JP3484351B2 - Routing table changing method and device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention provides an alternative method by detecting a failure at the destination of a call setting request message in a communication system in which a communication is established by establishing a connection to a destination after a communication request is issued. The present invention relates to a routing table changing method and apparatus for selecting a route to establish a communication path, and an An of a server used in a LAN emulation (LANE) service of an ATM network (ATM-NW).
The present invention relates to a routing table changing method and apparatus that can be used even when connecting to a server when there are a plurality of ycast connection addresses in the network.

[0002]

2. Description of the Related Art Conventionally, in the ATM-NW routing prior to PNNI1.0, a method of manually setting hop-by-hop routing using IISP has been the mainstream, and this method can detect only an abnormality of an adjacent device. I couldn't.

[0003]

Therefore, the destination is 1
When a switch that has only a route detects a failure, the connection cannot be actually set even if there is another route to the destination.

It is an object of the present invention to provide a routing table changing method and apparatus which can establish a connection using an SVC in an alternative route by solving the above-mentioned problems of the prior art.

[0005]

In order to achieve the above object, the present invention provides a routing in an exchange used in a communication system which establishes a connection to a destination after an information transmission request is made and performs communication. In the table changing method, as shown in FIG. 1, a message processing unit (101) that inserts information into a received message, and stores a call number and a called address when a call setting request message is received. A call number / destination address storage unit (102), a table control unit (103) that controls the routing table, a resource management unit (104) that manages the resources of the exchange, and a routing table that refers to a route for transmitting a message. (105), a timer (106) for measuring the holding time when the routing table is changed, and the resource management unit (10).
When 4) detects a network failure or port abnormality of the exchange, it sends a port abnormality notification including the port number to the table control unit (103), and the table control unit (103) receives it from the resource management unit (104). The routing table (1
When all the entries of (05) are not selectable and the exchange receives the call setup request message, it sends it to the message processing unit (101), and the message processing unit (101) calls the call number and the called address in the message. Number / destination address storage
The call number / destination address storage unit (102) stores the pair of the call number and the destination address sent from the message processing unit (101) as information, and the message processing unit (101) executes the routing. Search the route for sending the received call setup request message by referring to the table (105), and if the entry of the route to send the message is not selectable, generate the call setup reject message. The destination terminal is unreachable, the information is erased from the call number / destination address storage unit (102) for the call number of the message, and the exchange is assigned the information "unreachable terminal is unreachable". When the received call setup rejection message is received, the message is sent to the message processing unit (101), and the message processing unit (101) refers to the call number / destination address storage unit (102) to reject the received call setup rejection message. Call to message When the destination address of the fixed request message is calculated and "recipient terminal unreachable" information is detected from the received call setup rejection message, the information in the call number / destination address storage unit (102) for the call number of the message is deleted. The port number that received the call setup rejection message and the destination address that was calculated earlier are sent to the table control unit (103), and the table control unit (103) outputs the port number received from the message processing unit (101) and the calculated destination address. Based on the routing table (10
Refer to (5) to make the entry of the target port unselectable, start its own timer (106), and after the timer (106) measures the elapse of a certain time, the table control unit (103)
To the table control unit (103), the table control unit (103) rewrites the routing table (105) again after the timer (106) ends, and the table setting unit (103) selects it by receiving the call setup rejection message. The disabled entry is returned to the selectable state again.

According to the present invention, in an ATM-NW capable of coping with only a local failure due to hop-by-hop routing, only by extracting information of a reason display portion of a message based on a conventional specification called a call setup rejection message. , By detecting a remote failure and selecting an alternative route, the failure can be avoided and the connection by SVC can be established in a short time.

[0007]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the present invention will be described below with reference to the drawings.

FIG. 2 shows the network configuration used in the present invention, and FIG. 3 shows the state of the routing table. In FIG. 2, the terminal A is connected to the exchange SW1, and in the normal route, communication is performed from the exchange SW2 through the exchange SW4 to the terminal B connected to the exchange SW4. Further, the exchange SW1 has an exchange SW3 as another route, and the exchange SW3 is connected to the exchange SW4. In this figure, it is assumed that a failure has occurred between the exchange SW2 and the exchange SW4.
Further, the routing table in FIG. 3 considers that there are a plurality of output ports for one destination, and indicates whether or not these ports can be selected when transmitting information. If you cannot select, you can check the reason. Here, as the reasons why the selection is not possible, a physical abnormality at the output port, a call setup rejection message including the destination unreachability information are received, and an instruction from the NMS is given. If any of these reasons are checked, no message will be sent to that port.

An example of using the present invention in the network configuration of FIG. 2 is shown. When the terminal A sends a call setting request message, the message processing unit (301) of the exchange SW1 receives it,
Assigned call number (CR # 1) and destination address (470000000000000
00000000000aaaaaaaaaaaa00) to the call number / destination address storage unit (302). The message processing unit (301) refers to the routing table (305), selects the output port # 1 (see the table in FIG. 3), and sends the message to the exchange SW2. At this time, in the exchange SW2, the resort management unit (314) has detected the occurrence of a failure, and the table control unit (313) is notified of the failure, and the table control unit (313) displays the routing table (315) in FIG. 2) -Instruct to rewrite like (a) (This table shows (2)-(b) in Fig. 3 after the failure is recovered.
It is returned to the normal state like). Upon receiving the call setup request message from the exchange SW1, the message processing unit (311) of the exchange SW2 firstly includes the call number (CR # 1) and the destination address (47000000000000000000000000aaa) included in the received message.
aaaaaaaaa00) to the call number / destination address storage unit (312). Next, the rewritten routing table (315)
When it is detected that the terminal B cannot be reached due to a physical failure, the call number and the callee address entered previously are deleted from the call number / callee address storage unit (312), and the generated call setup rejection message Put the "unable to reach the other party" information in it and return it to the exchange SW1. In the exchange SW1, the message processing unit (301) that receives the call setup rejection message in which the "remote terminal unreachable" information is inserted refers to the call number / destination address storage unit (302) and The called address for the call number included in is read. Then, this destination address is notified to the table control unit (305), and at the same time, the corresponding entry (call number = CR # 1, destination address = 470000000000) in the call number / destination address storage unit (302).
(00000000000000aaaaaaaaaaaa00) is deleted. Next, the table control unit (305) receives the destination address (47000000000000000000000000aaaaaaa) received from the message processing unit (301).
The entry for aaaaa00) is rewritten as shown in (1)-(a) of FIG. 3, the call setup rejection message is returned to the terminal A, and at the same time, the timer (306) is started. Then terminal A again
When a call setup request message is sent to the, the next entry (output port = 2) is selected when the message is sent according to this routing table. After a certain period of time has passed, the timer (306) notifies the table control unit (303) that the timer (306) has been stopped, and the table control unit (303) sets the routing table (305) to the state before the failure. Return to the table ((1)-(b) in Figure 3).

FIG. 4 shows a basic sequence of signaling exchanged in the case of FIG. Terminal A
When setting a communication path from the terminal A to the terminal B, the priority route for the fixed route selection is the route 1 (via the exchange SW2) in the exchange SW1 that received the call setup request message from the terminal A.
If so, the call setup request message is transmitted to the exchange SW2. The exchange SW2, which has received this, tries to send a call setup request message to the exchange SW4, but cannot send it because a failure has occurred in the front. Therefore, a call setup rejection message is returned to the exchange SW1. In this call setup refusal message, "unreachable terminal unreachable" information indicating that the partner terminal cannot be reached is added.
Upon receiving this call setup rejection message, the exchange SW1 does not select the route 1 as the route to the terminal B. Next, when the terminal A sends a call setting request message by retry, the exchange SW1 selects route 2 (via the exchange SW3) as the transmission destination. As a result, the call setup request message is transmitted to the terminal B via the exchange SW3, and as a result, the connection is established through this route.

FIG. 5 shows a state in which the servers are housed in different exchanges SW in a network having a LANE server duplex configuration, and FIG. 6 shows the state of the routing table in the exchange SW2 at that time. It is a thing. In FIG. 5, there are two exchanges SW, an exchange SW1 and an exchange SW2, to which LANE servers A and B are connected. Each of the exchanges SW accommodates two terminals 1, 2, 3, and 4, respectively.

A terminal that receives the LANE service is always connected to the server. Therefore, the terminal must first establish a connection with the server. The server uses a special address called a well known address, which is an Anycast address. Therefore, the servers in FIG. 5 have the same address, and there are two destinations having the same address on the network.

In the dual server configuration of FIG. 5, it is assumed that one of the two servers A and B operates as an ACT system (main) and the other as an SBY system (standby). At this time, the server well
The output port of the switch SW2 for the known address is the first
The priority is port 1 (the port connected to the exchange SW1), and the second priority is the port 2 (the port connected to the spare server).
Here, description will be made focusing on the terminal 3 of the exchange SW2. Further, it is assumed here that the main server A has a failure.

From FIG. 5A, first, the terminal 3 connects to the server by a well known address (470079000000000000000000).
0000a03e00000100) as the destination address. The message processing unit (511) of the exchange SW2 that received the message indicates the call code included in the received message.
(CR # 1) and destination address (470079000000000000000000000)
0a03e00000100) is sent to the call number / destination address storage unit (512), stored therein, and the call setting request message is relayed to the exchange SW1 by the IISP. At the exchange SW1, the call number (CR # 1) and destination address included in the received message
(4700790000000000000000000000a03e00000100) is sent to the call number / destination address storage unit (512) and stored therein. Next, the resource management unit (504) notifies the table control unit (503) that a failure has occurred in the main server port, and the table control unit (503) notifies port 1 to the routing table (505). Included entries cannot be selected. At the same time, the previously stored entry of the call number / destination address storage unit (502) is erased, and a call setup refusal message including "unreachable terminal unreachable" information is returned to the exchange SW2. Switch SW2
The message processing unit (511) of the call number / destination address storage unit (5) based on the call number of the received call setup rejection message.
The destination address corresponding to the message is found by referring to (12), and at the same time the table control unit (513) is notified, the corresponding entry in the call number / destination address storage unit (512) is deleted. The table control unit (513) makes it impossible to select the port 1 corresponding to the received destination address due to the reason of "reception of call setup rejection message including reception of" remote terminal unreachable "information", and routing table (515). (See (a) in FIG. 6) and at the same time activates the timer (516), when the terminal 3 sends a call setup request message,
Based on this routing table (515), the message is forwarded to the spare server on port 2. Then, after a certain time has passed ((b) of FIG. 5), the timer (516) is the table control
(513) is notified that it will be stopped, so that the table control unit (513) cannot select it in the routing table (515) for the reason of "reception of call setup rejection message including reception of" unreachable terminal unreachable "information". The entry that is changed to the selectable state (see (b) in FIG. 6).

In this method, many terminals are designed to retry when connection to the server fails.

FIG. 7 shows the basic sequence of the message shown in FIG. Here, attention is paid to the terminal 3. When no failure occurs in the main server A, the call setting request message transmitted from the terminal 3 to the server is sent to the main server A, and communication is normally performed. Here, if a failure occurs in the main server or the connection line of the main server, the connection to the server is disconnected, and the terminal 3 retries the call setting request message transmission. The retryed message is connected to the spare server B because the routing table in the exchange SW2 has been changed as shown in FIGS. 5 and 6.

[0017]

As described above, according to the present invention, a specification used in the conventional SVC connection establishment called a call setup rejection message in an ATM-NW capable of handling only a local failure due to hop-by-hop routing. By extracting the information of the reason display part of the message based on
By detecting a remote failure and selecting an alternative route, the failure can be avoided and the SVC connection can be established in a short time.

[Brief description of drawings]

FIG. 1 is an explanatory view of a routing table changing device of the present invention and its operation.

FIG. 2 is an explanatory diagram of a failure avoidance method when the present invention is used.

FIG. 3 is a diagram showing a routing table in FIG.

4 is a diagram showing a message sequence for establishing a communication path in FIG.

FIG. 5 is an explanatory view of a routing table changing device and its operation in a LANE server redundant configuration.

FIG. 6 is a diagram showing a routing table in FIG.

FIG. 7 is a diagram showing a message sequence in the case of FIG.

[Explanation of symbols]

101 ... Message processing unit, 102 ... Call number / destination address storage unit, 103 ... Table control unit, 104 ... Resource management unit, 105
… Routing table, 106… Timer (106)

─────────────────────────────────────────────────── ─── Continuation of front page (56) Reference JP-A-9-294138 (JP, A) JP-A-10-23074 (JP, A) JP-A-9-261223 (JP, A) JP-A-10- 3440 (JP, A) JP 11-112504 (JP, A) JP 11-308236 (JP, A) JP 12-69044 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) H04L 12/56

Claims (3)

(57) [Claims] 1. In an ATM network that implements LAN emulation with a 2N spare configuration of a server in which a working and a spare LAN emulation server are installed in different ATM switches, the ATM switch accommodating the working server has a failure of the working server. When detecting the above, the ATM switch returns a call setup reject message in which "reason terminal unreachability" information is set in the reason indicating information element in response to the call setup request message from the LEC for setting up the connection, A routing table changing method, characterized in that an exchange containing the LEC which receives the message selects a port for transmitting the message to the spare server and transmits the call setup request message. 2. In a routing table changing method in an exchange used in a communication system for establishing a connection from a communication request to a destination and performing communication, the exchange sets a call setting request message for establishing a connection. When the message is received, the call number and the destination address included in the message are stored, and when the routing table is referred to when routing the message, the route entry for the destination address included in the received call setting request message is stored in the network. If it is not possible to select due to a failure or abnormal port of the exchange, when sending a call setup rejection message, the message "remote terminal is unreachable"
When information is inserted and returned, and the exchange receives a call setup reject message containing "unreachable terminal unreachable" information, the exchange stores the destination address included in the call setup request message for the received call setup reject message. The entry for the route that received the call setup rejection message in the routing table is set to the unselectable state and removed from the routing options, and at the same time, the "unable to reach the other terminal" information is included as the reason for canceling the transmission. A routing table changing method characterized by indicating that a call setup refusal message has been received, and after a lapse of a certain period of time, an entry made unselectable by receiving the call setup refusal message is returned to the selectable state again. 3. A routing table changing device in an exchange used in a communication system which establishes a connection to a destination and communicates after an information transmission request is generated, and a process of inserting information into a received message. A message processing unit to perform, a call number / destination address storage unit for storing a call number and a destination address when a call setting request message is received, a table control unit for controlling a routing table, and a resource of an exchange. It has a resource management unit that manages the network, a routing table for referring to the route for sending a message, and a timer that measures the holding time when the routing table is changed. When an error is detected, table control of port error notification including the port number When the switch receives the call setup request message, the table control unit disables all entries in the routing table including the port in which the error is detected based on the port error notification received from the resource management unit. It is sent to the message processing unit, and the message processing unit sends the call number and the called address in the message to the call number / called address storage unit, and the call number / called address storage unit calls the call sent from the message processing unit. The pair of number and destination address is stored as information, and the message processing unit refers to the routing table to search the route for transmitting the received call setting request message, and the entry of the route for transmitting the message cannot be selected. If it is, add the “unreachable terminal” information to the generated call setup rejection message and send it. If the exchange deletes the information in the call number / destination address storage section for the call number of the message and the exchange receives a call setup reject message with "unreachable terminal" information added, that message is sent to the message processing section. Then, the message processing unit refers to the call number / destination address storage unit to determine the destination address of the call setup request message for the received call setup rejection message, and then "unreachable terminal" from the received call setup rejection message. When the information is detected, the information in the call number / destination address storage section for the call number of the message is erased, and the port number that received the call setup rejection message and the destination address that was calculated earlier are sent to the table control section. Based on the port number received from the processing unit and the destination address calculated, Check the table to make the entry of the target port unselectable, start its own timer, and after the timer measures the elapse of a certain time, report to the table controller that the timer has stopped However, the table control unit rewrites the routing table again after the timer expires, and restores the entries that were made unselectable due to the reception of the call setup rejection message to the selectable state again. apparatus.