JP3794580B2 - DATA RELAY METHOD, DATA RELAY DEVICE, AND DATA RELAY SYSTEM USING THE DEVICE - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
          The present invention relates to a data relay method, a data relay device, and a device for relaying data constructed by a plurality of logical data relay devices that are logically set in a relationship between an active state and a standby state and perform main processing and backup processing. In particular, the present invention relates to a data relay method for relaying data by establishing a session between routers using MPLS (Multi Protocol Label Switching), a data relay device, and a data relay system using the device. It is.
[0002]
[Prior art]
          A conventional data relay method is, for example, an LSP (Label Switched) on an optimum route selected using a routing protocol such as OSPF (Open Shortest Path First) between routers in a single autonomous system (AS). By setting (Path), a session is established between the edge routers on the source user side and the destination user side. Then, by transferring the packet received from the network of the source user to the edge router on the destination user side (hereinafter referred to as “destination side edge router”) via the route from the edge router on the source user side, It was possible to control the packet communication path.
[0003]
          A typical control protocol for setting the LSP is the LDP (Label Distribution Protocol: RFC3036) protocol. In the conventional data relay method, as shown in FIG. 11, a plurality of routers 20 and 30 using the LDP signaling protocol are connected to the network 10 inside the AS, and the LDP processing unit 21a of these routers 20 and 30 is connected. , 22a sets the LSP on the optimum route based on the optimum route information including the next hop stored in the forwarding table (not shown) in the route information tables 21b, 22b, and transmits data between the routers 30, 20. I was relaying.
[0004]
          These routers 20 and 30 may duplicate the LDP protocol as a safety measure when a failure occurs. In this case, as shown in FIG. 13, for example, one router (hereinafter referred to as “physical router”) 20 is physically connected to two routers (hereinafter referred to as “physical routers”) in performing data relay operation. The logical routers 21 and 22 are connected to the other routers 30 and the routing control data (hereinafter referred to as “routes”) by the duplicated LDP protocol (built in the LDP processing units 21a and 22a). Some exchange information).
[0005]
          This logical router is in an active state and operates as a logical router (hereinafter referred to as “active side router”) 21 that performs main processing, and a logical router that operates in a standby state and performs backup processing (hereinafter referred to as “standby side”). 22), each logical router 21 and 22 is connected to another router 30, and the path information received from the other router 30 is immediately copied, and the LDP processing unit 21a of the logical routers 21 and 22 is connected. , 22a. Further, regarding the data generated on the active side router 21 side, when the routers 21 and 22 switch between the active state and the standby state, the standby side router 22 can immediately start the active state operation so that the routers 21 and 22 can start. It is necessary to synchronize sequentially.
[0006]
[Patent Document 1]
              JP 2000-31989 (page 4-6, FIG. 1)
[0007]
[Problems to be solved by the invention]
          However, in the above conventional example, if a failure occurs in the course of exchanging route information with another router and a packet loss occurs, the active router 21 and the standby router 22 cannot be completely synchronized, and at the time of the above state switching It becomes impossible to receive the route information accompanying the topology change. Therefore, in the signaling protocol LDP that determines the path information of MPLS, it is conceivable that the standby side router returns to the initial state by the initial designation and acquires the entire path information. Until completion, there was a problem that data communication via this router was stopped.
[0008]
          In Patent Document 1, an active data relay control unit and a standby data relay control unit are provided. When a failure occurs in the active data relay control unit, standby data relay control is performed using an ATM switch. However, in this case as well, it is conceivable that the entire route information can be obtained in the same manner as described above. There was a problem that data communication via the router would stop.
[0009]
          SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and a data relay method, a data relay device, and a data relay device that can easily duplicate data relay devices without stopping data relay processing as much as possible even when a failure occurs. An object of the present invention is to provide a data relay system using an apparatus.
[0010]
[Means for Solving the Problems]
          In order to achieve the above object, in the data relay method according to claim 1, the data relay device includes at least two first and second logical data relay devices logically set in a relationship between an active state and a standby state. In the data relay method for constructing and exchanging route information and data relay,eachLogical data relay deviceEstablish a session using the label switching method with other data relay devices,Processing steps for independently exchanging the route information and relaying dataA first monitoring step in which the logical data relay devices monitor each other's operation state; and when a failure occurring on the active side is detected from the monitoring result, the second logical data relay device in the standby state , Assigning new label information and transmitting to the other data relay device;A data relay method is provided.
[0011]
          According to the present invention, exchange of route information and data relay between the active side router and the standby side router, which are logical data relay apparatuses logically set in the relationship between the active state and the standby state, are performed regardless of the mutual operation. By performing it independently, data relay processing is performed without stopping as much as possible even when a failure occurs on the active side.At the same time, the active side router and standby side router monitor the operating status of each other, relay data using the label information assigned by the active side router during normal operation, and when the active side detects a failure, the standby side sends new label information. By assigning and transmitting to the partner data relay device, data relay processing is performed without stopping as much as possible even when a failure occurs on the active side.
[0012]
          Claims2Data relay methodIn the above invention,Each of the data relay devices detects a failure occurring in the connection with the active side when a second monitoring step of monitoring a connection state with the other data relay device, and the other data relay device detects the failure And a data relay step of performing data relay by a session established between the second logical data relay devices in the standby state.
[0013]
          According to the present invention, the connection state between the other side MPLS router which is another data relay device and the active side router and the standby side router is monitored, and when a failure occurs in the connection with the active side, the standby side router and the other side MPLS router By relaying data through sessions established between them, data relay processing is performed without stopping as much as possible even when a failure occurs on the active side.
[0014]
          Claims3The data relay apparatus according to the present invention has at least two first and second logical data relay apparatuses that are logically set in a relationship between an active state and a standby state, and the first and second logical data relay apparatuses In the data relay device for exchanging route information and performing data relay, each logical data relay device includes:With other data relay device To establish a session using the label switching method,Processing means for exchanging route information and relaying data independentlyAnd a first monitoring means for monitoring an operating state of the other logical data relay device, and when a failure is detected on the active side from the monitoring result, new label information is allocated, and the other data relay device A transmission means for transmitting toA data relay device is provided.
[0015]
          According to the present invention, the processing means of the active side router and the standby side router are operated independently, so that the down time at the time of switching the relay is shortened, and the data relay process is performed as much as possible even when an active side failure occurs. Do without stoppingAt the same time, the processing means of each router mutually monitor the operating state of the router, and when a failure occurs on the active side, the processing means on the standby side allocates new label information and transmits it to the partner data relay device. Data relay processing without interruption as much as possible even when a failure occurs.
[0016]
          Claims4In such a data relay system, in a data relay system in which a plurality of data relay devices are connected to exchange route information and data relay with other data relay devices,A processing unit that establishes a session by a label switching method with the other data relay device, and performs the exchange of the route information and the data relay independently of each other, and the operation state of the other logical data relay device A first monitoring means for monitoring the data, and a transmission means for allocating new label information and transmitting to the other data relay apparatus when a failure is detected on the active side from the monitoring result, and logically Having at least two first and second logical data relay devices that are set in a relationship between an active state and a standby state.Equipped with data relay deviceOctopusData relay system characterized byWill be provided.
[0017]
          According to the present invention, the exchange of the routing information and the data relay between the active side router and the standby side router are performed independently regardless of the operation of each other, or the operation state is monitored mutually, and when a failure occurs By switching the data relay to the other router, the data relay down time is shortened, and the data relay processing is performed as much as possible even when a failure occurs on the active side.
[0018]
DETAILED DESCRIPTION OF THE INVENTION
          A preferred embodiment of a data relay method, a data relay device, and a data relay system using the device according to the present invention will be described below with reference to the accompanying drawings of FIGS. In the following drawings, the same components as those in FIG. 11 are denoted by the same reference numerals for convenience of explanation.
[0019]
        (Embodiment 1)
          FIG. 1 is a system configuration diagram showing an example of a schematic configuration of a data relay system according to the present invention. In the figure, in the data relay system of this embodiment, as in FIG. 11, a plurality of routers 20 and 30 inside the AS are provided and connected on the network 10, and these routers 20 and 30 are faulty. As a safety measure at the time, the LDP protocol is duplicated, and each is set to operate independently.
[0020]
          The router 20 is a data relay device according to an embodiment of the present invention. The router 20 is an active side router 21 that is an active logical router equipped with the duplicated LDP protocol, and a standby side that is a standby state logical router. And a router 22.
[0021]
          In the MPLS router 20 according to this embodiment, as shown in FIG. 1, route information is acquired by exchanging route information separately between the active router 21 and the standby router 22 and the other routers 30. ing. Note that the MPLS router 22 can set, for example, the active router 21 and the standby as the next hop of the optimum route when the adjacent partner MPLS router 30 can set a plurality of next hops for the data reachable network. Each side router 22 is designated and stored in the path information table of the counterpart router 30. Whether this partner MPLS router 30 can set a plurality of next hops or one next hop for the destination network 40 is artificially set when the router 20 is connected to the network 10, for example. .
[0022]
          The active side router 21 and the standby side router 22 acquire route information on the network 10 side in the AS, and exchange route information with a router (not shown) in the network 40, so that other than AS Route information on the network 40 side is also acquired.
[0023]
          FIG. 2 is a configuration diagram showing a configuration related to the active side router 21 and the standby side router 22 in the router 20 shown in FIG. In the figure, the active router 21 performs data processing according to the LDP protocol.As a processing meansThe LDP processing unit 21a includes a route information table 21b that stores received route information. The standby side router 22 performs data processing according to the LDP protocol.As a processing meansThe LDP processing unit 22a includes a path information table 22b that stores received path information. The LDP processing units 21 a and 22 a assign label information to the partner MPLS router 30. In addition, LDP processing section21a, 22aPerforms a negotiation process with an adjacent router (a partner MPLS router or the like) to set, for example, a data communication speed.
[0024]
          Furthermore, the router 20 includes a relay processing unit 23 that relays data based on the label information notified from the LDP processing units 21a and 22a. The relay processing unit 23 stores the label information notified from the LDP processing units 21 a and 22 a of the active side router 21 and the standby side router 22 in association with the destination address, and stores the label information in the partner MPLS router 30. Each is notified. When the relay processing unit 23 receives the packet with the MPLS Shim header having the format shown in FIG. 3 from the partner MPLS router 30, the relay processing unit 23 extracts the label information in the header, and the extracted label information is used as the LDP process. If it matches the label information notified from the units 21a and 22a, it is transferred to the destination network 40.
[0025]
          As shown in FIG. 3, the MPLS Shim header includes a label area for storing label information, an Exp (Experiment) area allocated for various experiments, an S area for bottom-of-stack display, It is composed of a TTL area for storing the Time To Live value of this header. The MPLS Shim header is added and transmitted between the header of the layer 2 and the data area, for example, by a router in the AS.
[0026]
          Further, the LDP processing units 21a and 22a periodically transmit keep-alive signals to the partner MPLS router 30 to notify the router 30 that they are maintaining a normal operating state.
[0027]
          The partner MPLS router 30 is set so that a plurality of next hops can be registered with respect to a destination address in the network, for example, with respect to the network where the data can be reached. The router 30 transmits relay data to one or both of the active side router 21 and the standby side router 22 (including load distribution) based on the label information separately assigned from the MPLS router 20. Can do.
[0028]
          Next, the operation of the data relay system when a failure occurs on the active side will be described based on the schematic configuration diagram of the data relay system in FIG. 4 and the time charts of the data relay system in FIGS. 5 and 6. Here, the case without load balancing will be described with reference to FIG. 5, and the case with load balancing will be described with reference to FIG.
[0029]
          First, during normal operation without load balancing, as shown in FIG. 5, a keep-alive signal arrives from the active router 21 to the partner MPLS router 30, so that the router 30 is connected to the active router 21 in advance. Data relay is performed using an established session.
[0030]
          Next, when a failure occurs on the active side such as a failure of the active router 21 (see FIG. 4), the keep alive signal transmitted from the active router 21 does not reach the partner MPLS router 30 due to packet loss. If the router 30 cannot receive a keep-alive signal from the active router 21 even after the preset maximum packet loss time has elapsed, the router 30 determines that a failure has occurred on the active side, and The LDP session established with the router 21 is disconnected. Next, since the keep-alive signal arrives from the standby side router 22, the router 30 determines that the standby side router 22 is in a normal state and receives relay data to the network 40. Switching to the relay, this data is transferred using a session established with the standby side router 22 in advance.
[0031]
          When receiving this data, the standby side router 22 deletes the MPLS Shim header in the packet, and transfers this data to the terminal device of the corresponding destination address via the network 40.
[0032]
          Next, a case with load distribution will be described. When the load distribution is normal, as shown in FIG. 6, since the keep alive signal arrives at the partner MPLS router 30 from the active side router 21 and the standby side router 22, the router 30 is previously connected to the active side router 21. Data relaying is performed using a session established with the standby router 22.
[0033]
          Next, when a failure occurs on the active side such as a failure of the active router 21 (see FIG. 4), the keep alive signal transmitted from the active router 21 does not reach the partner MPLS router 30 due to packet loss. In this case as well, in the same way as described above, even if the preset maximum packet loss time has elapsed, if the keep-alive signal does not reach from the active router 21, a failure has occurred on the active side. Therefore, the LDP session established with the active router 21 is disconnected. Next, since the keep-alive signal arrives from the standby side router 22, the router 30 determines that the standby side router 22 is in a normal state and receives relay data to the network 40. Switching to the relay, the data is transferred using only the session established with the standby side router 22.
[0034]
          As described above, in this embodiment, the active side router and the standby side router are operated as different independent MPLS logical routers, and label information is separately assigned to networks where data can be reached from these routers. Since these label information is transmitted to the partner MPLS router for setting registration, relay data from the partner MPLS router can be transmitted to one or both of the active side and the standby side, and there is a failure on the active side. Even if it occurs, the data relay device can be easily duplicated without stopping the data relay processing.
[0035]
        (Embodiment 2)
          FIG. 7 is a system configuration diagram showing an example of a schematic configuration of the data relay system according to the present invention. 7 differs from the system of FIG. 1 in that the active router 21 and the standby router 22 periodically transmit, for example, keep-alive signals to monitor each other's operating state, and the partner MPLS router. 30 is that only one next hop can be set for the destination network 40. In the normal state where both the active side router 21 and the standby side router 22 are operating normally, for example, the active side router 21 is set as the next hop of the optimum route, and the route of the partner MPLS router 30 is set. When the active router 21 is in a failure state, the label information assigned by the standby router 22 is stored in the path information table of the partner MPLS router 30.
[0036]
          Therefore, as shown in the configuration diagram relating to the active side router 21 and the standby side router 22 in FIG. 8, the active side router 21 receives the keep alive signal output from the LDP processing unit 22 a of the standby side router 22. Monitoring the standby processing of the LDP processing unit 22aAs the first monitoring meansA standby process monitoring unit 21c is further provided. The standby side router 22 receives the keep alive signal output from the LDP processing unit 21a of the active side router 21, and monitors the active processing of the LDP processing unit 21a.As the first monitoring meansAn active process monitoring unit 22c is further provided.
[0037]
          The standby process monitoring unit 21c and the active process monitoring unit 22cAs a means of transmissionIf any one of the keep-alive signals from the LDP processing units 21a and 22a cannot be received even after the preset maximum packet loss time has elapsed, it is determined that a failure has occurred on the standby side or the active side. If the active process monitoring unit 22c determines that a failure has occurred on the active side, the LDP processing unit 22a transmits new label information assigned to the destination network 40 to the partner MPLS router 30, The label information corresponding to the destination network registered in the path information table of the partner MPLS router 30 is changed.
[0038]
          The partner MPLS router 30 is set so that only the next hop of the optimum route can be registered with respect to the network where data can be reached. The router 30 can transmit relay data to either the active side router 21 or the standby side router 22 based on the label information assigned from the MPLS router 20.
[0039]
          Next, the operation of the data relay system when a failure occurs on the active side will be described based on the time chart of the data relay system in FIG. 9 and the flowchart for explaining the processing operation of the standby side router 22 in FIG.
[0040]
          First, when the active router 21 is normal, as shown in FIG. 9, the keepalive signal arrives from the active router 21 to the partner MPLS router 30, so this router 30 is established with the active router 21 in advance. Data is relayed using the session.
[0041]
          Next, when a failure occurs on the active side such as a failure of the active router 21 (see FIG. 4), a keep-alive signal transmitted from the LDP processing unit 21a of the active router 21 is sent to the standby router 22 and the partner MPLS router 30. Will not reach. If the standby router 22 cannot receive the keep-alive signal from the active router 21 even after the preset maximum packet loss time has elapsed, the standby router 22 determines that a failure has occurred on the active router, The label information assigned to the network 40 is transmitted to the partner MPLS router 30.
[0042]
          YouNaThat is, in the standby side router 22, as shown in the flowchart of FIG. 10, the LDP processing unit 22a exchanges route information (step 101), calculates the optimum route, and creates the route information table 22b. (Step 102). Next, the active process monitoring unit 22c monitors the keep alive signal transmitted from the LDP processing unit 21a of the active side router 21 (step 103), and this keep alive signal is set to a preset maximum packet loss time. If it is not possible to receive even after elapse of time, it is determined that a failure has occurred on the active side, and the LDP processing unit 22a is notified of the failure. Upon receiving this notification, the LDP processing unit 22 a assigns label information to the destination network 40 and notifies the relay processing unit 23 of this label information.
[0043]
          When receiving the label information notification from the LDP processing unit 22a, the relay processing unit 23 stops data relay from the active side (step 104), and transmits the label information notified from the standby side router 22 to the partner MPLS router 30. (Step 105).
[0044]
          As shown in FIG. 9, when new label information is received from the standby side router 22, the counterpart MPLS router 30 changes the registered label information, and then keep-alive signal from the active side router 21. Is not received even after the preset maximum packet loss time has elapsed, the LDP session established with the active router 21 is disconnected. Then, the router 30 switches to data relay via the standby router 22 and performs data transfer using the relay data established in advance with the standby router 22.
[0045]
          As in the first embodiment, when receiving this data, the standby side router 22 deletes the MPLS Shim header in the packet and transfers this data to the terminal device of the corresponding destination address via the network 40. .
[0046]
          If the active side router 21 recovers when the active side router 21 detects the occurrence of a standby side failure, the active side router 21 assigns label information to the destination network 40 and relays this label information. The data is transmitted to the partner MPLS router 30 via the processing unit 23.
[0047]
          As described above, in this embodiment, the active router and the standby router mutually monitor the operation state of each other, and when a failure occurs on one logical router side, the other logical router is connected to the destination network. Since label information is assigned to these, and the label information is transmitted to the partner MPLS router for setting registration, relay data from the partner MPLS router can be transmitted to either the active side or the standby side, Even if a failure occurs on the active side, it is only necessary to stop the data relay processing for a short time when the newly assigned label information is registered in the partner MPLS router. Therefore, in this embodiment, it is possible to easily duplicate data relay apparatuses without stopping data relay processing as much as possible even when a failure occurs.
[0048]
          The present invention is not limited to these embodiments, and various modifications can be made without departing from the spirit of the present invention.
[0049]
【The invention's effect】
          As described above, in the present invention, path information exchange and data relay processing operations are performed independently by the logical data relay device logically set in the relationship between the active state and the standby state. However, it is possible to easily duplicate the data relay device without stopping the data relay processing as much as possible.
[0050]
          In the present invention, the logical data relay devices in the active state and the standby state monitor each other's operation state, and when a failure occurs on the active side, the standby side assigns new label information and transmits it to the partner data relay device. Therefore, it is possible to easily duplicate the data relay apparatus without stopping the data relay process as much as possible even when a failure occurs.
[Brief description of the drawings]
FIG. 1 is a system configuration diagram showing an example of a schematic configuration of a data relay system according to the present invention.
FIG. 2 is a configuration diagram illustrating an example of a configuration related to an active router and a standby router among the routers illustrated in FIG. 1;
FIG. 3 is a diagram illustrating a configuration format of an MPLS Shim header.
FIG. 4 is a schematic configuration diagram of a data relay system when an active failure occurs.
FIG. 5 is a diagram showing a time chart of the data relay system shown in FIG. 1 when there is no load distribution;
6 is a diagram showing a time chart of the data relay system shown in FIG. 1 when there is load distribution. FIG.
FIG. 7 is a system configuration diagram showing another example of the schematic configuration of the data relay system according to the present invention.
8 is a configuration diagram showing an example of a configuration related to an active side router and a standby side router among the routers shown in FIG. 7;
9 is a time chart of the data relay system shown in FIG.
10 is a flowchart for explaining the processing operation of the standby side router shown in FIG. 8;
FIG. 11 is a system configuration diagram showing a schematic configuration of a conventional example of a data relay system.
[Explanation of symbols]
          10,40 network
          20, 30 MPLS router
          21 Active side router (logical router)
          21c Standby processing monitoring unit
          21a, 22a LDP processing part
          21b, 22b route information table
          22 Standby router (logical router)
          22c Active process monitoring unit
          23 Relay processing section

Claims (4)

In a data relay method in which a data relay device constructs at least two first and second logical data relay devices that are logically set in a relationship between an active state and a standby state, and exchanges route information and relays data ,
A processing step in which each logical data relay device establishes a session by a label switching method with another data relay device, and exchanges the route information and relays data independently ,
A first monitoring step in which each logical data relay device monitors each other's operating state;
When detecting a failure occurring on the active side from the monitoring result, the second logical data relay device in the standby state allocates new label information and transmits to the other data relay device;
A data relay method comprising: In the data relay method, each data relay device monitors a connection state with the other data relay device;
When the other data relay device detects a failure occurring in the connection with the active side, a data relay step of performing data relay by a session established between the second logical data relay devices in the standby state; ,
The data relay method according to claim 1, further comprising: It has at least two first and second logical data relay devices that are logically set in a relationship between an active state and a standby state, and exchanges route data and data by the first and second logical data relay devices. In the data relay device that performs the relay,
Each logical data relay device establishes a session by a label switching method with another data relay device, and processing means for independently exchanging the route information and data relay ,
First monitoring means for monitoring the operating state of the other logical data relay device;
When a failure is detected on the active side from the monitoring result, new label information is allocated and transmitted to the other data relay device;
A data relay device comprising: In a data relay system in which a plurality of data relay devices are connected and exchange route information and data relay with other data relay devices.
A processing unit that establishes a session by a label switching method with the other data relay device, and performs the exchange of the route information and the data relay independently of each other, and the operation state of the other logical data relay device A first monitoring means for monitoring the data, and a transmission means for allocating new label information and transmitting to the other data relay apparatus when a failure is detected on the active side from the monitoring result, and logically data relay system comprising a kite comprising a data relay device comprising at least two first and second logic data relay apparatus which is set in relation active and standby state manner.

Images