Tenant redundancy system and tenant redundancy method

Abstract

A packet relay system (1) comprises: a relay device (41) that constitutes an autonomous system (72) and that is provided with a BGP unit (414) connected to an external device so as to perform path exchange; and a tenant provided with a plurality of containers (6a, 6b) that constitute an autonomous system (73) different from the autonomous system (72), that are connected to the relay device (41), and that are provided with BGP units (67) for exchanging path information and self-monitoring units (66) for monitoring network interfaces (61, 62).

Description

Technical Field

[0001] The present invention relates to a packet relay system that performs packet transfer in a network device and a packet relay method.

Background Art

[0002] With the development of SDN (Software Defined Network) technology and NFV (Network Function Virtualization) technology, it has been required to flexibly connect between a mobile terminal and a server on the cloud. Therefore, a service has emerged in which a network user controls a packet transfer destination for a packet relay device on demand to realize packet transfer between a mobile terminal and a server.

[0003] As an example of the conventional technology, a configuration is conceivable in which tunneling protocols such as IPsec (Internet Protocol Security), VXLAN, and GRE (Generic Routing Encapsulation) are used, and routing software such as Open vSwitch is used for the packet relay device.

[0004] When a mobile terminal and a server on the cloud communicate packets with each other, it is common to realize the service by using the IP address assigned to the packet relay device as the packet transmission destination for both. As a technology supporting such a service, for example, GRE described in Non-Patent Document 1 is available.

Prior Art Documents

Non-Patent Documents

[0005]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] In a system where tenants are composed of containers and virtual machines, the containers and virtual machines need to be stopped or relocated due to various factors. To address this, it is necessary to make the tenants redundant and, furthermore, promptly propagate the information of the tenant's network interface to the relay device as the containers and virtual machines are relocated.

[0007] Cases where containers and virtual machines need to be stopped or relocated include, for example, process failures of applications on containers, forced stops from the infrastructure due to overloading of containers, and forced stops from the infrastructure due to overloading of other containers on the same server.

[0008] Therefore, an object of the present invention is to promptly propagate the information of the tenant's network interface to the relay device.

Means for Solving the Problems

[0009] To solve the above problems, the tenant redundancy system of the present invention constitutes a first autonomous system and is connected to an external device to exchange information routes to do It includes a relay device having a route exchange unit, constitutes a second autonomous system different from the first autonomous system, is connected to the relay device, has a route exchange unit for exchanging route information, and a plurality of A tenant including a plurality of control units each having a self-monitoring unit for monitoring interfaces. Other means will be described in the mode for carrying out the invention.

Effect of the Invention

[0010] According to the present invention, it becomes possible to quickly propagate the information of the tenant's network interface to the relay device.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6A

Figure 6B

Figure 7A

Figure 7B

Figure 7C

Figure 8A

Figure 8B

Figure 9A

Figure 9B

Figure 10A

Figure 10B

Figure 10C

Figure 11A

Figure 11B

Figure 11C

Figure 12

Figure 13

Figure 14A

Figure 14B

Figure 14C

Figure 15A

Figure 15B

Figure 16A

Figure 16B

Figure 17A

Figure 17B

Figure 18A

Figure 18B

Figure 19

Figure 20A

Figure 20B

Figure 21A

Figure 21B

Embodiments for Carrying Out the Invention

[0012] Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to each drawing. 《Comparative Example》 FIG. 1 is a configuration diagram of a packet relay system 1A of a comparative example. The packet relay system 1A treats an IP address as a logical address, and controls each device in the packet relay system 1A to perform appropriate routing based on the logical address. The packet relay system 1A includes a container 6 constructed on a server 9 and relay devices 41 and 42 arranged before and after it. This packet relay system 1A is connected to a device 31 via a tunnel 81 and connected to a device 32 via a tunnel 82.

[0013] The container 6 is a control unit of a tenant configured as a Pod on kubernetes (registered trademark). The control unit of the tenant is single and not redundant. The container 6 performs transfer processing based on a logical address independent of the physical addresses of the relay devices 41 and 42. A plurality of tap devices 63 and 64 that terminate a tunneling protocol are constructed in the container 6. These tap devices 63 and 64 are virtual network devices.

[0014] Devices 31 and 32 are routers that are connected to relay devices 41 and 42 and exchange packets with each other via the packet relay system 1A. Devices 31 and 32 are connected to a management device (not shown) of the packet relay system 1 and instruct changes in the settings of communication between terminals.

[0015] Relay devices 41 and 42 are respectively connected to devices 31 and 32 via a network capable of transmitting and receiving IP packets. These relay devices 41 and 42 connect the container 6 to the outside. Relay device 41 includes network interfaces 411 and 412, and a FIB (Forwarding Information Base: routing control table) 413. Network interface 411 is an interface for connecting this relay device 41 to an external device 31. Network interface 412 is an interface for connecting this relay device 41 to the container 6. FIB 413 is a routing control table referred to when transferring packets passing through this relay device 41.

[0016] Similarly, relay device 42 includes network interfaces 421 and 422, and a FIB 423. Network interface 422 is an interface for connecting this relay device 42 to an external device 32. Network interface 421 is an interface for connecting this relay device 42 to the container 6. FIB 423 is a routing control table referred to when transferring packets passing through this relay device 42.

[0017] As a basic operation, relay devices 41 and 42 have the role of sending packets of the tunneling protocol transmitted from devices 31 and 32 to the container 6 and sending packets of the tunneling protocol transmitted from the container 6 to devices 31 and 32.

[0018] Container 6 has network interfaces 61 and 62 connected to relay devices 41 and 42 via a network capable of transmitting and receiving IP packets. The container 6 also has a FIB 65 and tap devices (denoted as "tap" in the figure) 63 and 64 that terminate the transfer protocol. These two tap devices 63 and 64 are virtual network devices that terminate the tunneling protocol. The FIB 65 is a routing control table referred to when transferring packets passing through this container 6.

[0019] As a basic operation, when a packet of the tunneling protocol transmitted from device 31 arrives at network interface 61 of container 6, container 6 terminates the tunneling protocol with tap device 63. Then, if necessary, container 6 performs encapsulation with the other tap device 64 and transfers it to the opposite device 32.

[0020] When a packet of the tunneling protocol transmitted from device 32 arrives at network interface 62 of container 6 and the tunneling protocol is received, container 6 terminates the tunneling protocol with tap device 64. Then, if necessary, container 6 performs encapsulation with the other tap device 63 and transfers it to the opposite device 31.

[0021] 《This Embodiment》 Figure 2 is a configuration diagram of the packet relay system 1 according to this embodiment. The packet relay system 1 connects and controls GRE tunnels from relay devices 41 and 42 to a loopback IP with the same IP address. The packet relay system 1 includes a plurality of containers 6a and 6b constructed in server 9, relay devices 41 and 42 arranged before and after them, and a control device 11. This packet relay system 1 is connected to device 31 via tunnel 81 and connected to device 32 via tunnel 82. This packet relay system 1 is a tenant redundancy system that makes tenants redundant.

[0022] The control device 11 realizes the survival check of the container and the automatic reconnection of the GRE tunnel by setting each part of the packet relay system 1. Devices 31 and 32 are routers that are connected to relay devices 41 and 42 and exchange packets with each other via the packet relay system 1, and respectively constitute autonomous systems 71 and 75 which are the third autonomous systems.

[0023] Device 31 is assigned an IP address 311 and has a BGP (Border Gateway Protocol) function by including a BGP part 312. Device 31 has an AS (autonomous system) number = 10 different from other components as a BGP setting. Thereby, device 31 constitutes autonomous system 71.

[0024] Also, device 31 enables the EBGP Multihop setting and establishes peers not only with the adjacent AS = 20 but also with AS = 21. Then, device 31 enables BFD (Bidirectional Forwarding Detection), checks the survival status of the transfer path between adjacent autonomous systems, and when it detects a failure at high speed, notifies the routing protocol.

[0025] Device 32 is assigned an IP address 321 and has a BGP function by including a BGP part 322. The BGP part 322 is a path exchange part that is connected to an external device and performs path exchange. Device 32 has an AS (autonomous system) number = 30 different from other components as a BGP setting. Thereby, device 32 constitutes autonomous system 75.

[0026] Also, device 32 enables the EBGP Multihop setting and establishes peers not only with autonomous system 74 with an adjacent AS number = 22 but also with autonomous system 73 with AS number = 21. Then, device 32 enables BFD, checks the survival status of the transfer path between adjacent autonomous systems, and when it detects a failure at high speed, notifies the routing protocol.

[0027] The relay devices 41 and 42 are each connected to the devices 31 and 32 via a network capable of transmitting and receiving IP packets. These relay devices 41 and 42 connect the container 6a or 6b to the outside.

[0028] The relay device 41 has a BGP function by including an FIB 413 and a BGP unit 414. The BGP unit 414 is a route exchange unit that is connected to an external device and performs route exchange. The relay device 41 has an AS (autonomous system) number = 20 different from other components as a BGP setting. Thereby, the relay device 41 constitutes the autonomous system 72 which is the first autonomous system.

[0029] Then, the relay device 41 enables BFD, checks the survival status of the transfer path between adjacent autonomous systems, and when a failure is detected at high speed, notifies the routing protocol. Based on the route information exchanged between BGP peers, the BGP unit 414 determines the routing destination of the packet addressed to the logical address 68 for the container 6a. The FIB 413 is a route control table referred to when the relay device 41 transfers a packet passing through itself.

[0030] The relay device 42 has a BGP function by including an FIB 423 and a BGP unit 424. The BGP unit 424 is a route exchange unit that is connected to an external device and performs route exchange. The relay device 42 has an AS (autonomous system) number = 22 different from other components as a BGP setting. Thereby, the relay device 42 constitutes the autonomous system 74.

[0031] Then, the relay device 42 enables BFD, checks the survival status of the transfer path between adjacent autonomous systems, and when a failure is detected at high speed, notifies the routing protocol. Based on the route information exchanged between BGP peers, the BGP unit 424 determines the routing destination of the packet addressed to the logical address 69 for the container 6a. The FIB 423 is a route control table referred to when the relay device 42 transfers a packet passing through itself.

[0032] Containers 6a and 6b are tenant control units, and perform transfer processing based on logical addresses 68 and 69 that are independent of the physical addresses of the relay devices 41 and 42. Note that the tenant control unit is not limited to this, and may be configured as a virtual machine.

[0033] Containers 6a and 6b have network interfaces 61 and 62, and have a BGP (Border Gateway Protocol) function by including a BGP unit 67. The BGP unit 67 is a path exchange unit that is connected to an external device and performs path exchange. Containers 6a and 6b have an AS (Autonomous System) number = 21 different from other components as a BGP setting. Thereby, containers 6a and 6b constitute an autonomous system 73 that is a second autonomous system different from the first autonomous system.

[0034] Then, containers 6a and 6b activate BFD, check the survival status of the transfer path between adjacent autonomous systems, and when a failure is detected at high speed, notify the routing protocol.

[0035] Containers 6a and 6b are constructed with a plurality of tap devices 63 and 64 that terminate the tunneling protocol, and further have a self-monitoring unit 66. The tap devices 63 and 64 are virtual network devices.

[0036] Containers 6a and 6b have a logical address 68 for notifying the device 31, and receive incoming packets at the network interface 61. Further, containers 6a and 6b have a logical address 69 for notifying the device 32, and receive incoming packets at the network interface 62.

[0037] The self - monitoring unit 66 monitors the network interfaces 61 and 62 to check whether there is any abnormality in either of the network interfaces. When the self - monitoring unit 66 detects an abnormality, it abnormally terminates the container (control unit) itself. A tunnel 81 is constructed between the container 6a and the device 31 in FIG. 2. And a tunnel 82 is constructed between the container 6a and the device 32.

[0038] As a basic operation, the relay devices 41 and 42 transmit the tunneling protocol packets sent from the devices 31 and 32 to the container 6, and transmit the tunneling protocol packets sent from the container 6 to the devices 31 and 32. The relay device 41 determines either of the containers 6a and 6b as the routing destination of the packet based on the route information exchanged by the BGP unit 414. The relay device 42 determines either of the containers 6a and 6b as the routing destination of the packet based on the route information exchanged by the BGP unit 424.

[0039] As a basic operation, when a tunneling protocol packet sent from the device 31 arrives at the network interface 61, the container 6 terminates the tunneling protocol with the tap device 63. Then, if necessary, the container 6 encapsulates with the other tap device 64 and transfers it to the opposite device 32.

[0040] Furthermore, when a tunneling protocol packet sent from the device 32 arrives at the network interface 62 and the tunneling protocol is received, the container 6 terminates the tunneling protocol with the tap device 64. Then, if necessary, the container 6 encapsulates with the other tap device 63 and transfers it to the opposite device 31.

[0041] 《The First Embodiment》 FIG. 3 is a configuration diagram of the packet relay system 1 according to the first embodiment. The packet relay system 1 includes a tenant 5 having a plurality of containers 6a and 6b, and protocol units 43 and 44 arranged before and after the tenant 5. The packet relay system 1 connects routers 33 and 34 arranged before and after the protocol units 43 and 44, and further connects terminals 21 and 22 arranged before and after the routers 33 and 34. The container 6a is connected to the router 33 via a tunnel 81. The containers 6a and 6b constitute an autonomous system 73.

[0042] The protocol units 43 and 44 correspond to the relay devices 41 and 42 in FIG. 2. The routers 33 and 34 correspond to the devices 31 and 32 in FIG. 2. The plurality of containers 6a and 6b constitute a second autonomous system, and are control units including a BGP unit 67 that is connected to the relay devices 41 and 42 and exchanges route information, and a self-monitoring unit 66 that monitors network interfaces 61 and 62.

[0043] The link connecting the terminal 21 and the router 33 has an address range of 192.168.120.0 / 24. The network interface on the right side of the terminal 21 is assigned an address of 192.168.120.78. The network interface on the left side of the router 33 is assigned an address of 192.168.120.88.

[0044] The link connecting the router 33 and the protocol unit 43 has an address range of 10.38.215.0 / 24. The network interface on the right side of the router 33 is assigned an address of 10.38.215.23. The network interface on the left side of the protocol unit 43 is assigned an address of 10.38.215.32.

[0045] The link connecting the protocol unit 43 and the containers 6a, 6b has an address range of 192.168.130.0 / 24. The network interface on the right side of the protocol unit 43 is assigned the address 192.168.130.42. The network interface on the left side of the container 6a is assigned the address 192.168.130.11. The network interface on the left side of the container 6b is assigned the address 192.168.130.12.

[0046] The link connecting the containers 6a, 6b and the protocol unit 44 has an address range of 192.168.110.0 / 24. The network interface on the right side of the container 6a is assigned the address 192.168.110.11. The network interface on the right side of the container 6b is assigned the address 192.168.110.12. The network interface on the left side of the protocol unit 44 is assigned the address 192.168.110.23.

[0047] The link connecting the protocol unit 44 and the router 34 has an address range of 192.168.170.0 / 24. The network interface on the right side of the protocol unit 44 is assigned the address 192.168.170.23. The network interface on the left side of the router 34 is assigned the address 192.168.170.24.

[0048] The link connecting the router 34 and the terminal 22 has an address range of 192.168.180.0 / 24. The network interface on the right side of the router 34 is assigned the address 192.168.180.24. The network interface on the left side of the terminal 22 is assigned the address 192.168.180.25.

[0049] A tunnel 81 is established between the network interface on the right side of the router 33 and the network interface on the left side of the container 6a. In addition to the start IP address 10.38.215.23 / 24 of the GRE tunnel, the GRE connection IP address 50.1.1.1 / 32 is assigned to the network interface on the right side of the router 33. In addition to the end IP address 40.1.1.1 / 24 of the GRE tunnel and the AnyCast IP, the GRE connection IP address 60.1.1.1 / 32 is assigned to the network interface on the left side of the container 6a. And in addition to the end IP address 40.1.1.1 / 24 of the GRE tunnel and the AnyCast IP, the GRE connection IP address 70.1.1.1 / 32 is assigned to the network interface of the container 6b.

[0050] Figure 4 is a flowchart showing the initial setting process. First, the control device 11 sets the GRE connection IP address between the router 33 and the container (step S10). Then, the control device 11 sets the AnyCast IP for the containers 6a and 6b (step S11), and sets the GRE connection IP address between the containers 6a and 6b and the router 33 (step S12).

[0051] Next, when the control device 11 performs GRE static setting on the container (step S13), it starts monitoring the survival status of the path with BFD described above (step S14), and ends the process of Figure 4. Thus, the initial setting of the packet relay system 1 is performed.

[0052] Figure 5 is a flowchart showing the self-monitoring process. The standby container checks the survival of the active container (step S20). Then, in step S21, if the standby container does not detect the stop of the active container (No), it returns to step S20. In step S21, if the standby container detects the stop of the active container (Yes), it proceeds to step S22. In step S22, when the standby container reconnects the right GRE tunneling and the left GRE tunneling and transitions itself to the active system (step S23), the process of FIG. 5 ends.

[0053] FIG. 6A is a diagram showing the BGP setting for the left protocol unit 43. The control device 11 acquires the current BGP setting by executing the "router bgp 20" command for the protocol unit 43. Note that the second argument of the router command is the AS number of the target device.

[0054] The control device 11 advertises the route information within its own AS as a BGP route by executing the "network 10.38.215.0 / 24" command for the protocol unit 43. Note that the first argument of the network command is the network address that advertises the route information within its own AS as a BGP route.

[0055] The control device 11 performs the setting of route filtering in BGP using a prefix-list by executing the "neighbor 192.168.130.11 remote-as 21" command for the protocol unit 43. Note that the first argument of the neighbor command is the network address of the target device to be advertised. The third argument of the neighbor command indicates the AS number.

[0056] The control device 11 performs the setting of route filtering in BGP using a prefix-list by executing the "neighbor 192.168.130.12 remote-as 21" command for the protocol unit 43.

[0057] The control device 11 advertises the route information within its own AS as a BGP route by executing the "network 192.168.130.0 / 24" command for the protocol unit 43.

[0058] The control device 11 executes the command "neighbor 10.38.215.23 remote-as 10" on the protocol unit 43 to perform the setting of route filtering in BGP using the prefix-list.

[0059] Then, the control device 11 executes the command "timers bgp 3 9" on the protocol unit 43 to set the BGP keepalive timer and holdtime timer. Specifically, the keepalive timer is set to 3 seconds and the holdtime timer is set to 9 seconds.

[0060] FIG. 6B is a diagram showing the routing table generated by the protocol unit 43 on the left side. The control device 11 executes the command "show ip bgp" on the protocol unit 43 to obtain the generated routing table. An example of the routing table of the execution result is shown below.

[0061] Network Next Hop ····· 40.1.1.0 / 24 192.168.130.11 40.1.1.0 / 24 192.168.130.12 50.1.1.0 / 32 10.38.215.23 60.1.1.0 / 32 192.168.130.11 70.1.1.0 / 32 192.168.130.12

[0062] FIG. 7A is a diagram showing the BGP setting for the router 33 on the left side. The control device 11 executes the command "router bgp 10" on the router 33 to obtain the current BGP setting.

[0063] The control device 11 advertises the route information within its own AS as a BGP route by executing the "network 192.168.120.0 / 24" command on the router 33.

[0064] The control device 11 performs the setting of route filtering in BGP using a prefix-list by executing the "neighbor 10.38.215.32 remote-as 20" command on the protocol unit 43.

[0065] The control device 11 advertises the route information within its own AS as a BGP route by executing the "network 50.1.1.1 / 32" command on the router 33.

[0066] The control device 11 performs the setting of route filtering in BGP using a prefix-list by executing the "neighbor 60.1.1.1 remote-as 21" command on the protocol unit 43.

[0067] The control device 11 performs the setting of route filtering in BGP using a prefix-list by executing the "neighbor 60.1.1.1 ebgp-multihop 255" command on the protocol unit 43.

[0068] The control device 11 performs the setting of route filtering in BGP using a prefix-list by executing the "neighbor 70.1.1.1 remote-as 21" command on the protocol unit 43.

[0069] The control device 11 performs the setting of route filtering in BGP using a prefix-list by executing the "neighbor 70.1.1.1 ebgp-multihop 255" command on the protocol unit 43.

[0070] FIG. 7B is a diagram showing a routing table generated by the left router 33. The control device 11 acquires the generated routing table by executing the "show ip bgp" command on the router 33. An example of the routing table of the execution result is shown below.

[0071] Network Next Hop 192.168.110.0 60.1.1.1 192.168.130.0 60.1.1.1 192.168.170.0 60.1.1.1 192.168.180.0 60.1.1.1 40.1.1.0 / 24 60.1.1.1

[0072] FIG. 7C is a diagram showing static settings and GRE settings for the left router 33. The control device 11 executes the "ip route 40.1.1.0 / 24 10.38.215.32" command on the router 33. This is a static setting for the router 33. Note that an IP address must be specified as the third argument here. The control device 11 executes the "ip route 60.1.1.1 / 32 gre" command on the router 33. This is a static setting for the router 33. The control device 11 executes the "ip route 70.1.1.1 / 32 gre" command on the router 33. This is a static setting for the router 33. The control device 11 executes the "ip link add gre type gre local 10.38.215.23 remote 40.1.1.1" command on the router 33. This is a GRE setting. The control device 11 executes the "ip addr add 50.1.1.1 / 32 dev gre" command on the router 33. This is a GRE setting.

[0073] FIG. 8A is a diagram showing the BGP settings for the protocol unit 44 on the right side. The control device 11 executes the "router bgp 22" command for the protocol unit 44 to obtain the current BGP settings. Note that the second argument of the router command is the AS number of the target device.

[0074] The control device 11 executes the "network 192.168.170.0 / 24" command for the protocol unit 44 to advertise the route information within its own AS as a BGP route.

[0075] The control device 11 executes the "neighbor 192.168.130.11 remote-as 21" command for the protocol unit 44 to perform the route filtering setting in BGP using the prefix-list.

[0076] The control device 11 executes the "neighbor 192.168.130.12 remote-as 21" command for the protocol unit 44 to perform the route filtering setting in BGP using the prefix-list.

[0077] The control device 11 executes the "network 192.168.130.0 / 24" command for the protocol unit 44 to advertise the route information within its own AS as a BGP route.

[0078] The control device 11 executes the "neighbor 10.38.215.23 remote-as 10" command for the protocol unit 44 to perform the route filtering setting in BGP using the prefix-list.

[0079] Then, the control device 11 sets the BGP keepalive timer and holdtime timer by executing the "timers bgp 3 9" command on the protocol unit 44. Specifically, the keepalive timer is set to 3 seconds and the holdtime timer is set to 9 seconds.

[0080] FIG. 8B is a diagram showing the routing table generated by the protocol unit 44 on the right side. The control device 11 obtains the generated routing table by executing the "show ip bgp" command on the protocol unit 44. An example of the routing table of the execution result is shown below.

[0081] Network Next Hop 192.168.130.0 192.168.110.11 192.168.130.0 192.168.110.12 192.168.180.0 192.168.170.24 192.168.120.0 192.168.110.11 192.168.120.0 192.168.110.12 10.38.215.0 / 24 192.168.110.11 10.38.215.0 / 24 192.168.110.12 40.1.1.0 / 24 192.168.110.11 50.1.1.0 / 24 192.168.110.11 60.1.1.0 / 24 192.168.110.11 70.1.1.0 / 24 192.168.110.12

[0082] FIG. 9A is a diagram showing the BGP settings for the router 34 on the right side. The control device 11 obtains the current BGP settings by executing the "router bgp 30" command on the router 34. Note that the second argument of the router command is the AS number of the target device.

[0083] The control device 11 advertises the route information within its own AS as a BGP route to the router 34 by executing the "network 192.168.180.0 / 24" command.

[0084] The control device 11 configures route filtering in BGP using a prefix-list by executing the "neighbor 192.168.170.23 remote-as 22" command on the router 34.

[0085] FIG. 9B is a diagram showing the routing table generated by the router 34 on the right side. The control device 11 obtains the generated routing table by executing the "show ip bgp" command on the router 34. An example of the routing table of the execution result is shown below.

[0086] Network Next Hop 192.168.110.0 192.168.170.23 192.168.130.0 192.168.170.23 10.38.215.0 / 24 192.168.170.23 192.168.120.0 192.168.170.23 40.1.1.0 / 24 192.168.170.23 50.1.1.0 / 24 192.168.170.23 60.1.1.0 / 24 192.168.170.23 70.1.1.0 / 24 192.168.170.23

[0087] FIG. 10A is a diagram showing the BGP settings for the upper container 6a. The control device 11 obtains the current BGP settings by executing the "router bgp 21" command on the upper container 6a. Note that the second argument of the router command is the AS number of the target device.

[0088] The control device 11 advertises the route information within its own AS as a BGP route by executing the "network 192.168.110.0 / 24" command on the container 6a.

[0089] The control device 11 configures route filtering in BGP using a prefix-list by executing the "neighbor 192.168.130.42 remote-as 20" command on the container 6a.

[0090] The control device 11 advertises the route information within its own AS as a BGP route by executing the "network 192.168.130.0 / 24" command on the container 6a.

[0091] The control device 11 configures route filtering in BGP using a prefix-list by executing the "neighbor 192.168.110.23 remote-as 22" command on the container 6a.

[0092] The control device 11 advertises the route information within its own AS as a BGP route by executing the "network 40.1.1.0 / 24" command on the container 6a.

[0093] The control device 11 advertises the route information within its own AS as a BGP route by executing the "network 60.1.1.1 / 32" command on the container 6a.

[0094] The control device 11 performs the setting of route filtering in BGP using a prefix-list by executing the command "neighbor 50.1.1.1 remote-as 10" on the container 6a.

[0095] The control device 11 performs the setting of route filtering in BGP using a prefix-list by executing the command "neighbor 50.1.1.1 ebgp-multihop 255" on the container 6a.

[0096] Then, the control device 11 sets the BGP keepalive timer and holdtime timer by executing the command "timers bgp 3 9" on the container 6a. Specifically, the keepalive timer is set to 3 seconds and the holdtime timer is set to 9 seconds.

[0097] FIG. 10B is a diagram showing the routing table generated in the upper container 6a. The control device 11 obtains the generated routing table by executing the command "show ip bgp" on the container 6a. An example of the routing table of the execution result is shown below.

[0098] Network Next Hop 10.38.215.0 / 24 192.168.130.42 192.168.120.0 50.1.1.1 192.168.170.0 192.168.110.23 192.168.180.0 192.168.110.23

[0099] FIG. 10C is a diagram showing the static setting and GRE setting for the upper container 6a. The control device 11 executes the command "ip addr add 40.1.1.1 / 32 dev lo" on the container 6a. This is the setting of AnyCastIP for the container 6a. The control device 11 executes the command "ip link add gre type gre local 40.1.1.1 remote 10.38.215.23" for the container 6a. This is the GRE setting for the container 6a. The control device 11 executes the command "ip addr add 60.1.1.1 / 32 dev gre" for the container 6a. This is the GRE setting for the container 6a. The control device 11 executes the command "ip route 50.1.1.1 / 32 gre" for the container 6a. This is a static setting.

[0100] Figure 11A is a diagram showing the BGP setting for the lower container 6b. The control device 11 obtains the current BGP setting by executing the command "router bgp 21" for the upper container 6b. Note that the second argument of the router command is the AS number of the target device.

[0101] The control device 11 advertises the route information within its own AS as a BGP route by executing the command "network 192.168.110.0 / 24" for the container 6b.

[0102] The control device 11 performs the setting of route filtering in BGP using the prefix-list by executing the command "neighbor 192.168.130.42 remote-as 20" for the container 6b.

[0103] The control device 11 advertises the route information within its own AS as a BGP route by executing the command "network 192.168.130.0 / 24" for the container 6b.

[0104] The control device 11 performs the setting of route filtering in BGP using a prefix-list by executing the command "neighbor 192.168.110.23 remote-as 22" on the container 6b.

[0105] The control device 11 advertises the route information within its own AS as a BGP route by executing the command "network 40.1.1.0 / 24" on the container 6b.

[0106] The control device 11 advertises the route information within its own AS as a BGP route by executing the command "network 70.1.1.1 / 32" on the container 6b.

[0107] The control device 11 performs the setting of route filtering in BGP using a prefix-list by executing the command "neighbor 50.1.1.1 remote-as 10" on the container 6b.

[0108] The control device 11 performs the setting of route filtering in BGP using a prefix-list by executing the command "neighbor 50.1.1.1 ebgp-multihop 255" on the container 6b.

[0109] Figure 11B is a diagram showing the routing table generated in the lower container 6b. The control device 11 obtains the generated routing table by executing the command "show ip bgp" on the container 6b. An example of the routing table of the execution result is shown below.

[0110] Network Next Hop 10.38.215.0 / 24 192.168.130.42 192.168.120.0 50.1.1.1 192.168.170.0 192.168.110.23 192.168.180.0 192.168.110.23

[0111] Figure 11C is a diagram showing the static settings and GRE settings for the lower container 6b. The control device 11 executes the command "ip addr add 40.1.1.1 / 32 dev lo" for the container 6b. This is the setting of AnyCastIP for the container 6b. The control device 11 executes the command "ip link add gre type gre local 40.1.1.1 remote 10.38.215.23" for the container 6b. This is the GRE setting for the container 6b. The control device 11 executes the command "ip addr add 70.1.1.1 / 32 dev gre" for the container 6b. This is the GRE setting for the container 6b. The control device 11 executes the command "ip route 50.1.1.1 / 32 gre" for the container 6b. This is the static setting.

[0112] 《Second Embodiment》 Figure 12 is a configuration diagram of the packet relay system 1 according to the second embodiment. The packet relay system 1 includes a plurality of containers 6a, 6b, and protocol units 43, 44 arranged before and after them. The packet relay system 1 connects routers 33, 34 arranged before and after the protocol units 43, 44, and further connects terminals 21, 22 arranged before and after the routers 33, 34. The container 6a is connected to the router 33 via a tunnel 81. The containers 6a, 6b constitute an autonomous system 73.

[0113] The protocol units 43, 44 correspond to the relay devices 41, 42 in FIG. 2. The routers 33, 34 correspond to the devices 31, 32 in FIG. 2.

[0114] Each part of the packet relay system 1 is assigned the same IP address as each part of the packet relay system 1 shown in FIG. 3. And the network interface on the left side of the protocol unit 43 is a NAT segment of 20.1.1.2 / 24. The network interface on the right side of the protocol unit 43 is a NAT segment of 20.1.1.3 / 24.

[0115] FIG. 13 is a flowchart showing the initial setting process. First, the control device 11 sets NAT (Network Address Translation) in the router 33 (step S30). Then, after the control device 11 enables BFD of each node to start monitoring the survival status of the path (step S31), the process of FIG. 13 ends. Here, the nodes where BFD is enabled are the router 33, the protocol unit 43, and the containers 6a and 6b.

[0116] FIG. 14A is a diagram showing the BGP setting for the left protocol unit 43. The control device 11 executes the "router bgp 20" command for the protocol unit 43 to obtain the current BGP setting. Note that the second argument of the router command is the AS number of the target device.

[0117] The control device 11 executes the "network 10.38.215.0 / 24" command for the protocol unit 43 to advertise the route information within its own AS as a BGP route.

[0118] The control device 11 executes the "neighbor 192.168.130.11 remote-as 21" command for the protocol unit 43 to perform the setting of route filtering in BGP using the prefix-list.

[0119] The control device 11 executes the "neighbor 192.168.130.12 remote-as 21" command on the protocol unit 43 to perform the setting of route filtering in BGP using the prefix-list.

[0120] The control device 11 executes the "network 192.168.130.0 / 24" command on the protocol unit 43 to advertise the route information within its own AS as a BGP route.

[0121] The control device 11 executes the "neighbor 10.38.215.23 remote-as 10" command on the protocol unit 43 to perform the setting of route filtering in BGP using the prefix-list.

[0122] The control device 11 executes the "network 20.1.1.0 / 24" command on the protocol unit 43 to advertise the route information within its own AS as a BGP route.

[0123] Figure 14B is a diagram showing the routing table generated by the protocol unit 43 on the left side. The control device 11 executes the "show ip bgp" command on the protocol unit 43 to obtain the generated routing table. An example of the routing table of the execution result is shown below.

[0124] Network Next Hop 192.168.110.0 192.168.130.11 192.168.110.0 192.168.130.12 192.168.120.0 10.38.215.23 192.168.170.0 192.168.130.11 192.168.170.0 192.168.130.12 192.168.180.0 192.168.130.11 192.168.180.0 192.168.130.12

[0125] Figure 14C is a diagram showing the NAT settings for the protocol section 43 on the left side. The control device 11 executes the command "iptables -t nat -A PREROUTING -d 20.1.1.3 -i eth2 -j DNAT --to-destination 192.168.180.25" for the protocol section 43. This is the DNAT setting for the router 33. The control device 11 executes the command "iptables -t nat -A POSTROUTING -s 192.168.180.25 -o eth2 -j SNAT --to-source 20.1.1.3" for the protocol section 43. This is the SNAT setting for the router 33.

[0126] The control device 11 executes the command "iptables -t nat -A PREROUTING -d 20.1.1.2 -i ens10 -j DNAT --to-destination 192.168.120.78" for the protocol section 43. This is the DNAT setting for the router 33. The control device 11 executes the command "iptables -t nat -A POSTROUTING -s 192.168.120.78 -o ens10 -j SNAT --to-source 20.1.1.2" for the protocol section 43. This is the SNAT setting for the router 33.

[0127] Figure 15A is a diagram showing the BGP settings for the router 33 on the left side. The control device 11 obtains the current BGP settings by executing the command "router bgp 10" for the router 33.

[0128] The control device 11 advertises the route information within its own AS as a BGP route by executing the "network 192.168.120.0 / 24" command on the router 33.

[0129] The control device 11 configures route filtering in BGP using a prefix-list by executing the "neighbor 10.38.215.32 remote-as 20" command on the router 33.

[0130] Figure 15B is a diagram showing the routing table generated by the router 33 on the left side. The control device 11 obtains the generated routing table by executing the "show ip bgp" command on the router 33. An example of the routing table of the execution result is shown below.

[0131] Network Next Hop 192.168.110.0 10.38.215.32 192.168.130.0 10.38.215.32 192.168.170.0 10.38.215.32 192.168.180.0 10.38.215.32 20.1.1.0 / 24 10.38.215.32

[0132] Figure 16A is a diagram showing the BGP settings for the protocol unit 44 on the right side. The control device 11 obtains the current BGP settings by executing the "router bgp 22" command on the protocol unit 44. Note that the second argument of the router command is the AS number of the target device.

[0133] The control device 11 advertises the route information within its own AS as a BGP route by executing the "network 192.168.170.0 / 24" command on the protocol unit 44.

[0134] The control device 11 executes the command "neighbor 192.168.110.11 remote-as 21" on the protocol unit 44 to perform the setting of route filtering in BGP using the prefix-list.

[0135] The control device 11 executes the command "neighbor 192.168.110.12 remote-as 21" on the protocol unit 44 to perform the setting of route filtering in BGP using the prefix-list.

[0136] The control device 11 executes the command "network 192.168.110.0 / 24" on the protocol unit 44 to advertise the route information within its own AS as a BGP route.

[0137] The control device 11 executes the command "neighbor 192.168.170.24 remote-as 30" on the protocol unit 44 to perform the setting of route filtering in BGP using the prefix-list.

[0138] Figure 16B is a diagram showing the routing table generated by the protocol unit 44 on the right side. The control device 11 executes the command "show ip bgp" on the protocol unit 44 to obtain the generated routing table. An example of the routing table of the execution result is shown below.

[0139] Network Next Hop 192.168.130.0 192.168.110.11 192.168.130.0 192.168.110.12 192.168.180.0 192.168.170.24 192.168.120.0 192.168.110.11 192.168.120.0 192.168.110.12 10.38.215.0 / 24 192.168.110.11 10.38.215.0 / 24 192.168.110.12 20.1.1.0 / 24 192.168.110.11 20.1.1.0 / 24 192.168.110.12

[0140] Figure 17A is a diagram showing the BGP settings for the router 34 on the right side. The control device 11 obtains the current BGP settings by executing the "router bgp 30" command on the router 34. Note that the second argument of the router command is the AS number of the target device.

[0141] The control device 11 advertises the route information within its own AS as a BGP route by executing the "network 192.168.180.0 / 24" command on the router 34.

[0142] The control device 11 performs the setting of route filtering in BGP using the prefix-list by executing the "neighbor 192.168.170.23 remote-as 22" command on the router 34.

[0143] Figure 17B is a diagram showing the routing table generated by the router 34 on the right side. The control device 11 obtains the generated routing table by executing the "show ip bgp" command on the router 34. An example of the routing table of the execution result is shown below.

[0144] Network Next Hop 192.168.110.0 192.168.170.23 192.168.130.0 192.168.170.23 10.38.215.0 / 24 192.168.170.23 192.168.120.0 192.168.170.23 20.1.1.0 / 24 192.168.170.23

[0145] Figure 18A is a diagram showing the BGP settings for the upper container 6a. The control device 11 executes the "router bgp 21" command for the upper container 6a to obtain the current BGP settings. Note that the second argument of the router command is the AS number of the target device.

[0146] The control device 11 executes the "network 192.168.110.0 / 24" command for the container 6a to advertise the route information within its own AS as a BGP route.

[0147] The control device 11 executes the "neighbor 192.168.130.42 remote-as 20" command for the container 6a to perform route filtering settings in BGP using the prefix-list.

[0148] The control device 11 executes the "network 192.168.130.0 / 24" command for the container 6a to advertise the route information within its own AS as a BGP route.

[0149] The control device 11 executes the "neighbor 192.168.110.23 remote-as 22" command for the container 6a to perform route filtering settings in BGP using the prefix-list.

[0150] Figure 18B is a diagram showing the routing table generated in the upper container 6a. The control device 11 executes the "show ip bgp" command for the container 6a to obtain the generated routing table. An example of the routing table of the execution result is shown below.

[0151] Network Next Hop 10.38.215.0 / 24 192.168.130.42 192.168.120.0 192.168.130.42 192.168.170.0 192.168.110.23 192.168.180.0 192.168.110.23 20.1.1.0 / 24 192.168.130.42

[0152] Embodiment 3 The packet relay system is one in which a plurality of tenants are connected in parallel. As a result, services can be provided to a certain pair of terminals and another pair of terminals respectively.

[0153] FIG. 19 is a configuration diagram of a packet relay system 1B according to the third embodiment. The packet relay system 1B includes a tenant 5a including a plurality of containers 6a and 6b, a tenant 5b including a plurality of containers 6c and 6d, and protocol units 45 and 46 and a control device 11 arranged before and after them. The packet relay system 1B connects routers 35 and 36 arranged before and after the protocol units 45 and 46, and further connects a pair of terminals 23 and 24 and a pair of terminals 25 and 26 arranged before and after the routers 35 and 36.

[0154] The protocol units 45 and 46 correspond to the protocol units 43 and 44 of the first embodiment. The routers 35 and 36 correspond to the routers 33 and 34 of the first embodiment.

[0155] FIG. 20A is a diagram showing the BGP settings for the containers 6a and 6b belonging to the upper tenant 5a. The control device 11 performs BGP configuration by executing the "neighbor {Protocol Unit A IP} distribute-list 100 in" command on the containers 6a and 6b. Here, {Protocol Unit A IP} refers to the IP address of the protocol unit 45.

[0156] The control device 11 performs BGP configuration by executing the "neighbor {Protocol Unit B IP} distribute-list 100 in" command on the containers 6a and 6b. Here, {Protocol Unit B IP} refers to the IP address of the protocol unit 46.

[0157] The control device 11 executes the "access-list 100 permit ip {Terminal A Network IP} {Wildcard Mask} any" command on the containers 6a and 6b. Here, {Terminal A Network IP} refers to the IP address of the terminal 23. {Wildcard Mask} refers to information on which part of the IP addresses of the terminals targeted by the containers 6a and 6b is to be read.

[0158] The control device 11 executes the "access-list 101 permit ip {Terminal B Network IP} {Wildcard Mask} any" command on the containers 6a and 6b. Here, {Terminal B Network IP} refers to the IP address of the terminal 23.

[0159] Figure 20B is a diagram showing the routing table generated in the containers 6a and 6b belonging to the upper tenant 5a. The control device 11 obtains the generated routing table by executing the "show ip bgp" command on the containers 6a and 6b. An example of the routing table of the execution result is shown below.

[0160] Network Next Hop {Terminal A Network IP} {Protocol Unit A IP} {Terminal B Network IP} {Protocol Section B IP}

[0161] It should be noted that the network related to terminals 25 and 26 is not set here. Tenant 5a sets only the network related to terminals 23 and 24.

[0162] Figure 21A is a diagram showing the BGP settings for containers 6c and 6d belonging to the lower tenant 5b. The control device 11 performs BGP settings for containers 6c and 6d by executing the command "neighbor {Protocol Section A IP} distribute-list 102 in". Here, {Protocol Section A IP} refers to the IP address of protocol section 45.

[0163] The control device 11 performs BGP settings for containers 6c and 6d by executing the command "neighbor {Protocol Section B IP} distribute-list 103 in". Here, {Protocol Section B IP} refers to the IP address of protocol section 46.

[0164] The control device 11 executes the command "access-list 102 permit ip {Terminal C Network IP} {Wildcard Mask} any" for containers 6c and 6d. Here, {Terminal C Network IP} refers to the IP address of terminal 25. {Wildcard Mask} refers to information on which part of the IP address of the terminal targeted by containers 6c and 6d is to be read.

[0165] The control device 11 executes the command "access-list 103 permit ip {Terminal D Network IP} {Wildcard Mask} any" for containers 6a and 6b. Here, {Terminal D Network IP} refers to the IP address of terminal 26.

[0166] FIG. 21B is a diagram showing a routing table generated in containers 6c and 6d belonging to the lower tenant 5b. The control device 11 obtains the generated routing table by executing the "show ip bgp" command for the containers 6c and 6d. An example of the routing table of the execution result is shown below.

[0167] Network Next Hop {Terminal C network IP} {Protocol section A IP} {Terminal D network IP} {Protocol section B IP}

[0168] It should be noted that the networks related to terminals 23 and 24 are not set here. Tenant 5b sets only the networks related to terminals 25 and 26.

[0169] 《Effect》 Hereinafter, the effects of the tenant redundancy system and the like according to the present invention will be described.

[0170] 《Claim 1》 A relay device that constitutes a first autonomous system and includes a path exchange unit that is connected to an external device and performs path exchange, A plurality of control units that constitute a second autonomous system different from the first autonomous system, are connected to the relay device, exchange path information, and include a self-monitoring unit that monitors an interface. Tenant, A tenant redundancy system characterized by comprising:

[0171] By configuring in this way, along with the rearrangement of the control elements of the tenant, the information of the network interface of the tenant can be quickly propagated to the relay device.

[0172] 《Claim 2》 The control unit of each said tenant is constructed as a container or a virtual machine, The tenant redundancy system according to claim 1, characterized in that:

[0173] As described herein, the tenant control element can be constructed as a container or a virtual machine.

[0174] Claim 3 The self-monitoring unit of each of the control units monitors each interface, and when an abnormality is detected in any interface, terminates the control unit itself abnormally. The tenant redundancy system according to claim 2, characterized in that.

[0175] By configuring in this way, the tenant control element can quickly terminate itself when an abnormality occurs.

[0176] Claim 4 Each of the control units constructs a plurality of virtual network devices that terminate the tunneling protocol. The tenant redundancy system according to claim 1, characterized in that.

[0177] By configuring in this way, the tenant control element can establish a tunneling protocol with an external device.

[0178] Claim 5 The relay device determines, based on the route information exchanged by the route exchange unit, any one of the tenant control units as the routing destination of the packet. The tenant redundancy system according to claim 1, characterized in that.

[0179] By configuring in this way, the tenant control element can easily establish a tunneling protocol with an external device during relocation.

[0180] Claim 6 The external device constitutes a third autonomous system. The first to third autonomous systems each have a different autonomous system number. The tenant redundancy system according to claim 1, characterized in that...

[0181] By configuring in this way, route information can be easily exchanged between autonomous systems.

[0182] 《Claim 7》 The relay device and the control unit of the tenant each activate BFD (Bidirectional Forwarding Detection). The tenant redundancy system according to claim 1, characterized in that...

[0183] By configuring in this way, the relay device and the control unit of the tenant can check the survival status of the transfer path, detect failures at high speed, and notify the routing protocol.

[0184] 《Claim 8》 Steps for the control unit of the active system of the tenant to construct a tunnel through the relay device; Steps for the route exchange part of the relay device and the route exchange part of the control unit of the active system to exchange routes; Steps for the self-monitoring part of the control unit of the standby system of the tenant to monitor the interface; When the self-monitoring part of the control unit of the standby system detects an abnormality in any of the interfaces, steps to abnormally terminate the control unit of the active system; Steps for the control unit of the standby system to reconstruct the tunnel through the relay device; Steps for the control unit of the standby system to transition to the active system; A tenant redundancy method characterized by executing the following.

[0185] By configuring in this way, when the control element of the tenant detects an abnormality in the active system, the standby system can appropriately transition to the active system.

Explanation of Signs

[0186] 1A, 1 Packet Relay System (Tenant Redundancy System) 21 to 26 Terminals 31 Device 311 IP Address 312 BGP Section 32 Device 321 IP Address 322 BGP Section (Route Exchange Section) 33 to 36 Routers 41 Relay Device 411, 412 Network Interfaces 413 FIB 414 BGP Section (Route Exchange Section) 42 Relay Device 421, 422 Network Interfaces 423 FIB 424 BGP Section (Route Exchange Section) 43 to 46 Protocol Sections 5a, 5b Tenants 6, 6a, 6b, 6c, 6d Containers (Control Units) 61, 62 Network Interfaces 63, 64 tap Devices 65 FIB 66 Self - Monitoring Section 67 BGP Section (Route Exchange Section) 68, 69 Logical Addresses 71 to 75 Autonomous Systems 81, 82 Tunnels 9 Server

Claims

1. A relay device that constitutes a first autonomous system and includes a path exchange unit connected to an external device to exchange path information, and A tenant that constitutes a second autonomous system different from the first autonomous system, is connected to the relay device, includes a path exchange unit that exchanges path information, and a plurality of control units that each include a self-monitoring unit that monitors a plurality of interfaces. A tenant redundancy system, characterized by comprising the above.

2. The control unit of each tenant is constructed as a container or a virtual machine. The tenant redundancy system according to claim 1, characterized by the above.

3. The self-monitoring unit of each control unit monitors each interface, and when an abnormality is detected in any interface, terminates the control unit itself abnormally. The tenant redundancy system according to claim 2, characterized by the above.

4. Each control unit constructs a plurality of virtual network devices that terminate a tunneling protocol. The tenant redundancy system according to claim 1, characterized by the above.

5. Based on the path information exchanged by the path exchange unit, the relay device determines any one of the control units of the tenant as the routing destination of the packet. The tenant redundancy system according to claim 1, characterized by the above.

6. The external device constitutes a third autonomous system, and The first to third autonomous systems each have a different autonomous system number. The tenant redundancy system according to claim 1, characterized by the above.

7. The relay device and the control unit of the tenant each have BFD (Bidirectional Forwarding Detection) enabled. The tenant redundancy system according to claim 1, characterized by the above.

8. Steps for the control unit of the active system of the tenant to construct a tunnel via the relay device; Steps for the path exchange unit of the relay device and the path exchange unit of the control unit of the active system to exchange path information; Steps for the self-monitoring unit of the control unit of the active system of the tenant to monitor a plurality of interfaces; Steps for the self-monitoring unit of the control unit of the active system to terminate the control unit of the active system abnormally when an abnormality is detected in any of the plurality of interfaces; Steps for the control unit of the standby system to reconstruct the tunnel via the relay device. The step in which the control unit of the standby system transitions to the active system, and A tenant redundancy method characterized by executing.

Images