Method and system for processing RapidIO network main and standby synchronization exception

Abstract

The invention provides a processing method and system for a RapidIO network main and standby synchronization abnormity. The processing method comprises the following steps: identifying a stage in which an abnormity occurs in a RapidIO main network; aiming at anomalies of the RapidIO main network in different stages, a standby network manager in the standby network adopts different processing methods to perform corresponding processing; wherein when the RapidIO main network is abnormal in the initialization stage, the standby network manager scans the whole RapidIO network again; and when the RapidIO main network is abnormal in the operation stage, the standby network manager ensures the integrity of RapidIO network information through information synchronized by the main network manager and one-time state scanning. According to the method, the running stages in the RapidIO network are classified, and the processing methods of the standby network management in the standby network are different under the condition that the RapidIO main network management in different stages is abnormal, so that the completeness of RapidIO network information is ensured, and the stability of the system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of RapidIO network, and in particular to a method and system for processing abnormal synchronization of a master and a standby of a RapidIO network. Background Art

[0002] In embedded system applications, all hardware components are connected based on bus mode. Due to the limitations of transmission rate and system topology, traditional interconnection buses can no longer meet the growing needs of processor frequency and performance. Therefore, in embedded environments, RapidIO interconnection architecture is basically adopted. RapidIO interconnection architecture is compatible with communication processors, digital signal processors, etc., and meets the requirements of high-performance embedded industry for reliability, high bandwidth, and high speed in system internal interconnection.

[0003] In the current integrated electronic information system, RapidIO technology can achieve high-speed data transmission between various processors in the system, supporting multiple rates, such as 1.25Gbaud, 5.0Gbaud, 6.25Gbaud, and 1X\2X\4X modes. The integrated electronic information system contains multiple PPC, DSP, and FPGA processors, which are connected in series by one or more domestic SRIO switching chips to form a RapidIO-based network system.

[0004] In large-scale integrated electronic information systems, in order to maintain the stability of the system, the RapidIO active-standby hot backup method is adopted to ensure that the RapidIO network can quickly switch to the backup RapidIO network when the main device fails, thereby minimizing downtime and reducing the risk of data loss. The active-standby hot backup of the RapidIO network system requires that the primary and backup devices have the same configuration, including the settings of the RapidIO interface, to ensure that they can be replaced seamlessly. In the RapidIO network active-standby hot backup system, the active and standby network managers in the active and standby RapidIO networks synchronize data in real time through the RapidIO bus. The heartbeat detection is performed between the active and standby RapidIO network managers. Once the standby network manager fails to detect the heartbeat of the RapidIO main network within a certain period of time, the RapidIO standby network manager is triggered to take over the operation of the entire RapidIO system.

[0005] The disadvantages of RapidIO network system active / standby hot backup are as follows:

[0006] (1) The RapidIO main network is abnormal during the initialization phase, such as the RapidIO node is not in the main system and cannot be synchronized to the RapidIO backup network management. Abnormal initialization will cause a certain degree of data loss.

[0007] (2) The RapidIO main network is abnormal during operation. During this period, a RapidIO node attempts to access the network. The main network cannot handle it and cannot synchronize it to the backup network manager. Once the backup network manager takes over the entire network, the RapidIO node information is lost. Summary of the invention

[0008] The present invention aims to provide a method and system for processing abnormal synchronization of the primary and standby of a RapidIO network, so as to overcome the above-mentioned disadvantages of the primary and standby hot backup of the RapidIO network system.

[0009] The present invention provides a method for processing a RapidIO network master / slave synchronization anomaly, comprising:

[0010] Identify the stage where the RapidIO main network is abnormal;

[0011] The backup network management system in the backup network uses different processing methods to handle abnormalities at different stages of the RapidIO primary network.

[0012] Furthermore, when the RapidIO main network is abnormal during the initialization phase, the backup network manager is asked to rescan the entire RapidIO network to ensure the integrity of the RapidIO network information.

[0013] Furthermore, when the RapidIO main network is abnormal during operation, the backup network manager ensures the integrity of the RapidIO network information through information synchronized with the main network manager and a status scan.

[0014] In one embodiment, when the processing method is used to initialize multiple SRIO domestic switching chips, the following steps are included:

[0015] S100, after the system is powered on, the backup network manager of the backup network waits for three types of information from the primary network manager: RapidIO node information, periodic heartbeat information, and primary network manager operation stage information;

[0016] S200: If the backup network manager does not receive the heartbeat information from the primary network manager within a period of time, it is considered that the primary network manager is abnormal and can take over the entire RapidIO network.

[0017] S300, judging the stage where the main network management is abnormal according to the information of the main network management operation stage:

[0018] If the active NM fails during the initialization phase, the standby NM discards the received RapidIO node information and rescans the entire RapidIO network;

[0019] If the active network management system encounters an exception during operation, that is, the standby network management system receives the initialization completion information, the standby network management system uses the synchronized RapidIO node information to manage the entire RapidIO network. Then, the standby network management system scans the RIOPort Error and Status CSR port link status of all RapidIO switch chips. If any link status is Link, the standby network management system checks whether there is corresponding RapidIO node information in the synchronized RapidIO node information. If so, no processing is performed. If not, the RapidIO node needs to be added to the network.

[0020] Furthermore, in step S300, if the standby network manager does not receive the running phase information of the main network manager or receives the initialization start information, then an abnormality occurs in the initialization phase of the main network manager.

[0021] Furthermore, in step S300, if the standby network manager receives the initialization completion information, an abnormality occurs in the active network manager during the operation phase.

[0022] The present invention also provides a RapidIO network master-slave synchronization abnormality processing system, comprising:

[0023] The abnormal phase identification module is used to identify the phase where the RapidIO main network is abnormal;

[0024] The processing module is used to process the abnormalities of the RapidIO main network at different stages by using different processing methods by the backup network management in the backup network.

[0025] Furthermore, the processing module includes:

[0026] The first processing module is used to ensure the integrity of RapidIO network information by letting the backup network manager rescan the entire RapidIO network when the RapidIO main network is abnormal during the initialization phase;

[0027] The second processing module is used for ensuring the integrity of RapidIO network information by performing a status scan through information synchronized by the main network manager when the main RapidIO network is abnormal during operation.

[0028] In one embodiment, when the processing system is used to initialize multiple SRIO domestic switching chips, the execution process is as follows:

[0029] S100, after the system is powered on, the backup network manager of the backup network waits for three types of information from the primary network manager: RapidIO node information, periodic heartbeat information, and primary network manager operation stage information;

[0030] S200: If the backup network manager does not receive the heartbeat information from the primary network manager within a period of time, it is considered that the primary network manager is abnormal and can take over the entire RapidIO network.

[0031] S300, the abnormal stage identification module determines the stage at which the main network management is abnormal according to the main network management operation stage information:

[0032] If the active network management system fails during the initialization phase, the first processing module in the standby network management system discards the received RapidIO node information and rescans the entire RapidIO network.

[0033] If an exception occurs in the active network management during operation, that is, the standby network management receives the initialization completion information, the standby network management uses the synchronized RapidIO node information to manage the entire RapidIO network; then, the second processing module in the standby network management scans the link status of all RapidIO switch chips RIO Port Error and Status CSR ports. If the link status is Link, check whether there is corresponding RapidIO node information in the synchronized RapidIO node information. If so, no processing is performed; if not, the RapidIO node needs to be connected to the network.

[0034] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0035] By classifying the operation stages in the RapidIO network, the backup network manager in the backup network uses different processing methods for abnormal conditions of the RapidIO primary network manager at different stages, thus ensuring the integrity of the RapidIO network information and improving the stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 The present invention is a flowchart of a method for processing a RapidIO network master / slave synchronization anomaly in an embodiment of the present invention.

[0037] Figure 2 The flowchart is an application example of a method for processing a RapidIO network master / slave synchronization exception in an embodiment of the present invention.

[0038] Figure 3 The diagram is a schematic diagram of a system for processing a RapidIO network master / slave synchronization anomaly according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0040] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] Example

[0042] like Figure 1 As shown, this embodiment provides a method for processing a RapidIO network master / slave synchronization exception, including:

[0043] Identify the stage where the RapidIO main network is abnormal;

[0044] The backup network management system in the backup network uses different processing methods to handle abnormalities at different stages of the RapidIO primary network.

[0045] Therefore, by classifying the operation stages in the RapidIO network, the backup network manager in the backup network uses different processing methods for abnormal conditions of the RapidIO primary network manager at different stages, thereby ensuring the integrity of the RapidIO network information and improving the stability of the system.

[0046] Correspondingly, the phases in which the RapidIO main network is abnormal include the initialization phase and the operation phase, and the processing is as follows:

[0047] When the RapidIO main network fails during the initialization phase, the system switches to the backup network, and some RapidIO node information is lost. The backup network manager is asked to rescan the entire RapidIO network to ensure the integrity of the RapidIO network information.

[0048] When an abnormality occurs when a node accesses the RapidIO main network during the operation phase, the backup network manager uses the information synchronized by the main network manager and a status scan to ensure the integrity of the RapidIO network information.

[0049] Figure 2 An application example of the method for handling the abnormal synchronization of the master and standby of the RapidIO network is shown, and the initialization of multiple SRIO domestic switching chips includes the following steps:

[0050] S100, after the system is powered on, the backup network manager of the backup network waits for three types of information from the primary network manager: RapidIO node information, periodic heartbeat information, and primary network manager operation stage information;

[0051] S200, if the standby network manager does not receive the heartbeat information from the active network manager within a period of time (which can be set as needed), it is considered that the active network manager is abnormal and can take over the entire RapidIO network;

[0052] S300, judging the stage where the main network management is abnormal according to the information of the main network management operation stage:

[0053] If the active NM fails during the initialization phase, that is, the standby NM does not receive the active NM running phase information or the initialization start information, the standby NM discards the received RapidIO node information and rescans the entire RapidIO network;

[0054] If the active network management system encounters an exception during operation, that is, the standby network management system receives the initialization completion information, the standby network management system uses the synchronized RapidIO node information to manage the entire RapidIO network. Then, the standby network management system scans the RIOPort Error and Status CSR port link status of all RapidIO switch chips. If any link status is Link, the standby network management system checks whether there is corresponding RapidIO node information in the synchronized RapidIO node information. If so, no processing is performed. If not, the RapidIO node needs to be added to the network.

[0055] Corresponding to the above-mentioned RapidIO network master / slave synchronization abnormality processing method, this embodiment also provides a RapidIO network master / slave synchronization abnormality processing system, such as Figure 3 As shown, including:

[0056] The abnormal phase identification module is used to identify the phase where the RapidIO main network is abnormal;

[0057] The processing module is used to process the abnormalities of the RapidIO main network at different stages by using different processing methods by the backup network management in the backup network.

[0058] Correspondingly, the phases in which the RapidIO main network is abnormal include an initialization phase and an operation phase, and thus, the processing module includes:

[0059] The first processing module is used to ensure the integrity of RapidIO network information by letting the backup network manager rescan the entire RapidIO network when the RapidIO main network is abnormal during the initialization phase;

[0060] The second processing module is used for ensuring the integrity of RapidIO network information by performing a status scan through information synchronized by the main network manager when the main RapidIO network is abnormal during operation.

[0061] The specific application of the processing system for the abnormal synchronization between the master and standby of the RapidIO network can refer to the above. The processing system can be set in the standby network management of the standby network. In the same application example, the execution flow of the processing system for initializing multiple SRIO domestic switching chips is as follows:

[0062] S100, after the system is powered on, the backup network manager of the backup network waits for three types of information from the primary network manager: RapidIO node information, periodic heartbeat information, and primary network manager operation stage information;

[0063] S200, if the standby network manager does not receive the heartbeat information from the active network manager within a period of time (which can be set as needed), it is considered that the active network manager is abnormal and can take over the entire RapidIO network;

[0064] S300, the abnormal stage identification module determines the stage at which the main network management is abnormal according to the main network management operation stage information:

[0065] If the active network management system fails during the initialization phase, that is, the standby network management system does not receive the active network management system operation phase information or the initialization start information, the first processing module in the standby network management system discards the received RapidIO node information and rescans the entire RapidIO network.

[0066] If the active network management encounters an abnormality during the operation phase, that is, the standby network management receives the initialization completion information, the standby network management uses the synchronized RapidIO node information to manage the entire RapidIO network; then, the second processing module scans the RIO Port Error and Status CSR port link status of all RapidIO switch chips. If the link status is Link, check whether there is corresponding RapidIO node information in the synchronized RapidIO node information. If so, no processing is performed; if not, the RapidIO node needs to be connected to the network.

[0067] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for processing abnormal synchronization between a master and a standby in a RapidIO network, characterized in that: include: Identify the stage where the RapidIO main network is abnormal; The backup network management system in the backup network uses different processing methods to handle abnormalities at different stages of the RapidIO primary network.

2. The method for processing RapidIO network master / slave synchronization anomaly according to claim 1, characterized in that: When the RapidIO main network fails during the initialization phase, the backup network manager can rescan the entire RapidIO network to ensure the integrity of the RapidIO network information.

3. The method for processing the abnormal synchronization between the master and standby nodes of a RapidIO network according to claim 1, characterized in that: When the RapidIO main network is abnormal during operation, the backup network manager uses the information synchronized with the main network manager and a status scan to ensure the integrity of the RapidIO network information.

4. The method for processing the abnormality of the master / slave synchronization of the RapidIO network according to any one of claims 1 to 3, characterized in that: When the processing method is used to initialize multiple SRIO domestic switching chips, the following steps are included: S100, after the system is powered on, the backup network manager of the backup network waits for three types of information from the primary network manager: RapidIO node information, periodic heartbeat information, and primary network manager operation stage information; S200: If the backup network manager does not receive the heartbeat information from the primary network manager within a period of time, it is considered that the primary network manager is abnormal and can take over the entire RapidIO network. S300, judging the stage where the main network management is abnormal according to the information of the main network management operation stage: If the active NM fails during the initialization phase, the standby NM discards the received RapidIO node information and rescans the entire RapidIO network; If the active network management system encounters an exception during operation, that is, the standby network management system receives the initialization completion information, the standby network management system uses the synchronized RapidIO node information to manage the entire RapidIO network. Then, the standby network management system scans the link status of all RapidIO switch chips RIO Port Error and Status CSR ports. If the link status is Link, the standby network management system checks whether there is corresponding RapidIO node information in the synchronized RapidIO node information. If so, no processing is performed. If not, the RapidIO node needs to be connected to the network.

5. The method for processing the abnormal synchronization between the master and standby nodes of a RapidIO network according to claim 4, characterized in that: In step S300, if the standby network manager does not receive the running phase information of the active network manager or receives the initialization start information, then an abnormality occurs in the initialization phase of the active network manager.

6. The method for processing the abnormal synchronization between the master and standby nodes of a RapidIO network according to claim 4, characterized in that: In step S300, if the standby network manager receives the initialization completion information, an abnormality occurs in the active network manager during the operation phase.

7. A RapidIO network master-slave synchronization abnormality processing system, characterized in that: include: The abnormal phase identification module is used to identify the phase where the RapidIO main network is abnormal; The processing module is used to process the abnormalities of the RapidIO main network at different stages by using different processing methods by the backup network management in the backup network.

8. The RapidIO network master / slave synchronization abnormality processing system according to claim 7, characterized in that: The processing module comprises: The first processing module is used to ensure the integrity of the RapidIO network information by allowing the backup network manager to rescan the entire RapidIO network when the RapidIO main network is abnormal during the initialization phase.

9. The RapidIO network master / slave synchronization abnormality processing system according to claim 7, characterized in that: The processing module comprises: The second processing module is used for ensuring the integrity of RapidIO network information by performing a status scan through information synchronized by the main network manager when the main RapidIO network is abnormal during operation.

10. The RapidIO network master / slave synchronization abnormality processing system according to any one of claims 7 to 9, characterized in that: When the processing system is used to initialize multiple SRIO domestic switching chips, the execution process is as follows: S100, after the system is powered on, the backup network manager of the backup network waits for three types of information from the primary network manager: RapidIO node information, periodic heartbeat information, and primary network manager operation stage information; S200: If the backup network manager does not receive the heartbeat information from the primary network manager within a period of time, it is considered that the primary network manager is abnormal and can take over the entire RapidIO network. S300, the abnormal stage identification module determines the stage at which the main network management is abnormal according to the main network management operation stage information: If the active network management system fails during the initialization phase, the first processing module in the standby network management system discards the received RapidIO node information and rescans the entire RapidIO network. If an exception occurs in the active network management during operation, that is, the standby network management receives the initialization completion information, the standby network management uses the synchronized RapidIO node information to manage the entire RapidIO network; then, the second processing module in the standby network management scans the link status of all RapidIO switch chips RIO Port Error and Status CSR ports. If the link status is Link, check whether there is corresponding RapidIO node information in the synchronized RapidIO node information. If so, no processing is performed; if not, the RapidIO node needs to be connected to the network.

Images