CN102752144B - Method and device for processing topology change in soft restart in-service software upgrade process - Google Patents

Abstract

The invention discloses a method and a device for processing topology change in a soft restart in-service software upgrade process. According to the scheme, in a soft restart process of an interface board, if a proxy board receives a topology change message, all STP (Spanning Tree Protocol) embodiments having the topology change are determined according to the topology change message, and topology change information corresponding to each STP embodiment having the topology change is recorded in an overall topology change state list; and after the soft restart process of the interface board, the proxy board refreshes a transfer list item corresponding to the STP embodiment of a port of the STP embodiment according to the topology change information corresponding to each STP embodiment in the overall topology change state list, and the topology change message is sent through the port of the STP embodiment. With the adoption of the method and the device provided by the invention, the transfer abnormal conditions caused by the topology change in the soft restart ISSU (In-Service Software Upgrade) process can be solved.

Description

Method and device for processing topology change in soft restart uninterrupted service upgrading process

Technical Field

The present application relates to the field of communications technologies, and In particular, to a method and an apparatus for processing topology changes during an In-service software Upgrade (ISSU) process.

Background

The Spanning Tree Protocol (STP), defined by IEEE802.1d standards, is a two-layer management Protocol for eliminating data link layer physical loops in a local area network. The equipment (STP equipment) running the STP protocol discovers loops in the network by mutually exchanging information, selectively blocks certain ports, and finally prunes the loop network structure into a loop-free tree network structure, thereby preventing messages from continuously increasing and infinitely circulating in the loop network and avoiding the problem of message processing capacity reduction caused by repeatedly receiving the same messages by the equipment.

The STP device performs role negotiation through a Bridge Protocol Data Unit (BPDU) message. And each STP equipment calculates and determines the role of the equipment in the STP network and the role of each STP port in the equipment according to the received BPDU message, so that the STP network is finally stabilized in a loop-free state. Topology changes exist before and after calculation, some forwarding table entries may be learned before topology stabilization, and a refreshing of the forwarding table entries needs to be performed on the whole network after topology stabilization. The STP equipment carries out notification of refreshing the forwarding table entry of the whole network through the topology change message, and the STP equipment and the port which receive the topology change message can actively carry out the cleaning and the refreshing on the forwarding table entry of the equipment and the port so as to ensure the accuracy of the forwarding table entry. The Topology Change message is a BPDU message carrying a Topology Change (TC) flag, or a Topology Change Notification (TCN) message.

The ISSU is a high-reliability software starting mode for upgrading equipment, and a series of methods are used for ensuring that the service is not interrupted or the interruption time is short in the upgrading process. The current common ISSU technologies mainly include: the method comprises a process level increment ISSU, a soft restart ISSU and a main/standby switch ISSU. The essence of the soft restart ISSU is that data (operation data, configuration data and hardware data) and states at the moment of system operation are all stored in a memory; restarting the CPU by using new software, wherein the hardware continues to provide forwarding capability during the period of restarting the CPU; and after the CPU is restarted, the CPU continues running by using the stored data and the state. And for the protocol message needing to be interacted with the opposite terminal equipment in real time to keep connection, ensuring that the connection and the protocol state are not influenced in the soft restart upgrading process through the protocol agent process.

The soft restart ISSU is generally performed by taking an interface board as a unit, a proxy board is used for service management, monitoring and proxy in the soft restart process, and software on the interface board of the soft restart cannot respond to external event processing.

Referring to fig. 1, fig. 1 is a schematic diagram of STP networking in the prior art, as shown in fig. 1, a total of 4 STP devices are included in an MST domain: equipment A, equipment B, equipment C, equipment D, equipment A is the root bridge, includes 4 ports: a1-a4, device B includes 3 ports: B1-B3; device C includes 4 ports: C1-C4, device D includes 3 ports: D1-D3, wherein the ports B1, C1, D2 are root ports, the ports A3, a4, B3, C2 are designated ports, the ports a2 and D2 are replacement ports, the port B2 is a backup port, the port D3 is an edge port, the port a1 is both a primary port and a domain boundary port, and the ports C3, C4 are domain boundary ports.

Assuming ports B1 and B3 on device B in fig. 1 are located on interface board 1, a soft restart upgrade needs to be performed on interface board 1. In the soft restart process of the interface board 1, if a link between the device C and the device D fails, the replacement port D1 on the device D will be directly converted into a new root port to enter a forwarding state, and send a TC message of a topology change notification to notify each device in the MST domain to refresh a forwarding table entry.

However, since the interface board on which the port B1 on the device B is located is performing soft restart, it is not able to process TC messages, so that the forwarding table entry on the device B after soft restart is inconsistent with the final topology of the STP network, and a service forwarding exception occurs. In addition, suppose there is a device E between the device B and the device C, at this time, the device C cannot notify the topology change message to the outside because the port C2 fails to trigger the port role change, and the device B cannot transmit the TC message to the device E because the interface board 1 is in soft restart, and the device E cannot sense the topology change, so that the forwarding table entry on the device E is inconsistent with the final topology of the STP network, and the service forwarding exception occurs.

Disclosure of Invention

In view of this, the present invention provides a method for processing topology change in a soft restart ISSU process, which can solve the technical problem of service forwarding exception caused by topology change in the soft restart ISSU process.

In order to achieve the above object, the present invention provides a method for processing topology change in a soft restart ISSU process, the method comprising:

in the soft restarting process of an interface board, when an agent board receives a topology change message, determining all Spanning Tree Protocol (STP) instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in a global topology change state table;

after the soft restart of the interface board is finished, the proxy board refreshes the forwarding table items corresponding to the STP instance on the port belonging to the STP instance according to the topology change information corresponding to each STP instance in the global topology change state table, and sends the topology change messages from the port belonging to the STP instance.

The method further comprises the following steps:

before the interface board is in soft restart, emptying a global topology change state table;

when the soft restart of the interface board is started, message interception table entries which are used for sending messages received by ports on the interface board to the proxy board for processing are set, wherein the message interception table entries comprise message characteristic information and proxy board numbers, and the message characteristic information comprises destination Media Access Control (MAC) addresses of the messages; when the port on the interface board receives the message, if the message accords with the message characteristic information in the message interception table item, the message is sent to the agent board according to the agent board number in the message interception table item.

Preferably, the topology change information corresponding to the STP instance includes the STP instance identifier, or the STP instance identifier and port information;

the topology change message received by the proxy board is a Bridge Protocol Data Unit (BPDU) message carrying a topology change TC mark;

the method for determining all the STP instances with topology changes according to the topology change messages and recording the topology change information corresponding to each STP instance with topology changes in the global topology change state table comprises the following steps: determining the STP instance to which the TC mark carried in the topology change message belongs, and adding the identifier of the STP instance or the identifier of the STP instance and port information for receiving the topology change message in a global topology change state table aiming at the STP instance to which each TC mark belongs;

when the topology change information corresponding to the STP instance includes the STP instance identifier and the port information, before the sending the topology change packet from the port belonging to the STP instance, the method further includes: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

Preferably, the topology change information corresponding to the STP instance includes the STP instance identifier, or the STP instance identifier and port information;

the topology change message received by the agent board is a topology change notification TCN message;

the method for determining all the STP instances with topology changes according to the topology change messages and recording the topology change information corresponding to each STP instance with topology changes in the global topology change state table comprises the following steps: determining that the STP instance with the topology change is 0 instance, adding the identifiers of all STP instances in a global topology change state table, or the identifier of each STP instance in all STP instances and port information for receiving the topology change message;

when the topology change information corresponding to the STP instance includes the STP instance identifier and the port information, before the sending the topology change packet from the port belonging to the STP instance, the method further includes: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

Preferably, the topology change information corresponding to the STP instance includes the STP instance identifier, or the STP instance identifier and port information;

the topology change message received by the agent board is a topology change notification TCN message;

the method for determining all the STP instances with topology changes according to the topology change messages and recording the topology change information corresponding to each STP instance with topology changes in the global topology change state table comprises the following steps: determining that the STP instance with the topology change is a 0 instance, and adding the identifier of the STP instance or the identifier of the STP instance and port information for receiving the topology change message in a global topology change state table;

before the refreshing the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance, the method further includes: and judging whether the STP instance is a 0 instance, if so, refreshing forwarding table entries corresponding to all STP instances on all ports, and otherwise, refreshing the forwarding table entries corresponding to the STP instances on the ports belonging to the STP instance.

When the topology change information corresponding to the STP instance includes the STP instance identifier and the port information, before the sending the topology change packet from the port belonging to the STP instance, the method further includes: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

When the topology change information corresponding to the STP instance includes the STP instance identifier and the port information, before the recording the topology change information corresponding to each STP instance having topology change in the global topology change state table, the method further includes: judging whether the topology change information corresponding to the STP instance exists in the global topology change state table, if so, updating the port information in the topology change information corresponding to the STP instance to the port information of the topology change message carrying the STP instance information, and if not, recording the topology change information corresponding to the STP instance in the global topology change state table.

The invention also provides a device for processing topology change in the process of soft restart ISSU, which comprises: interface board and agent board, the agent board includes: the device comprises a message receiving unit, a message processing unit, a table item refreshing unit and a message sending unit;

the message receiving unit is used for receiving a topology change message in the soft restart process of the interface board;

the message processing unit is used for determining all STP instances with topology change according to the topology change message when the message receiving unit receives the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table;

the table entry refreshing unit is used for refreshing the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance according to the topology change information corresponding to each STP instance in the global topology change state table after the soft restart of the interface board is finished;

and the message sending unit is used for sending the topology change message from the port belonging to the STP instance according to the topology change information corresponding to each STP instance in the global topology change state table after the soft restart of the interface board is finished.

The interface board comprises a message uploading unit;

the message processing unit is further configured to empty the global topology change state table before the interface board is soft restarted;

the message uploading unit is used for setting a message interception table entry for uploading a message received by a port on the interface board to the proxy board for processing when the soft restart of the interface board is started, wherein the message interception table entry comprises message characteristic information and a proxy board number, and the message characteristic information comprises a destination MAC address of the message; and the message receiving unit is used for sending the message to the proxy board according to the proxy board number in the message interception table item if the message accords with the message characteristic information in the message interception table item when the port on the interface board receives the message.

In the above apparatus, the topology change information corresponding to the STP instance includes an identifier of the STP instance;

the topology change message received by the message receiving unit is a BPDU message carrying a TC mark;

the message processing unit is used for determining all STP instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table, and is used for: and determining the STP instance to which the TC mark carried in the topology change message belongs, and adding the STP instance identifier in a global topology change state table aiming at the STP instance to which each TC mark belongs.

In the above device, the topology change information corresponding to the STP instance includes an identifier of the STP instance and port information;

the topology change message received by the message receiving unit is a BPDU message carrying a TC mark;

when the message processing unit records topology change information corresponding to the STP instance to which each TC tag belongs, carried in the topology change message, in the global topology change state table, the message processing unit is configured to: determining the STP instance to which the TC mark carried in the topology change message belongs, adding the STP instance identifier in a global topology change state table aiming at the STP instance to which each TC mark belongs, and receiving port information of the topology change message;

before the packet sending unit sends the topology change packet from the port belonging to the STP instance, the packet sending unit is further configured to: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

In the above apparatus, the topology change information corresponding to the STP instance includes an identifier of the STP instance;

the topology change message received by the message receiving unit is a topology change notification TCN message;

the message processing unit is used for determining all STP instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table, and is used for: and determining that the STP instance with the topology change is a 0 instance, and adding the identifications of all the STP instances in the global topology change state table.

In the above device, the topology change information corresponding to the STP instance includes an identifier of the STP instance and port information;

the topology change message received by the message receiving unit is a topology change notification TCN message;

the message processing unit is used for determining all STP instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table, and is used for: determining that the STP instance with the topology change is a 0 instance, and adding the identifier of the STP instance and receiving the port information of the topology change message in a global topology change state table aiming at each STP instance in all the STP instances;

before the packet sending unit sends the topology change packet from the port belonging to the STP instance, the packet sending unit is further configured to: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

In the above apparatus, the topology change information corresponding to the STP instance includes an identifier of the STP instance;

the topology change message received by the message receiving unit is a topology change notification TCN message;

the message processing unit is used for determining all STP instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table, and is used for: determining that the STP instance with the topology change is a 0 instance, and adding the identifier of the STP instance in a global topology change state table;

before the table entry refreshing unit refreshes the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance, the table entry refreshing unit is further configured to: and judging whether the STP instance is a 0 instance, if so, refreshing forwarding table entries corresponding to all STP instances on all ports, and otherwise, refreshing the forwarding table entries corresponding to the STP instances on the ports belonging to the STP instance.

In the above device, the topology change information corresponding to the STP instance includes an identifier of the STP instance and port information;

the topology change message received by the message receiving unit is a topology change notification TCN message;

the message processing unit is used for determining all STP instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table, and is used for: the STP instance of the topology change message is 0 instance, the identification of the STP instance is added in a global topology change state table, and the port information of the topology change message is received;

before the table entry refreshing unit refreshes the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance, the table entry refreshing unit is further configured to: judging whether the STP instance is a 0 instance, if so, refreshing forwarding table entries corresponding to all STP instances on all ports, otherwise, refreshing the forwarding table entries corresponding to the STP instances on the ports belonging to the STP instance;

before the packet sending unit sends the topology change packet from the port belonging to the STP instance, the packet sending unit is further configured to: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

Before the packet processing unit records topology change information corresponding to each STP instance having topology change in the global topology change state table, the packet processing unit is further configured to: judging whether the topology change information corresponding to the STP instance exists in the global topology change state table, if so, updating the port information in the topology change information corresponding to the STP instance to the port information of the topology change message carrying the STP instance information, and if not, recording the topology change information corresponding to the STP instance in the global topology change state table.

According to the technical scheme, in the soft restart process of the interface board, when the proxy board receives the topology change message, the STP instance with the topology change is determined according to the topology change message, and the topology change information corresponding to each STP instance with the topology change is recorded to the global topology change state table; after the soft restart of the interface board is finished, the proxy board refreshes the forwarding table items corresponding to the STP instance on the port belonging to the STP instance according to the topology change information corresponding to each STP instance recorded in the global topology change state table, and sends the topology change messages to the outside from the port belonging to the STP instance. After the interface board is soft restarted, the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance is refreshed according to the topology change information of the STP instance recorded in the soft restarting process of the interface board, so that the problem of service forwarding abnormity caused by topology change in the ISSU process can be solved.

Drawings

Figure 1 is a schematic diagram of a prior art STP networking;

FIG. 2 is a flowchart of a method for processing topology change in a soft restart ISSU process according to an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of a processing apparatus for processing topology change during a soft restart ISSU according to an embodiment of the present invention;

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more clearly apparent, the technical solutions of the present invention are described in detail below with reference to the accompanying drawings and examples.

Referring to fig. 2, fig. 2 is a flowchart of a processing method for topology change in the process of soft restart ISSU according to an embodiment of the present invention, which mainly includes the following steps:

step 201, in the soft restart process of the interface board, when the proxy board receives the topology change message, determining all STP instances in which the topology change occurs according to the topology change message, and recording topology change information corresponding to each STP instance in which the topology change occurs in the global topology change state table.

In practical application, after detecting the topology change, the device may send a topology change message to notify of the refreshing of the forwarding table entry of the whole network. Meanwhile, topology change messages sent by other equipment after detecting topology changes can be received.

In this embodiment, when the interface board of the device is performing soft restart, the topology change packet cannot be processed. In order to avoid inconsistency between the forwarding table entry of each port after the soft restart of the interface board and the final topology of the network, the proxy board may be used to process the topology change packet instead, and record the topology change information corresponding to each STP instance in which the topology change occurs to the global topology change state table, so as to refresh the forwarding table entry on each port after the soft restart is completed.

Step 202, after the soft restart of the interface board is finished, the proxy board refreshes the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance according to the topology change information corresponding to each STP instance in the global topology change state table, and sends the topology change message from the port belonging to the STP instance.

After the interface board is in soft restart, the proxy board can determine the STP instance with the topology change according to the topology change information corresponding to the STP instance recorded in the global topology change state table, and then can refresh the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance with the topology change, so that the situation that the forwarding table entry corresponding to the STP instance on the port is inconsistent with the final topology of the network can be avoided.

Here, the refreshing of the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance includes two ways, one is to clear the renegotiation, and the other is to actively refresh, where clearing the renegotiation means to directly delete the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance; the active refresh means that an ARP request message is sent for each forwarding entry, if an ARP response message is received, the forwarding entry is retained, and if an ARP response message is not received within a certain time, the forwarding entry is deleted. This is in the prior art.

In the embodiment of the present invention shown in fig. 2, before the interface board is soft restarted, the global topology change state table also needs to be cleared first. In addition, because the message cannot be processed after the soft restart of the interface board, it is necessary to set a message interception table entry for sending the message received by the port on the interface board to the proxy board for processing when the soft restart of the interface board is started. Here, the message interception table entry includes message characteristic information and a proxy board number, where the message characteristic information includes a destination MAC of the message. Thus, when a message is received at a port on an interface board, if the message conforms to the message feature information in the message interception table entry (for example, if the destination MAC of the message is the same as the destination MAC in the message feature information, it indicates that the message conforms to the message feature information in the message interception table entry), the message is sent to the proxy board according to the proxy board number in the message interception table entry, so that the proxy board can process the message instead.

In practical applications, the topology change message can be divided into two types, one is a BPDU message carrying one or more TC tags, and the other is a TCN message. Each TC mark belongs to one STP instance in the BPDU message carrying the TC mark, when the device receives the BPDU message carrying the TC mark, the device can determine the STP instance with topology change according to the TC mark carried in the BPDU message, and then refresh the forwarding table entry corresponding to the STP instance with topology change after the soft restart of the interface board is finished. When the device receives the TCN message, it indicates that the STP instance with the topology change on the device is 0 instance, and at this time, forwarding table entries corresponding to all STP instances on all ports need to be refreshed.

In addition, in order to keep the forwarding table entry on the device consistent with the final topology of the network after the soft restart of the interface board of the device is finished, topology change information corresponding to the STP instance having the topology change needs to be recorded, and the topology change information may include an identifier of the STP instance having the topology change. After receiving the topology change message at a port, the port does not need to forward the topology change message outwards, and port information can be added to the topology change information corresponding to the STP instance based on the consideration of avoiding resource waste.

As can be seen from the above two paragraphs, according to the difference of the received topology change messages and the difference of the topology change information corresponding to the STP instance in which the topology change occurs, there may be at least the following four combinations:

the combination one, the topology change information corresponding to the STP instance includes the STP instance identification; the topology change message received by the agent board is a BPDU message carrying TC marks.

In this case, the method for determining all STP instances having topology changes according to the topology change packet and recording topology change information corresponding to each STP instance having topology changes in the global topology change state table includes: and determining the STP instance to which the TC mark carried in the topology change message belongs, and adding the identifier of the STP instance in a global topology change state table aiming at the STP instance to which each TC mark belongs.

Thus, after the soft restart of the interface board is finished, the proxy board can determine the STP instance with the topology change according to the STP instance identifier in the global topology change state table, then, for each STP instance with the topology change, refresh the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance, and send the topology change message to the outside from the port belonging to the STP instance.

The proxy board records the identifier of the STP instance with the topology change in the global topology change state table in the soft restart process of the interface board, and refreshes the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance according to the identifier of each STP instance recorded in the global topology change state table after the restart of the interface board is finished, so that the forwarding table entry of the STP instance is prevented from being inconsistent with the final topology of the network, and the technical problem of abnormal service forwarding caused by the topology change in the soft restart ISSU process of the interface board can be solved.

Combining two, the topology change information corresponding to the STP instance comprises the STP instance identification and the port information; the topology change message received by the agent board is a BPDU message carrying TC marks.

In this case, the method for determining all STP instances having topology changes according to the topology change packet and recording topology change information corresponding to each STP instance having topology changes in the global topology change state table includes: and determining the STP instance to which the TC mark carried in the topology change message belongs, and adding the identifier of the STP instance to which the TC mark belongs and receiving the port information of the topology change message in a global topology change state table aiming at the STP instance to which each TC mark belongs.

In addition, after receiving the topology change message at a certain port, the device does not need to forward the topology change message from the port. Therefore, before sending the topology change message from the port belonging to the STP instance, the method further includes: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

It should be noted that, when the topology change packet is a BPDU packet carrying a TC tag, and after the soft restart of the interface board is completed, the proxy board transmits the topology change packet from the port belonging to the STP instance according to the topology change information corresponding to each STP instance in the global topology change state table, and if a plurality of STP instances on the port have topology change, the transmitted topology change packet may carry a plurality of TC tags, where each TC tag belongs to one STP instance of the plurality of STP instances.

In addition, in the soft restart process of the interface board, for the same STP instance, if the proxy board receives the topology change messages for multiple times and can determine that the STP instance has topology change according to the topology change messages, only the port in the topology change information corresponding to the STP instance in the global topology change state table needs to be updated. Therefore, before the recording the topology change information corresponding to each STP instance having a topology change in the global topology change state table, the method further includes: judging whether the topology change information corresponding to the STP instance exists in the global topology change state table, if so, updating the port information in the topology change information corresponding to the STP instance to be the port information for receiving the topology change message, and if not, recording the topology change information corresponding to the TP instance in the global topology change state table.

Under the combination, the proxy board records the identifier of the STP instance with the topology change in the global topology change state table and the port information of the topology change message indicating the topology change of the STP instance for the last time in the soft restart process of the interface board, and refreshes the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance according to the identifier of each STP instance recorded in the global topology change state table after the restart of the interface board is finished, so that the inconsistency between the forwarding table entry of the STP instance and the final topology of the network can be avoided, and the technical problem of abnormal service forwarding caused by the topology change in the soft restart ISSU process of the interface board can be solved. In addition, the proxy board does not send the topology change message indicating that the STP has topology change from the port according to the recorded port information of the topology change message indicating that the STP instance has topology change, thereby reducing unnecessary resource waste.

Combining the topology change information corresponding to the STP instance with the STP instance identifier; the topology change message received by the agent board is a topology change notification TCN message.

In this case, the method for determining all STP instances having topology changes according to the topology change packet and recording topology change information corresponding to each STP instance having topology changes in the global topology change state table includes: the STP instance in which the topology change occurs is determined to be a 0 instance, because the forwarding table entries corresponding to all STP instances need to be refreshed, the identifiers of all STP instances can be added in the global topology change state table.

In fact, the method for determining all STP instances having topology changes according to the topology change packet and recording topology change information corresponding to each STP instance having topology changes in the global topology change state table may further be: and determining that the STP instance with the topology change is a 0 instance, and adding the identification of the STP instance in the global topology change state table. In this case, before the refreshing the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance, the method further includes: and judging whether the STP instance is a 0 instance, if so, refreshing forwarding table entries corresponding to all STP instances on all ports, and otherwise, refreshing the forwarding table entries corresponding to the STP instances on the ports belonging to the STP instance.

Under the combination, the proxy board records the identifier of the STP instance with the topology change in the global topology change state table in the soft restart process of the interface board, and refreshes the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance according to the identifier of each STP instance recorded in the global topology change state table after the restart of the interface board is finished, so that the inconsistency between the forwarding table entry of the STP instance and the final topology of the network can be avoided, and the technical problem of abnormal service forwarding caused by the topology change in the soft restart ISSU process of the interface board can be solved.

Combining the topology change information corresponding to the STP instance with the STP instance identifier and port information; the topology change message received by the agent board is a topology change notification TCN message.

In this case, the method for determining all STP instances having topology changes according to the topology change packet and recording topology change information corresponding to each STP instance having topology changes in the global topology change state table includes: the STP instance with the topology change is determined to be the 0 instance, and since the forwarding table entries corresponding to all STP instances need to be refreshed, the identifier of the STP instance and the port information of the topology change packet may be added to the global topology change state table for each STP instance in all STP instances.

In fact, the method for determining all STP instances having topology changes according to the topology change packet and recording topology change information corresponding to each STP instance having topology changes in the global topology change state table may further be: and determining that the STP instance with the topology change is a 0 instance, adding the identifier of the STP instance in a global topology change state table, and receiving port information of the topology change message. In this case, before the refreshing the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance, the method further includes: and judging whether the STP instance is a 0 instance, if so, refreshing forwarding table entries corresponding to all STP instances on all ports, and otherwise, refreshing the forwarding table entries corresponding to the STP instances on the ports belonging to the STP instance.

In addition, before the sending the topology change packet from the port belonging to the STP instance, the method further includes: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

In addition, in the soft restart process of the interface board, for the same STP instance, if the proxy board receives the topology change messages for multiple times and can determine that the STP instance has topology change according to the topology change messages, only the port in the topology change information corresponding to the STP instance in the global topology change state table needs to be updated. Therefore, before the recording the topology change information corresponding to each STP instance having a topology change in the global topology change state table, the method further includes: judging whether the topology change information corresponding to the STP instance exists in the global topology change state table, if so, updating the port information in the topology change information corresponding to the STP instance to be the port information for receiving the topology change message, and if not, recording the topology change information corresponding to the TP instance in the global topology change state table.

Under the combination, the proxy board records the identifier of the STP instance with the topology change in the global topology change state table and the port information of the topology change message indicating the topology change of the STP instance for the last time in the soft restart process of the interface board, and refreshes the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance according to the identifier of each STP instance recorded in the global topology change state table after the restart of the interface board is finished, so that the inconsistency between the forwarding table entry of the STP instance and the final topology of the network can be avoided, and the technical problem of abnormal service forwarding caused by the topology change in the soft restart ISSU process of the interface board can be solved. In addition, the proxy board does not send the topology change message indicating that the STP has topology change from the port according to the recorded port information of the topology change message indicating that the STP instance has topology change, thereby reducing unnecessary resource waste.

It should be noted that, in the embodiment of the present invention shown in fig. 2, after the soft restart is completed, when the proxy board sends the topology change packet from the port belonging to the STP instance according to the topology change information corresponding to each STP instance in the global topology change state table, the sent topology change packet may be a TCN packet or a BPDU packet carrying a TC tag, where the specific situation is related to the role of the port, the STP version used by the device, and the STP version used by the opposite end, which is the same as that in the prior art.

The above describes in detail a method for processing topology change in the soft restart ISSU process according to an embodiment of the present invention, and the present invention also provides a device for processing topology change in the soft restart ISSU process, which is described below with reference to fig. 3.

Fig. 3 is a schematic structural diagram of a processing apparatus for processing topology change in a soft restart ISSU process according to an embodiment of the present invention, and as shown in fig. 3, the apparatus includes an interface board and a proxy board, where the proxy board includes: a message receiving unit 301, a message processing unit 302, a table entry refreshing unit 303 and a message sending unit 304; wherein,

a message receiving unit 301, configured to receive a topology change message in a soft restart process of the interface board;

a message processing unit 302, configured to determine all STP instances in which topology changes occur according to the topology change message when the message receiving unit 301 receives the topology change message, and record topology change information corresponding to each STP instance in which topology changes occur in the global topology change state table;

an entry refreshing unit 303, configured to refresh, after the soft restart of the interface board is completed, a forwarding entry corresponding to each STP instance on a port belonging to the STP instance according to topology change information corresponding to each STP instance in the global topology change state table;

a message sending unit 304, configured to send a topology change message from a port belonging to each STP instance in the global topology change state table according to topology change information corresponding to the STP instance after the soft restart of the interface board is completed.

The interface board includes a message upload unit 305;

the message processing unit 302 is further configured to empty a global topology change state table before the interface board is soft restarted;

the message uploading unit 305 is configured to set a message interception table entry for uploading a message received by a port on the interface board to the proxy board for processing when the soft restart of the interface board starts, where the message interception table entry includes message characteristic information and a proxy board number, and the message characteristic information includes a destination MAC of the message; when a port on an interface board receives a message, if the message conforms to the message feature information in the message interception table entry, the message receiving unit 301 sends the message to the proxy board according to the proxy board number in the message interception table entry.

Here, the message uploading unit 305 uploads the message received at the port of the interface board to the proxy board according to the set message interception table entry, so that the proxy board can perform message processing instead of the interface board in the soft restart process of the interface board.

In the above-mentioned device, the liquid crystal display device,

the topology change information corresponding to the STP instance comprises the STP instance identifier;

the topology change message received by the message receiving unit 301 is a BPDU message carrying a TC tag;

when determining all STP instances having topology changes according to the topology change packet and recording topology change information corresponding to each STP instance having topology changes in the global topology change state table, the packet processing unit 302 is configured to: and determining the STP instance to which the TC mark carried in the topology change message belongs, and adding the STP instance identifier in a global topology change state table aiming at the STP instance to which each TC mark belongs.

In the above-mentioned device, the liquid crystal display device,

the topology change information corresponding to the STP instance comprises the STP instance identifier and port information;

the topology change message received by the message receiving unit 301 is a BPDU message carrying a TC tag;

when the packet processing unit 302 records topology change information corresponding to the STP instance to which each TC tag belongs, carried in the topology change packet, in the global topology change state table, the packet processing unit is configured to: determining the STP instance to which the TC mark carried in the topology change message belongs, adding the STP instance identifier in a global topology change state table aiming at the STP instance to which each TC mark belongs, and receiving port information of the topology change message;

the message sending unit 304 is further configured to, before sending the topology change message from the port belonging to the STP instance: judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, otherwise, sending the topology change message from the port belonging to the STP instance

In the above-mentioned device, the liquid crystal display device,

the topology change information corresponding to the STP instance comprises the STP instance identifier;

the topology change message received by the message receiving unit 301 is a topology change notification TCN message;

when determining all STP instances having topology changes according to the topology change packet and recording topology change information corresponding to each STP instance having topology changes in the global topology change state table, the packet processing unit 302 is configured to: and determining that all the STP instances have topology changes, and adding the identifier of the STP instance in the global topology change state table aiming at each STP instance.

In the above-mentioned device, the liquid crystal display device,

the topology change information corresponding to the STP instance comprises the STP instance identifier and port information;

the topology change message received by the message receiving unit 301 is a topology change notification TCN message;

when determining all STP instances having topology changes according to the topology change packet and recording topology change information corresponding to each STP instance having topology changes in the global topology change state table, the packet processing unit 302 is configured to: determining that all STP instances have topology change, adding the identifier of the STP instance in a global topology change state table for each STP instance, and receiving port information of the topology change message;

the message sending unit 304 is further configured to, before sending the topology change message from the port belonging to the STP instance: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

In the above-mentioned device, the liquid crystal display device,

the topology change information corresponding to the STP instance comprises the STP instance identifier;

the topology change message received by the message receiving unit 301 is a topology change notification TCN message;

when determining all STP instances having topology changes according to the topology change packet and recording topology change information corresponding to each STP instance having topology changes in the global topology change state table, the packet processing unit 302 is configured to: determining that the STP instance with the topology change is a 0 instance, and adding the identifier of the STP instance in a global topology change state table;

before the table entry refreshing unit 303 refreshes the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance, the table entry refreshing unit is further configured to: and judging whether the STP instance is a 0 instance, if so, refreshing forwarding table entries corresponding to all STP instances on all ports, and otherwise, refreshing the forwarding table entries corresponding to the STP instances on the ports belonging to the STP instance.

In the above-mentioned device, the liquid crystal display device,

the topology change information corresponding to the STP instance comprises the STP instance identifier and port information;

the topology change message received by the message receiving unit 301 is a topology change notification TCN message;

when determining all STP instances having topology changes according to the topology change packet and recording topology change information corresponding to each STP instance having topology changes in the global topology change state table, the packet processing unit 302 is configured to: determining that the STP instance with the topology change is a 0 instance, adding the identifier of the STP instance in a global topology change state table, and receiving port information of the topology change message;

before the table entry refreshing unit 303 refreshes the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance, the table entry refreshing unit is further configured to: judging whether the STP instance is a 0 instance, if so, refreshing forwarding table entries corresponding to all STP instances on all ports, otherwise, refreshing the forwarding table entries corresponding to the STP instances on the ports belonging to the STP instance;

before the packet sending unit sends the topology change packet from the port belonging to the STP instance, the packet sending unit is further configured to: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

In the above-mentioned device, the liquid crystal display device,

before the packet processing unit 302 records the topology change information corresponding to each STP instance having topology change in the global topology change state table, the packet processing unit is further configured to: judging whether the topology change information corresponding to the STP instance exists in the global topology change state table, if so, updating the port information in the topology change information corresponding to the STP instance to the port information of the topology change message carrying the STP instance information, and if not, recording the topology change information corresponding to the STP instance in the global topology change state table.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (15)

1. A method for processing topology change in the process of upgrading ISSU by soft restart uninterrupted service (ISSU), which is characterized in that the method comprises the following steps:

in the soft restarting process of an interface board, when an agent board receives a topology change message, determining all Spanning Tree Protocol (STP) instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in a global topology change state table;

after the soft restart of the interface board is finished, the proxy board refreshes the forwarding table items corresponding to the STP instance on the port belonging to the STP instance according to the topology change information corresponding to each STP instance in the global topology change state table, and sends the topology change messages from the port belonging to the STP instance.

2. The method for handling topology change during soft restart ISSU according to claim 1, further comprising:

before the interface board is in soft restart, emptying a global topology change state table;

when the soft restart of the interface board is started, message interception table entries which are used for sending messages received by ports on the interface board to the proxy board for processing are set, wherein the message interception table entries comprise message characteristic information and proxy board numbers, and the message characteristic information comprises destination Media Access Control (MAC) addresses of the messages; when the port on the interface board receives the message, if the message accords with the message characteristic information in the message interception table item, the message is sent to the agent board according to the agent board number in the message interception table item.

3. The method for handling topology changes during soft restart ISSU according to claim 1,

the topology change information corresponding to the STP instance comprises the STP instance identifier, or the STP instance identifier and port information;

the topology change message received by the proxy board is a Bridge Protocol Data Unit (BPDU) message carrying a topology change TC mark;

the method for determining all the STP instances with topology changes according to the topology change messages and recording the topology change information corresponding to each STP instance with topology changes in the global topology change state table comprises the following steps: determining the STP instance to which the TC mark carried in the topology change message belongs, and adding the identifier of the STP instance or the identifier of the STP instance and port information for receiving the topology change message in a global topology change state table aiming at the STP instance to which each TC mark belongs;

when the topology change information corresponding to the STP instance includes the STP instance identifier and the port information, before the sending the topology change packet from the port belonging to the STP instance, the method further includes: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

4. The method for handling topology changes during soft restart ISSU according to claim 1,

the topology change information corresponding to the STP instance comprises the STP instance identifier, or the STP instance identifier and port information;

the topology change message received by the agent board is a topology change notification TCN message;

the method for determining all the STP instances with topology changes according to the topology change messages and recording the topology change information corresponding to each STP instance with topology changes in the global topology change state table comprises the following steps: determining that the STP instance with the topology change is 0 instance, adding the identifiers of all STP instances in a global topology change state table, or the identifier of each STP instance in all STP instances and port information for receiving the topology change message;

when the topology change information corresponding to the STP instance includes the STP instance identifier and the port information, before the sending the topology change packet from the port belonging to the STP instance, the method further includes: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

5. The method for handling topology changes during soft restart ISSU according to claim 1,

the topology change information corresponding to the STP instance comprises the STP instance identifier, or the STP instance identifier and port information;

the topology change message received by the agent board is a topology change notification TCN message;

the method for determining all the STP instances with topology changes according to the topology change messages and recording the topology change information corresponding to each STP instance with topology changes in the global topology change state table comprises the following steps: determining that the STP instance with the topology change is a 0 instance, and adding the identifier of the STP instance or the identifier of the STP instance and port information for receiving the topology change message in a global topology change state table;

before the refreshing the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance, the method further includes: judging whether the STP instance is a 0 instance, if so, refreshing forwarding table entries corresponding to all STP instances on all ports, otherwise, refreshing the forwarding table entries corresponding to the STP instances on the ports belonging to the STP instance;

when the topology change information corresponding to the STP instance includes the STP instance identifier and the port information, before the sending the topology change packet from the port belonging to the STP instance, the method further includes: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

6. The method for processing topology change during soft restart ISSU according to claim 3, 4 or 5,

when the topology change information corresponding to the STP instance includes the STP instance identifier and the port information, before the recording the topology change information corresponding to each STP instance having topology change in the global topology change state table, the method further includes: judging whether the topology change information corresponding to the STP instance exists in the global topology change state table, if so, updating the port information in the topology change information corresponding to the STP instance to the port information of the topology change message carrying the STP instance information, and if not, recording the topology change information corresponding to the STP instance in the global topology change state table.

7. A processing device for topology change in the process of upgrading ISSU by soft restart uninterrupted service is characterized in that the device comprises an interface board and a proxy board, and the proxy board comprises: the device comprises a message receiving unit, a message processing unit, a table item refreshing unit and a message sending unit;

the message receiving unit is used for receiving a topology change message in the soft restart process of the interface board;

the message processing unit is used for determining all STP instances with topology change according to the topology change message when the message receiving unit receives the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table;

the table entry refreshing unit is used for refreshing the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance according to the topology change information corresponding to each STP instance in the global topology change state table after the soft restart of the interface board is finished;

and the message sending unit is used for sending the topology change message from the port belonging to the STP instance according to the topology change information corresponding to each STP instance in the global topology change state table after the soft restart of the interface board is finished.

8. The apparatus for processing topology change during soft restart ISSU according to claim 7, wherein the interface board comprises a message forwarding unit;

the message processing unit is further configured to empty the global topology change state table before the interface board is soft restarted;

the message uploading unit is used for setting a message interception table entry for uploading a message received by a port on the interface board to the proxy board for processing when the soft restart of the interface board is started, wherein the message interception table entry comprises message characteristic information and a proxy board number, and the message characteristic information comprises a destination MAC address of the message; and the message receiving unit is used for sending the message to the proxy board according to the proxy board number in the message interception table item if the message accords with the message characteristic information in the message interception table item when the port on the interface board receives the message.

9. The apparatus for processing topology change during soft restart ISSU according to claim 7,

the topology change information corresponding to the STP instance comprises the STP instance identifier;

the topology change message received by the message receiving unit is a BPDU message carrying a TC mark;

the message processing unit is used for determining all STP instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table, and is used for: and determining the STP instance to which the TC mark carried in the topology change message belongs, and adding the STP instance identifier in a global topology change state table aiming at the STP instance to which each TC mark belongs.

10. The apparatus for processing topology change during soft restart ISSU according to claim 7,

the topology change information corresponding to the STP instance comprises the STP instance identifier and port information;

the topology change message received by the message receiving unit is a BPDU message carrying a TC mark;

the message processing unit is used for determining all STP instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table, and is used for: determining the STP instance to which the TC mark carried in the topology change message belongs, adding the STP instance identifier in a global topology change state table aiming at the STP instance to which each TC mark belongs, and receiving port information of the topology change message;

before the packet sending unit sends the topology change packet from the port belonging to the STP instance, the packet sending unit is further configured to: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

11. The apparatus for processing topology change during soft restart ISSU according to claim 7,

the topology change information corresponding to the STP instance comprises the STP instance identifier;

the topology change message received by the message receiving unit is a topology change notification TCN message;

the message processing unit is used for determining all STP instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table, and is used for: and determining that the STP instance with the topology change is a 0 instance, and adding the identifications of all the STP instances in the global topology change state table.

12. The apparatus for processing topology change during soft restart ISSU according to claim 7,

the topology change information corresponding to the STP instance comprises the STP instance identifier and port information;

the topology change message received by the message receiving unit is a topology change notification TCN message;

the message processing unit is used for determining all STP instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table, and is used for: determining that the STP instance with the topology change is a 0 instance, and adding the identifier of the STP instance and receiving the port information of the topology change message in a global topology change state table aiming at each STP instance in all the STP instances;

before the packet sending unit sends the topology change packet from the port belonging to the STP instance, the packet sending unit is further configured to: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

13. The apparatus for processing topology change during soft restart ISSU according to claim 7,

the topology change information corresponding to the STP instance comprises the STP instance identifier;

the topology change message received by the message receiving unit is a topology change notification TCN message;

the message processing unit is used for determining all STP instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table, and is used for: determining that the STP instance with the topology change is a 0 instance, and adding the identifier of the STP instance in a global topology change state table;

before the table entry refreshing unit refreshes the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance, the table entry refreshing unit is further configured to: and judging whether the STP instance is a 0 instance, if so, refreshing forwarding table entries corresponding to all STP instances on all ports, and otherwise, refreshing the forwarding table entries corresponding to the STP instances on the ports belonging to the STP instance.

14. The apparatus for processing topology change during soft restart ISSU according to claim 7,

the topology change information corresponding to the STP instance comprises the STP instance identifier and port information;

the topology change message received by the message receiving unit is a topology change notification TCN message;

the message processing unit is used for determining all STP instances with topology change according to the topology change message, and recording topology change information corresponding to each STP instance with topology change in the global topology change state table, and is used for: determining that the STP instance with the topology change is a 0 instance, adding the identifier of the STP instance in a global topology change state table, and receiving port information of the topology change message;

before the table entry refreshing unit refreshes the forwarding table entry corresponding to the STP instance on the port belonging to the STP instance, the table entry refreshing unit is further configured to: judging whether the STP instance is a 0 instance, if so, refreshing forwarding table entries corresponding to all STP instances on all ports, otherwise, refreshing the forwarding table entries corresponding to the STP instances on the ports belonging to the STP instance;

before the packet sending unit sends the topology change packet from the port belonging to the STP instance, the packet sending unit is further configured to: and judging whether the port is the same as the port in the topology change information corresponding to the STP instance in the global topology change state table, if so, not sending the topology change message from the port belonging to the STP instance, and otherwise, sending the topology change message from the port belonging to the STP instance.

15. The apparatus for processing topology change during soft restart ISSU according to claim 10, 12 or 14,

before the packet processing unit records topology change information corresponding to each STP instance having topology change in the global topology change state table, the packet processing unit is further configured to: judging whether the topology change information corresponding to the STP instance exists in the global topology change state table, if so, updating the port information in the topology change information corresponding to the STP instance to the port information of the topology change message carrying the STP instance information, and if not, recording the topology change information corresponding to the STP instance in the global topology change state table.