposite server measurement packet reception unit 309 (step S704).

  Next, as the processing of the counter server measurement result acquisition unit 310, the “measurement result” of the record corresponding to the virtual server IP address set as the destination address of the measurement packet received in step S 704 from the database table of the measurement result 314. The value of “is obtained. That is, in FIG. The value of “measurement result” is extracted from the table structure of the measurement result 314 of 3 (step S705).

  And as a process of the opposing server response packet transmission part 311, the following measurement response packets are produced | generated (step S706). First, a measurement response packet is generated in which the measurement result acquired in step S705 and the transmission source time information included in the measurement packet received in step S704 are stored. Next, a value indicating a reply (that is, a measurement response packet) is set in the field indicating the packet type in the measurement response packet. In addition, the physical server IP address of the transmission source physical server 201 set as the transmission source address of the measurement packet received in step S704 is set as the destination address of the measurement response packet. In addition, the virtual server IP address of the measurement target virtual server 202 set as the destination address of the measurement packet received in step S704 is set as the transmission source address of the measurement response packet. Further, a number corresponding to the processing of the measurement response packet is set in the field indicating the destination port number in the measurement response packet. Details of a packet format example of the measurement response packet will be described later with reference to FIG.

When the transmission of the measurement response packet is completed, the process returns to the packet reception waiting state in step S701.
As described above, in this embodiment, the following operation is realized by one probe 204 arranged as a function of the hypervisor 203 for one physical server 201. Upon receiving a request from the probe 204 arranged one by one in the other physical server 201, the measurement result and response time regarding the communication reachability of the virtual server 202 included in the own physical server 201 are returned to the physical server 201. It becomes possible.

  FIG. 8 is a diagram illustrating an example of a measurement packet format between physical servers in the present embodiment. FIG. 8 shows an example in which a measurement packet is transferred using a UDP (User Datagram Protocol) datagram format.

  In FIG. 8, the first 14 bytes are a header portion corresponding to a physical layer such as Ethernet (registered trademark). The 6-byte “Destination MAC Address” field stores the virtual server MAC address of the virtual server 202 from which the measurement result is acquired as destination physical address information. Using this information, routing to a destination server on a physical network such as Ethernet is performed. In the “Source MAC Address” field composed of the second 6 bytes of the physical layer header portion, the physical server MAC address of the own physical server 201 is stored as the transmission source physical address information of the measurement packet. Using this information, when returning a measurement response packet, return routing to the own physical server 201 on a physical network such as Ethernet is executed. A value “0x0800” designating that IPv4 (Internet Protocol version 4) is used as the Ethernet type is stored in the “Ether Type” field consisting of the third 4 bytes of the physical layer header portion. Here, “0x” indicates that the subsequent value “0800” is a hexadecimal number. IPv6 (Internet protocol version 6) may be adopted.

  Next, a 20-byte IP header portion (an area starting with “IP Header” in the figure) following the 14-byte physical layer header portion is a header portion of an IP packet based on the Internet protocol. A value “0x11” indicating that the packet is a UDP packet is stored in the “Protocol” field including the first byte. In the “Source IP Address” field consisting of the second 4 bytes of the IP header part, the physical server IP address of the own physical server 201 is stored as the source IP address information of the measurement packet. In the “Destination IP Address” field consisting of the third 4 bytes of the IP header part, the virtual server IP address of the virtual server 202 operating on the physical server 201 from which the measurement result is obtained is stored as destination IP address information. . As a result, the packet is routed to the destination virtual server 202 via the IP network.

  Next, an 8-byte UDP header portion (an area starting with “UDP Header” in the figure) following the 20-byte IP header portion is a header portion of a UDP datagram based on the UDP protocol. An arbitrary value is stored in the “Source Port Number” field consisting of the first two bytes. In the “Destination Port Number” field composed of the second 2 bytes of the UDP header part, a specific value indicating that the process is a measurement packet for communication reachability confirmation is determined and stored.

  The portion following the 8-byte UDP header portion (region starting with “UDP Data” in the figure) is the data portion of the UDP datagram based on the UDP protocol. In this part, information defined by this embodiment is stored. A value “0x01” indicating a request (that is, a measurement packet) is stored as a packet type in an area consisting of the first byte. In the area following 1 byte, the time information when the probe 204 of the physical server 201 that is the transmission source of the measurement packet transmits the measurement packet is stored as a value including up to milliseconds (ms).

  The measurement packet having the packet format shown in FIG. 8 is generated by the function of the opposite server measurement packet transmission unit 306 in FIG. 3, which is the function of the probe 204 of the physical server 201, or the processing in step S603 in FIG. Sent. Then, this measurement packet is received by the function of the counter server measurement packet receiving unit 309 in FIG. 3 which is the function of the probe 204 of the destination physical server 201 or the processing in step S704 in FIG. From the virtual server IP address in the “Destination IP Address” field of the IP header part of the measurement packet, the receiving physical server 201 can recognize the packet addressed to the virtual server 202 operating on the server itself in step S702 of FIG. Also, the physical server 201 on the receiving side can recognize the measurement packet in step S703 of FIG. 7 from “Destination Port Number” in the UDP header portion of the measurement packet and “Packet type” in the UDP data portion.

  FIG. 9 is a diagram illustrating an example of a measurement response packet format between physical servers in the present embodiment. FIG. 9 also shows an example in which the measurement response packet is transferred using the UDP datagram format, as in FIG.

  In FIG. 9, the first 14 bytes are a header portion corresponding to a physical layer such as Ethernet. The 6-byte “Destination MAC Address” field stores the physical server MAC address of the physical server 201 that requested the measurement result of the destination of the measurement response packet as the destination physical address information. The function of the probe 204 of the physical server 201 on the receiving side is the same as the source physical server MAC address of the measurement packet stored in the “Source MAC Address” field (FIG. 8) of the physical layer header part of the received measurement packet. Store in the field. Using this information, return routing to the requesting physical server 201 on the physical network such as Ethernet is performed. In the “Source MAC Address” field consisting of the second 6 bytes of the physical layer header part, the virtual server MAC address of the virtual server 202 from which the measurement result is acquired is stored as the transmission source physical address information of the measurement response packet. . A value “0x0800” designating that IPv4 is used as the Ethernet type is stored in the “Ether Type” field consisting of the third 4 bytes of the physical layer header portion. IPv6 (Internet protocol version 6) may be adopted.

  Next, a 20-byte IP header portion (an area starting with “IP Header” in the figure) following the 14-byte physical layer header portion is a header portion of an IP packet based on the Internet protocol. A value “0x11” indicating that the packet is a UDP packet is stored in the “Protocol” field including the first byte. In the “Source IP Address” field composed of the second 4 bytes of the IP header portion, the virtual server IP address of the virtual server 202 from which the measurement result is acquired is stored. The physical server 201 on the receiving side stores the virtual server IP address of the virtual server 202 from which the measurement result stored in the “Destination IP Address” field of the IP header portion of the received measurement packet is acquired in the above field. In the “Destination IP Address” field composed of the third 4 bytes of the IP header portion, the physical server IP address of the physical server 201 that requested the measurement result is stored as destination IP address information. The physical server 201 on the receiving side stores the physical server IP address stored in the “Source IP Address” field (FIG. 8) of the IP header portion of the received measurement packet in the above field. As a result, this packet is routed to the physical server 201 that requested the measurement result via the IP network.

  Next, an 8-byte UDP header portion (an area starting with “UDP Header” in the figure) following the 20-byte IP header portion is a header portion of a UDP datagram based on the UDP protocol. In the “Source Port Number” field consisting of the first two bytes, a specific value indicating the processing of the measurement response packet for communication reachability confirmation is determined and stored. An arbitrary value is stored in the “Destination Port Number” field composed of the second 2 bytes of the UDP header part.

  The portion following the 8-byte UDP header portion (region starting with “UDP Data” in the figure) is the data portion of the UDP datagram based on the UDP protocol. In this part, information defined by this embodiment is stored. A value “0x02” indicating a reply (that is, a measurement response packet) is stored as a packet type in an area consisting of the first byte. In the 1-byte area following the 1-byte, a response result, that is, a value “0x00” indicating “normal” (“OK”) or a value “0x01” indicating “abnormal” (“NG”) is stored as a measurement result. The Further, in the area following the 1 byte, the time information when the probe 204 of the physical server 201 that requested the measurement result transmits the measurement packet is stored as a value including up to milliseconds (ms). The physical server 201 on the receiving side stores the time information (FIG. 8) at the time of the measurement result request stored in the UDP data part of the received measurement packet in the field as it is.

  The measurement response packet having the packet format shown in FIG. 9 is transmitted by the function of the counter server response packet transmission unit 311 in FIG. 3, which is the function of the probe 204 of the physical server 201, or the processing in step S706 in FIG. The Then, this measurement response packet is received by the function of the counter server response packet receiving unit 307 in FIG. 3 or the processing in step S608 in FIG. 6, which is the function of the probe 204 of the physical server 201 that is the request source of the measurement result. The physical server 201 that has requested the measurement result can recognize the packet addressed to the own physical server 201 in step S607 of FIG. 6 from the physical server IP address in the “Destination IP Address” field of the IP header portion of the measurement response packet. Further, the physical server 201 that has requested the measurement result recognizes the measurement response packet from the “Source Port Number” in the UDP header portion of the measurement response packet and the “packet type” in the UDP data portion in step S607 of FIG. it can. Furthermore, the physical server 201 requesting the measurement result has received the virtual requesting the measurement result as a difference time between the reception time information of the measurement response packet and the time information of the transmission source included in the UDP data portion of the measurement response packet. The response time of the server 202 can be calculated.

  In the embodiment described above, for example, the virtual server 202 operating in each physical server 201 in FIG. 2 may be added, abolished, or moved. In such a case, the registration contents of the corresponding record in the database table of the virtual server information 312 of FIG. 3 held in the hypervisor 203 of FIG. 2 are updated by the function of the virtual server 202 itself of FIG. Will be. Therefore, in the present embodiment, the virtual server information acquisition unit 301 operating with the probe 204 having the configuration of FIG. 3 periodically monitors the virtual server information 312 to detect changes. Thereby, the list of the target virtual server group 202 to which the measurement packet is transmitted is automatically updated. Based on the update result, the virtual server measurement packet transmission unit 302, the virtual server response packet reception unit 303, and the virtual server measurement result storage unit 304 are operated again to update the measurement result 314. It becomes possible.

By using the probe 204 in the above embodiment, the following effects can be obtained.
It is possible to confirm the normality of communication addressed to all the virtual servers 202 only by arranging the probe function on the physical server 201 without arranging the probe functions individually on the virtual servers 202.

It is possible to reduce the number of combinations for transmitting and receiving measurement packets, and to reduce the network load and the server processing load.
For example, consider a network in which 100 physical servers 201 each having 30 virtual servers 202 deployed on one physical server 201 are configured.

As a generally considered method, it is assumed that probe functions are arranged one by one on the virtual server 202 and communication reachability is confirmed for all the opposing virtual servers 202. In this case, the number of communications is
3,000 × 2,999 = 8,997,000 times. On the other hand, the number of communications by the method of this embodiment is
Number of communication within physical server = 30 × 100 = 3,000 times Number of communication between physical servers = 99 × 30 × 100 = 297,000 times Total = 3,000 + 297,000 = 300,000 times. Therefore, in this example, it is possible to reduce the communication amount to about 300,000 / 8,997,000 = 0.033 = 3%.

FIG. 10 is a diagram illustrating a hardware configuration example of the physical server 201 in FIG. 2 in the embodiment.
10 includes a CPU 1001, a memory 1002, an input device 1003, an output device 1004, an external storage device 1005, a portable recording medium driving device 1006 into which a portable recording medium 1009 is inserted, and a communication interface 1007. , These are connected to each other by a bus 1008. The configuration shown in the figure is an example of a computer that can implement the above system, and such a computer is not limited to this configuration.

  The CPU 1001 controls the entire computer. The memory 1002 is a memory such as a RAM that temporarily stores a program or data stored in the external storage device 1005 (or the portable recording medium 1009) when executing a program, updating data, or the like. The CUP 1001 performs overall control by reading the program into the memory 1002 and executing it.

  The input / output device 1003 detects an input operation by a user using a keyboard, a mouse, or the like, notifies the CPU 1001 of the detection result, and outputs data sent under the control of the CPU 1001 to a display device or a printing device.

The external storage device 1005 is, for example, a hard disk storage device. Mainly used for storing various data and programs.
The portable recording medium driving device 1006 accommodates a portable recording medium 1009 such as an optical disc, SDRAM, or Compact Flash (registered trademark), and has an auxiliary role for the external storage device 1005.

The communication interface 1007 is a device for connecting, for example, a LAN (local area network) or WAN (wide area network) communication line.
The system according to the present embodiment is realized by the CPU 1001 executing a program having functions realized by the flowcharts of FIGS. 5, 6, and 7. For example, the program may be recorded and distributed in the external storage device 1005 or the portable recording medium 1009, or may be acquired from the network by the network connection device 1007.

Regarding the above embodiment, the following additional notes are disclosed.
(Appendix 1)
In each physical server in which a plurality of virtual servers are operated and connected to each other via a network, a measurement result acquisition step in the physical server that acquires and manages information related to communication reachability for each virtual server operating on the own physical server; ,
In each of the physical servers, an inter-physical server measurement result acquisition step for requesting and acquiring information regarding communication reachability regarding each of the virtual servers operating on the other physical server,
In each of the physical servers, in accordance with a request from another physical server, obtains information on communication reachability for each virtual server operating on the own physical server, and transmits / receives a measurement result between physical servers responding to the other physical server; ,
A network failure detection method in a virtual machine environment characterized by executing
(Appendix 2)
In the physical server internal measurement result acquisition step,
A virtual server operating on the own physical server is identified by referring to a database table of virtual server information that is information on the virtual server, and a measurement result and response time regarding communication reachability to the virtual server operating on the own physical server And store the acquisition results in a database table of measurement results,
In the inter-physical server measurement result acquisition step, a virtual server operating on the other physical server is identified by referring to a database table of server information in the network that is information relating to the virtual server operating on the other physical server, and the other Information on the communication reachability of the virtual server included in the other physical server is requested to the physical server in the measurement packet, and information on the communication reachability regarding the virtual server included in the other physical server is requested from the other physical server. Is stored as a measurement response packet, and the received information on communication reachability is stored in a database table of measurement results,
In the inter-physical server measurement result step, according to the request by the measurement packet received from the other physical server, the information on the communication reachability of the virtual server included in the own physical server is acquired from the measurement result database table, Reply as the measurement response packet;
The network failure detection method in the virtual machine environment according to appendix 1, wherein:
(Appendix 3)
In the physical server internal measurement result acquisition step,
Referring to the database table of the virtual server information, obtain virtual server information that identifies a virtual server running on the own physical server,
To the specified virtual server, send an echo request to the virtual server measurement packet,
By receiving an echo reply corresponding to the echo request, receiving a virtual server response packet to obtain information on the communication reachability to the virtual server running on the own physical server,
Storing the information on the reachability of the communication with respect to the acquired virtual server operating in the acquired physical server in the measurement result database table, and storing the measurement result against the virtual server;
The network failure detection method in the virtual machine environment according to appendix 2, wherein:
(Appendix 4)
In the measurement result acquisition step between the physical servers,
With reference to the database table of the server information in the network, the server information in the network that identifies the virtual server operating on the other physical server is acquired,
Transmitting the opposite server measurement packet for transmitting the measurement packet for requesting the communication reachability information toward the virtual server operating on the specified other physical server,
Opposing server response packet reception (307) for monitoring and receiving a response of the measurement response packet storing the communication reachability information transmitted by the target physical server for the measurement packet;
Obtaining information on the communication reachability from the received measurement response packet, calculating a response time from receiving the measurement packet to receiving the corresponding measurement response packet, and relating to the obtained communication reachability Storing the opposite server measurement result for storing information and the calculated response time in the measurement result database table;
The network failure detection method in the virtual machine environment according to any one of appendix 2 or 3, characterized in that:
(Appendix 5)
In the measurement result transmission step between physical servers,
Receiving an opposite server measurement packet for receiving a measurement result request addressed to a virtual server operating on the own physical server by the measurement packet from the other physical server;
Obtain the measurement result for the virtual server requested by the received measurement packet, obtain the opposite server measurement result from the measurement result database table,
Transmitting the opposite server response packet that returns the measurement response packet storing the acquired measurement result to the physical server that has transmitted the received measurement packet;
The network failure detection method in the virtual machine environment according to any one of appendices 2 to 4, characterized in that:
(Appendix 6)
In each physical server, monitor the operating status of the virtual server in the local physical server,
According to the operating status of the virtual server, update information related to communication reachability for each virtual server operating on the own physical server,
The network failure detection method in a virtual machine environment according to any one of appendices 1 to 5, characterized in that:
(Appendix 7)
Acquisition of measurement results in the physical server that manages multiple virtual servers that operate on each physical server that is connected to each other via a network and that acquire and manage information about communication reachability for each virtual server that operates on its own physical server And
In each of the physical servers, an inter-physical server measurement result acquisition unit that requests and acquires information on communication reachability regarding each of the virtual servers operating on the other physical server,
In each physical server, in accordance with a request from another physical server, obtains information on communication reachability for each virtual server running on the own physical server, and transmits an inter-physical server measurement result transmission unit that responds to the other physical server; ,
A network failure detection apparatus in a virtual machine environment characterized by comprising:
(Appendix 8)
The physical server measurement result acquisition unit includes:
A virtual server operating on the own physical server is identified by referring to a database table of virtual server information that is information on the virtual server, and a measurement result and response time regarding communication reachability to the virtual server operating on the own physical server And store the acquisition results in a database table of measurement results,
The inter-physical server measurement result acquisition unit identifies a virtual server operating on the other physical server by referring to a database table of server information in the network that is information on the virtual server operating on the other physical server, and Information on the communication reachability of the virtual server included in the other physical server is requested to the physical server in the measurement packet, and information on the communication reachability regarding the virtual server included in the other physical server is requested from the other physical server. Is stored as a measurement response packet, and the received information on communication reachability is stored in a database table of measurement results,
In the inter-physical server measurement result section, according to the request by the measurement packet received from the other physical server, the information on the communication reachability of the virtual server included in the own physical server is acquired from the measurement result database table, Reply as the measurement response packet;
The network failure detection apparatus in the virtual machine environment according to appendix 7, wherein
(Appendix 9)
The physical server measurement result acquisition unit includes:
With reference to the database table of the virtual server information, a virtual server information acquisition unit that identifies a virtual server operating on the own physical server;
A virtual server measurement packet transmitter for transmitting an echo request to the identified virtual server,
By monitoring an echo reply corresponding to the echo request, a virtual server response packet receiving unit for acquiring information on the communication reachability with respect to a virtual server operating on the own physical server;
A virtual server measurement result accumulation unit that accumulates information on the communication reachability for the acquired virtual server running on the acquired physical server in the measurement result database table;
The network failure detection apparatus in the virtual machine environment according to appendix 8, characterized by including:
(Appendix 10)
The inter-physical server measurement result acquisition unit
With reference to the database table of the server information in the network, the server information acquisition unit in the network that identifies the virtual server operating in the other physical server,
An opposite server measurement packet transmitter that transmits the measurement packet that requests the communication reachability information toward a virtual server that operates on the specified other physical server,
An opposite server response packet receiving unit that monitors and receives a response of the measurement response packet in which the communication reachability information transmitted by the target physical server with respect to the measurement packet is stored;
Obtaining information on the communication reachability from the received measurement response packet, calculating a response time from receiving the measurement packet to receiving the corresponding measurement response packet, and relating to the obtained communication reachability An opposing server measurement result storage unit that stores information and the calculated response time in the measurement result database table;
The network failure detection apparatus in the virtual machine environment according to any one of appendix 8 or 9, characterized by including:
(Appendix 11)
The inter-physical server measurement result transmitter
An opposite server measurement packet receiver that receives a measurement result request addressed to a virtual server operating on the own physical server by the measurement packet from the other physical server;
An opposite server measurement result acquisition unit that acquires a measurement result related to the virtual server requested by the received measurement packet from the database table of the measurement results;
The opposite server response packet transmitter that returns the measurement response packet storing the acquired measurement result to the physical server that has transmitted the received measurement packet;
11. A network failure detection apparatus in a virtual machine environment according to any one of appendixes 8 to 10, characterized by comprising:
(Appendix 12)
In each physical server, monitor the operating status of the virtual server in the local physical server,
According to the operating status of the virtual server, update information related to communication reachability for each virtual server operating on the own physical server,
The network failure detection apparatus in a virtual machine environment according to any one of appendices 7 to 11, characterized in that
(Appendix 13)
In-physical server measurement results that acquire and manage information related to communication reachability for each virtual server running on its own physical server in each physical server computer that is running multiple virtual servers and connected to each other via a network Processing,
In each physical server, inter-physical server measurement result acquisition processing (300-2) for requesting and acquiring information on communication reachability regarding each virtual server operating on the other physical server from the other physical server;
In each physical server, in accordance with a request from another physical server, obtains information on communication reachability for each virtual server operating on the own physical server, and transmits a measurement result between physical servers that responds to the other physical server. ,
A program for running
(Appendix 14)
In the physical server measurement result acquisition process,
A virtual server operating on the own physical server is identified by referring to a database table of virtual server information that is information on the virtual server, and a measurement result and response time regarding communication reachability to the virtual server operating on the own physical server And store the acquisition results in a database table of measurement results,
In the inter-physical server measurement result acquisition processing, the virtual server operating on the other physical server is identified by referring to a database table of server information in the network that is information on the virtual server operating on the other physical server, and the other Information on the communication reachability of the virtual server included in the other physical server is requested to the physical server in the measurement packet, and information on the communication reachability regarding the virtual server included in the other physical server is requested from the other physical server. Is stored as a measurement response packet, and the received information on communication reachability is stored in a database table of measurement results,
In the inter-physical server measurement result processing, according to the request by the measurement packet received from the other physical server, the information on the communication reachability of the virtual server included in the own physical server is acquired from the measurement result database table, Reply as the measurement response packet;
Program for executing processing.
(Appendix 15)
In the physical server measurement result acquisition process,
With reference to the database table of the virtual server information, a virtual server information acquisition process for specifying a virtual server operating on the own physical server;
A virtual server measurement packet transmission process for transmitting an echo request to the identified virtual server;
By monitoring an echo reply corresponding to the echo request, a virtual server response packet reception process for acquiring information related to the communication reachability to a virtual server operating on the own physical server,
A virtual server measurement result accumulation process for accumulating information on the communication reachability to the virtual server operating on the acquired physical server in the measurement result database table;
A program for running
(Appendix 16)
In the measurement result acquisition process between physical servers,
With reference to the database table of the server information in the network, server information acquisition processing in the network that identifies the virtual server operating on the other physical server,
Opposing server measurement packet transmission processing for transmitting the measurement packet for requesting the communication reachability information toward a virtual server operating on the specified other physical server,
Opposite server response packet reception processing for monitoring and receiving a reply of the measurement response packet storing the communication reachability information transmitted by the target physical server for the measurement packet;
Obtaining information on the communication reachability from the received measurement response packet, calculating a response time from receiving the measurement packet to receiving the corresponding measurement response packet, and relating to the obtained communication reachability Opposite server measurement result accumulation processing for accumulating information and the calculated response time in the measurement result database table;
A program for running
(Appendix 17)
In the measurement result transmission process between physical servers,
Opposing server measurement packet reception processing for receiving a measurement result request addressed to a virtual server operating on the own physical server by the measurement packet from the other physical server;
Opposing server measurement result acquisition processing for acquiring a measurement result related to the virtual server requested by the received measurement packet from the database table of the measurement result;
Opposing server response packet transmission processing for returning the measurement response packet storing the acquired measurement result to the physical server that has transmitted the received measurement packet;
A program for running
(Appendix 18)
In each physical server, monitor the operating status of the virtual server in the local physical server,
According to the operating status of the virtual server, update information related to communication reachability for each virtual server operating on the own physical server,
Program for executing processing.

201 Physical Server 202 Virtual Server 203 Hypervisor 204 Probe 205 Virtual Switch 206 NIC
207 Router 300-1 Measurement result acquisition unit in physical server 300-2 Measurement result acquisition unit between physical servers 300-3 Measurement result transmission unit between physical servers 301 Virtual server information acquisition unit 302 Virtual server measurement packet transmission unit 303 Virtual server Response packet reception unit 304 Virtual server measurement result storage unit 305 In-network server information acquisition unit 306 Counter server measurement packet transmission unit 307 Counter server response packet reception unit 308 Counter server measurement result storage unit 309 Counter server measurement packet reception unit 310 Counter server Measurement result acquisition unit 311 Counter server response packet transmission unit 312 Virtual server information 313 Network server information 314 Measurement result

Claims (14)

For each physical server that has multiple virtual servers and is connected to each other via a network, refer to the database table of the virtual server information, which is information related to the virtual server running on the own physical server, and run on the own physical server. A physical server measurement result acquisition step for identifying a virtual server to be acquired, acquiring a measurement result and a response time related to communication reachability to a virtual server operating on the own physical server, and storing the acquisition result in a database table of the measurement result; ,
In each of the physical servers, the virtual server operating on the other physical server is identified by referring to a database table of server information in the network that is information on the virtual server operating on the other physical server, and the identified other physical server A measurement packet requesting communication reachability information of the virtual server operating on the specified other physical server is transmitted to the virtual server operating on the specified physical server, and the target physical server transmits the measurement packet. and the communication reachability information to monitor the response measurement response packet stored received was obtains the information about the communication reachability from measuring the received response packets, corresponding from the transmission of the measurement packet The response time until the measurement response packet is received is calculated, and the acquired information regarding communication reachability The response time was pre the calculated, and the physical servers inter measurement result acquisition step of storing in a database table of the measurement result,
In each of the physical servers, in accordance with the request by the measurement packet received from the other physical server, information on the communication reachability of the virtual server included in the own physical server is acquired from the database table of the measurement result, and the measurement response packet As a physical server inter-physical server measurement result transmission step to return to the other physical server as
Network failure detection method in a virtual machine environment characterized in that the run. For each physical server that has multiple virtual servers and is connected to each other via a network, refer to the database table of the virtual server information, which is information related to the virtual server running on the own physical server, and run on the own physical server. A physical server measurement result acquisition step for identifying a virtual server to be acquired, acquiring a measurement result and a response time related to communication reachability to a virtual server operating on the own physical server, and storing the acquisition result in a database table of the measurement result; ,
In each of the physical servers, the virtual server operating on the other physical server is identified by referring to the database table of the server information in the network that is information on the virtual server operating on the other physical server, Requesting information related to communication reachability of the virtual server included in the other physical server in a measurement packet, and sending information related to communication reachability related to the virtual server included in the other physical server from the other physical server to the measurement response packet. The physical server inter-physical server measurement result acquisition step for storing the received information about the communication reachability in a measurement result database table;
In each physical server, wherein by measuring the packet receives the measurement result request addressed to virtual servers running on its own physical server, measurement results relating to the virtual server that is requested by the measurement the received packet from the another physical server and retrieved from the database table of the measurement results, the measurement response packet storing the acquired measurement result, and the physical servers inter measurement result transmission step of replying toward the physical server that has transmitted the measurement the received packet ,
Network failure detection method in a virtual machine environment characterized in that the run. In the physical server internal measurement result acquisition step,
Above with reference to a database table of the virtual server information, the specified virtual server running in its own physical server,
Wherein for a particular virtual server, sends an echo request,
By monitoring the echo reply corresponding to the echo request, to obtain information about the communication reachability to the virtual servers running on the host physical server,
According to claim 1 or 2, characterized in that to perform the accumulating information about the communication reachability to the virtual servers running on the obtained the own physical server in a database table of the measurement result Network failure detection method in virtual machine environment. In each physical server, monitor the operating status of the virtual server in the local physical server,
According to the operating status of the virtual server, update information related to communication reachability for each virtual server operating on the own physical server,
The network failure detection method in a virtual machine environment according to any one of claims 1 to 3 . For each physical server that has multiple virtual servers and is connected to each other via a network, refer to the database table of the virtual server information, which is information related to the virtual server running on the own physical server, and run on the own physical server. identifying a virtual server, the self physical servers running the acquired measurement result and response time to a communication reachability for each virtual server, the acquisition result of measurement in the physical server accumulates the measurement result database tables result acquisition to And
In each of the physical servers, the virtual server operating on the other physical server is identified by referring to a database table of server information in the network that is information on the virtual server operating on the other physical server, and the identified other physical server Transmitting a measurement packet for requesting communication reachability information of each of the virtual servers operating on the specified other physical server toward the virtual server operating in the specified physical server Monitoring and receiving a reply of the measurement response packet in which the communication reachability information transmitted is stored , obtaining the communication reachability information from the received measurement response packet, and transmitting the measurement packet Calculate the response time until receiving the corresponding measurement response packet, and relate to the acquired communication reachability The response time information and the said calculated, and the physical servers between the measurement result acquisition section for storing in a database table of the measurement result,
In each physical server, the following other physical requirements by the server the measurement packet received from, acquires information regarding the communication reach of virtual servers included in the own physical server from a database table of the measurement result, the measurement response packet As an inter-physical server measurement result transmission unit that returns to the other physical server as
A network failure detection apparatus in a virtual machine environment characterized by comprising: By referring to a database table of virtual server information, which is information related to the virtual server operating on the own physical server, in the computer of each physical server in which a plurality of virtual servers are operated and connected to each other via a network, the own physical server The virtual server operating in the physical server is identified, the measurement result and response time regarding the communication reachability for each virtual server operating on the own physical server is acquired, and the acquisition result is stored in the database table of the measurement result. Acquisition process,
In the computer of each physical server , referring to the database table of the server information in the network that is information related to the virtual server operating on the other physical server, the virtual server operating on the other physical server is specified, and the specified A measurement packet that requests communication reachability information of each virtual server that operates on the specified other physical server is transmitted to a virtual server that operates on the other physical server, and is targeted for the measurement packet Monitors and receives the response of the measurement response packet in which the communication reachability information transmitted by the physical server is stored , acquires information on the communication reachability from the received measurement response packet, and transmits the measurement packet And calculating the response time until receiving the corresponding measurement response packet. The information regarding the communication reachability and the calculated response time, and the measurement result between physical servers measurements accumulated in a database table of results acquisition process,
Wherein in said computer for each physical server, the following other physical requirements by the server the measurement packet received from, acquires information regarding the communication reach of virtual servers included in the own physical server from a database table of the measurement result, the A measurement result transmission process between physical servers that returns a measurement response packet to the other physical server;
A program for running For each physical server that has multiple virtual servers and is connected to each other via a network, refer to the database table of the virtual server information, which is information related to the virtual server running on the own physical server, and run on the own physical server. A measurement result acquisition unit in a physical server that identifies a virtual server to acquire, acquires a measurement result and a response time related to communication reachability to a virtual server operating on the own physical server, and stores the acquisition result in a database table of the measurement result; ,
In each of the physical servers, the virtual server operating on the other physical server is identified by referring to the database table of the server information in the network that is information on the virtual server operating on the other physical server, Requesting information related to communication reachability of the virtual server included in the other physical server in a measurement packet, and sending information related to communication reachability related to the virtual server included in the other physical server from the other physical server to the measurement response packet. The physical server inter-physical server measurement result acquisition unit that stores the received information on the communication reachability in the measurement result database table;
Each physical server receives a measurement result request addressed to a virtual server operating on the own physical server by the measurement packet from the other physical server, and a measurement result related to the virtual server requested by the received measurement packet The inter-physical server measurement result transmitting unit that returns the measurement response packet storing the acquired measurement result to the physical server that transmitted the received measurement packet; ,
A network failure detection apparatus in a virtual machine environment characterized by comprising: By referring to a database table of virtual server information, which is information related to the virtual server operating on the own physical server, in the computer of each physical server in which a plurality of virtual servers are operated and connected to each other via a network, the own physical server Acquire the measurement result and response time related to communication reachability to the virtual server running on the local physical server, identify the virtual server running in the local physical server, and store the obtained result in the database table of the measurement result Processing,
In the computer of each physical server, the virtual server operating on the other physical server is identified by referring to a database table of server information in the network that is information related to the virtual server operating on the other physical server, and the other physical server In response to the request for information on the communication reachability of the virtual server included in the other physical server in a measurement packet, the information on the communication reachability of the virtual server included in the other physical server from the other physical server, A measurement result acquisition process between physical servers that is received as a measurement response packet and stores the received communication reachability information in a database table of measurement results;
In the computer of each physical server, the virtual server requested by the received measurement packet is received by the measurement packet addressed to the virtual server operating on the own physical server by the measurement packet from the other physical server The measurement result between the physical servers that obtains the measurement result related to the measurement result from the database table of the measurement result and returns the measurement response packet storing the obtained measurement result to the physical server that transmitted the received measurement packet. Send processing,
A program for running For each physical server that has multiple virtual servers running and connected to each other via a network, obtain information on communication reachability for each virtual server running on its own physical server,
In each of the physical servers, a measurement packet for requesting communication reachability information regarding each virtual server operating on the other physical server is transmitted to each virtual server operating on a physical server other than the own physical server,
In each of the physical servers, a measurement response packet storing communication reachability information about the virtual server, transmitted as a response to the measurement packet by another physical server in which the virtual server that is the destination of the measurement packet is operating, is received. ,
In each physical server, from the received measurement response packet, obtain information on communication reachability for each virtual server running on the other physical server,
A network failure detection method characterized by the above. For each physical server that has multiple virtual servers running and connected to each other via a network, obtain information on communication reachability for each virtual server running on its own physical server,
In each of the physical servers, information on communication reachability regarding each virtual server running on a physical server other than the own physical server is requested and acquired from the other physical server,
In each of the physical servers, receiving from the other physical server a measurement packet that requests communication reachability information related to the virtual server operating on the own physical server that is addressed to the virtual server operating on the own physical server,
In each of the physical servers, a measurement response packet storing communication reachability information related to the virtual server requested by the received measurement packet is transmitted to the other physical server that has transmitted the received measurement packet.
A network failure detection method in a virtual machine environment. In each physical server in which a plurality of virtual servers are operated and connected to each other via a network, a first acquisition unit that acquires information on communication reachability for each virtual server operating on the own physical server;
Measurement in which each physical server transmits a measurement packet for requesting communication reachability information regarding each virtual server operating on the other physical server to each virtual server operating on a physical server other than the own physical server A packet transmitter;
Each physical server receives a measurement response packet storing communication reachability information related to the virtual server, which is transmitted as a response to the measurement packet by another physical server in which the virtual server that is the destination of the measurement packet is operating. A measurement response packet receiver;
In each of the physical servers, from the received measurement response packet, a second acquisition unit that acquires information on communication reachability regarding each virtual server operating on the other physical server;
A network failure detection apparatus in a virtual machine environment characterized by comprising: In each physical server in which a plurality of virtual servers are operated and connected to each other via a network, a first acquisition unit that acquires information on communication reachability for each virtual server operating on the own physical server;
In each of the physical servers, a second acquisition unit that requests and acquires information on communication reachability regarding each virtual server operating on a physical server other than the own physical server from the other physical server;
In each of the physical servers, a receiving unit that receives, from the other physical server, a measurement packet that requests communication reachability information related to the virtual server operating on the physical server that is addressed to the virtual server operating on the local physical server When,
In each of the physical servers, a transmission for transmitting a measurement response packet storing communication reachability information related to the virtual server requested by the received measurement packet toward the other physical server that has transmitted the received measurement packet. And
A network failure detection apparatus in a virtual machine environment characterized by comprising: In each physical server computer that is operated by a plurality of virtual servers and connected to each other via a network, processing for obtaining information on communication reachability for each virtual server operating on the own physical server;
In the computer of each physical server, a measurement packet for requesting communication reachability information regarding each virtual server running on the other physical server is addressed to each virtual server running on another physical server other than the own physical server. Processing to send,
In the computer of each physical server, a measurement response packet storing communication reachability information related to the virtual server, which is transmitted as a response to the measurement packet by another physical server in which the virtual server that is the destination of the measurement packet is operating Processing to receive
In the computer of each physical server, from the received measurement response packet, processing for obtaining information on communication reachability for each virtual server operating on the other physical server;
A program for running In each physical server computer that is operated by a plurality of virtual servers and connected to each other via a network, processing for obtaining information on communication reachability for each virtual server operating on the own physical server;
In the computer of each physical server, a process for requesting and acquiring information about communication reachability regarding each virtual server operating on a physical server other than the own physical server from the other physical server;
The computer of each physical server receives a measurement packet requesting communication reachability information related to the virtual server running on the own physical server that is addressed to the virtual server running on the own physical server from the other physical server Processing to
In the computer of each physical server, a measurement response packet storing communication reachability information related to the virtual server requested by the received measurement packet is directed to the other physical server that has transmitted the received measurement packet. Processing to send,
A program for running