JP2000069019A - Redundant method of ATM signaling protocol software - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To realize a dual protocol stack by handling only external interfaces and handling only external interfaces without relying on hardware and without intervening inside provided software or packaged products. The present invention provides a method for duplicating ATM signaling protocol software. In a network device, a signaling event detected in an active system is transferred to a standby system, and an equivalent signaling event is generated in the standby system to simulate a standby software to use a standby protocol stack. The system is made to be in the same state as the system, and switching of the working spare due to a failure or the like is performed quickly and without disconnecting the call. Further, by establishing synchronization with the protocol stack that simulates the lower layer of the partner device facing the protection side protocol, the information of the working side that is taken over without causing a failure due to the protocol synchronization deviation is retained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an asynchronous transfer mode (hereinafter referred to as ATM)
(Switched Virtual Connect
The present invention relates to a method of operating arbitrary software for realizing a signaling protocol procedure in a signaling function for realizing the ion function.

[0002]

2. Description of the Related Art In order to duplicate a communication protocol in an ATM network, information on a protocol stack that is currently operating and is changing sequentially is reflected on a spare protocol stack at any time, and when a failure occurs, the protocol stack is switched to an adjacent device. It is necessary to recover so that the failure is not noticed.

When the protocol is implemented by software, the state of the working / spare memory (data) and the state of the task are matched, and the state of the already established SVC call and the state of the link connection are held in the spare. Need to be kept.

[0004] First, the content of the standby protocol stack is duplicated by duplicating the CPU and the memory and duplicating the active / standby memory by the active CPU, and when the active side fails, it is switched in hardware. There is a way. This method is effective because it depends heavily on hardware and does not depend on software, but it is highly dependent on vendor equipment and has a tightly coupled system configuration, making it difficult to use in a distributed system. Further, it is difficult to realize this in a system having working / spare in the same apparatus (for example, realizing a working / spare protocol stack by dividing a logical space).

[0005] Next, there is a method for realizing duplication by directly taking out data in the working protocol stack and copying it to a spare area. With this method, a program that intervenes inside software that implements the relevant protocol, examines and understands the data structure, and copies the necessary data to a spare stack area must be considered. It is not a means of duplicating any protocol software, but requires extensive rework and investigation every time the installed software changes. In particular, when the data area is software that uses not only the static area but also the memory area secured from the OS, this method is almost difficult.

[0006]

The ATM signaling function is the ATM Forum UNI 3.0 / 3.1 / 4.
Any function that implements the communication protocol specified by 0 is available, and software that implements the protocol is sold and sold in the form of packages or sources from many vendors. When constructing a system, consider combining these software with functions, and without intervening inside the software that implements the protocol,
It is necessary to handle only the external interface to realize the duplex of the protocol stack.

The ATM signaling protocol is generally realized by the following three protocol stacks.

[0008] SSCOP (Service Specific Connect)
ion Oriented Protocol): Implements a protocol for establishing a layer 2 level (lower layer) connection with the other party, that is, a path for transmitting and receiving data in line units. In addition to transmitting and receiving periodic confirmation packets, the reliability of the link is constantly confirmed, and control for sequentially transmitting data in response to a data transmission request from an upper layer is contracted.

[0009] Q93B / Q2931: A protocol for interpreting and realizing signaling messages at the level 3 for establishing / releasing SVC calls. After the layer 2 level with the other party is established in SSCOP, this protocol sends and receives messages to and from each other and performs a series of procedures for establishing / releasing SVC calls. Although the name changes depending on the UNI version, it is hereinafter referred to as Q93B for convenience.

[0010] Upper scenario (SCC: Switched Call)
Control): Control to execute a signaling operation scenario above Q93B. In the case of a switch, it controls where the line is routed in response to a call establishment request (Setup) from a certain line side, and in the case of a terminal, it controls whether or not to accept the call establishment request. This is the top scenario part.

In order to realize a duplex signaling protocol stack, it is necessary for each of these three stacks to successively hold the same information as that of the working side and to operate and hold it in a thermal reserve state.

An object of the present invention is to provide an ATM signaling protocol software which realizes duplication of a protocol stack by satisfactorily handling only an external interface without intervening in software or packaged products provided in consideration of maintainability and man-hours. It is to provide a duplication method.

[0013]

In order to achieve the above-mentioned object, at the same time as the system is started, the standby protocol stack is also initialized in the same manner as the active system, and is operated in the event waiting state. The signaling protocol transmits and receives a signaling event (packet) from the line side, processes the event and follows a protocol procedure to realize a call establishment / release sequence. A similar signaling event is transferred to the protection system when the call is established / released on the working side, and is injected into the protection system protocol stack, whereby a call equivalent to the working side is pseudo-established / released. Also, a stack that simulates the protocol stack of the partner device with which the working protocol stack exchanges (hereinafter referred to as “pseudo-protocol stack” for convenience) is added to the standby system, and is also opposed to the standby protocol stack. Synchronization is also simulated. This prevents failure detection due to protocol out-of-synchronization, and makes it possible to make the standby-side protocol stack wait in almost the same state as that of the active protocol stack.

If the active side becomes faulty, the transmission / reception route of packets on the standby side is switched from the pseudo protocol stack to the partner apparatus, and continuous operation can be performed without disconnecting the call.

[0015]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 shows a network device system 00 to which the present invention is applied and a general network device system 01.
And are connected to each other by a transmission line (line) K2. In both systems, signaling protocol stacks for realizing SVC are S0, S
There is one. For a system 01 without the present invention, this stack is active only. Normally, data flows between the two protocol stacks through the transmission line K2, and the establishment and release of the SVC call are realized. Actually, the signal from the transmission path reaches the SSCOP via the physical layer, the ATM layer, and the ATM adaptation layer, but is omitted here. The dotted arrows in the figure indicate the flow of data / processing.

In the system 00, a standby protocol stack S2 is generated in the same manner as the active S1, and is initialized and set to a program operable state. Furthermore, SSCOP is used as a simulation protocol stack to simulate the partner system.
S3 mounted with only the same is generated in the same manner. S2 and S
3 is a packet transmission line K1 that both SSCOPs transmit and receive.
Prepare Note that this transmission path is software-like, and the data transmitted by the other SSCOP may be simply injected directly into the other SSCOP, and there is no need to intervene a normal physical layer or the like at all.

The "pseudo Q93B" in S3 is a control for easily generating a signaling message for requesting call establishment / release.

The procedure for reflecting the call information in the backup protocol stack is as follows.

(1) A signaling protocol operates between S0 and S1 to establish a call.

(2) Upon the above-mentioned call establishment, an event is notified from the active system to the standby system, and the pseudo Q93B in S3 transmits a message (Setup) of call establishment.

(3) The message is P31 → P30 → K
1 → P20 → P21 → P22 Responses are processed in the reverse order.

(4) As a result, the protocol stack S2
The data of the currently established call is generated in.

The procedure for releasing a call is similar to the above procedure.

Here, the pseudo Q93B is simple, but must generate a series of signaling messages to realize a call establishment / release sequence. However, it is sufficient to process each call one by one, and since it is known that the processing is normally completed, a very simple protocol is sufficient.

In the spare stack S2, the SSCOP packet is simply injected and the processing is performed as it is in the current operation. Therefore, software (SSCOP, Q93B, SCC) for realizing each protocol can be used as it is without any modification.

The reason that only the SSCOP is provided with the normal protocol stack in S3 is because an abnormality is not detected in the health check (that is, the procedure of periodically transmitting confirmation packets and confirming the partner state) performed in the SSCOP. Since the SSCOP data transmission / reception is realized by strict sequence control, only SSCOP is mounted on S3 and the sequence control is performed by a regular procedure.
This is because the “pseudo Q93B” prepared by itself can be simplified.

The information of the call generated in the active S1 by the above procedure is reflected on the standby side S2 as needed, and the standby side can be switched at any time without causing an abnormality (health check error) due to loss of synchronization on the standby side. Can hold.

Next, a procedure in which the active side becomes a failure and switches will be described with reference to FIG. When the working protocol stack S1 fails, S2 becomes a new working. Further, the transmission line K1, which previously connected S3 and S2, is disconnected and the entry to the transmission line K2 is switched to S2 so that packets from the system 01 can be transmitted and received to S2. In S2, information of the transferred call is stored in each protocol (P20, P21, P2).
22), and the call can be held as it is without any information mismatch with the opposite system 01.

However, depending on the timing, SSCOP
In some cases, the sequence number of the health check of the SSCOP does not match. When this is detected, SSCOP resynchronizes and matches the sequence number. Along with that Q93
Although the call state confirmation sequence operates at the B level, as described above, since the information is consistent between the systems, both the established calls are held as they are and are not disconnected.

[0031]

According to the present invention, it is possible to duplicate software for realizing an arbitrary ATM signaling protocol without intervening therein, and to quickly and without a call disconnection in the event of a failure on the working side. It is possible to switch.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating the configuration and operation of an embodiment of the present invention.

FIG. 2 is a diagram illustrating an operation when a working protocol stack is switched to a spare due to a failure in an embodiment of the present invention.

[Explanation of symbols]

00, 01 ... network device system, S0, S1 ...
Active signaling protocol stack, S2: standby signaling protocol stack, S3: partner signaling protocol stack, P00, P10, P2
0, P30 ... SSCOP, P01, P11, P21 ... Q
93B, P02, P12, P22 ... SCC, P31 ... pseudo Q93B, K1, K2 ... data transmission paths.

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Aki Suzuki 810 Shimoimaizumi, Ebina-shi, Kanagawa F-term in Hitachi Server Co., Ltd. Server Development Division (Reference) 5K030 GA12 HA10 LB19 MD02

Claims (2)

[Claims] In a network device which implements arbitrary software for realizing an ATM signaling protocol and has a duplicated structure of a basic unit, a detection is made in an active system with respect to a standby (standby) side of a duplicated basic unit. By transferring the signaling event, and generating a signaling event inside the standby system and simulating the protection side software, the protection system is brought into the same protocol state as the working without modifying the software inside, A method for duplexing ATM signaling protocol software, wherein switching of a working spare due to a failure or the like is realized quickly and without disconnection of a call. 2. The working side which establishes synchronization of a lower layer by adding a function of simulating the operation of a partner device of a lower layer of the signaling protocol in a standby system, and takes over without causing a failure due to protocol synchronization deviation. A method for duplicating ATM signaling protocol software, characterized by holding the following information: