Consensus process recovery method and related node

1. A consensus method implemented by a first node, comprising:

sending a service request to M backup nodes, wherein the first node is a primary node running an instance, wherein the M backup nodes are backup nodes running the instance, and wherein M is a positive integer;

receiving N pieces of first authentication information from N backup nodes, wherein the N pieces of the first authentication information comprise N signatures of the N backup nodes generating the N pieces of the first authentication information, wherein the N pieces of the first authentication information indicate that the N backup nodes successfully authenticated the service request, and wherein N is a positive integer less than or equal to M;

identifying that N exceeds a first preset threshold;

performing the service request in response to identifying that N exceeds the first preset threshold; and

sending first acknowledgement information to the M backup nodes, wherein the first acknowledgement information comprises the N signatures, and wherein the first acknowledgement information enables the M backup nodes to perform the service request.

2. The consensus method of claim 1 , wherein a node cluster comprises the first node and the M backup nodes, wherein the first node and the M backup nodes jointly run the instance, wherein a plurality of instances comprises a primary instance and a secondary instance, wherein the first node is the primary node running a target instance of the instances, wherein the M backup nodes are the backup nodes running the target instance, wherein the consensus method further comprises obtaining, using the M backup nodes, a consensus to replace the primary instance with the target instance as a new primary instance, and wherein a throughput of the target instance is higher than a throughput of the primary instance.

3. The consensus method of claim 2 , wherein

L

≥

⌊

K

-

1

3

⌋

+

1

,

wherein L represents a quantity of the instances, wherein K represents a quantity of nodes in the node cluster, wherein K=M+1, and wherein K is an integer greater than or equal to three.

4. The consensus method of claim 1 , wherein a backup node running the instance is an alternate primary node of the instance, and wherein the alternate primary node is configured to replace the primary node running the instance to become a new primary node running the instance.

5. The consensus method of claim 4 , wherein alternate primary nodes running two different instances are different.

6. The consensus method of claim 4 , wherein alternate primary nodes running the instance comprise different priorities, and wherein an alternate primary node comprising a highest priority is configured to replace the primary node to become the new primary node.

7. A consensus method implemented by a second node, comprising:

receiving a service request from a first node, wherein the first node is a primary node running an instance, wherein the second node is one of M backup nodes running the instance, and wherein M is a positive integer;

authenticating the service request to generate first authentication information, wherein the first authentication information comprises a signature of the second node, and wherein the first authentication information indicates that the second node successfully authenticated the service request;

sending the first authentication information to the first node;

receiving a first acknowledgement message from the first node, wherein the first acknowledgement message comprises respective signatures of N backup nodes, wherein the first acknowledgement message indicates that a quantity N of pieces of first authentication information received by the first node exceeds a first preset threshold, and wherein N is a positive integer less than or equal to M; and

performing the service request.

8. The consensus method of claim 7 , wherein a node cluster comprises the first node and the M backup nodes, wherein the first node and the M backup nodes jointly run the instance, wherein a plurality of instances comprises a primary instance and a secondary instance, wherein the first node is the primary node running a target instance of the instances, wherein the second node is one of the M backup nodes running the target instance, wherein the consensus method further comprises obtaining, using the first node, a consensus to replace the primary instance with the target instance as a new primary instance, and wherein a throughput of the target instance is higher than a throughput of the primary instance.

9. The consensus method of claim 8 , wherein

L

≥

⌊

K

-

1

3

⌋

+

1

,

wherein L represents a quantity of the instances, wherein K represents a quantity of nodes in the node cluster, wherein K=M+1, and wherein K is an integer greater than or equal to three.

10. The consensus method of claim 7 , wherein a backup node running the instance is an alternate primary node of the instance, and wherein the alternate primary node is configured to replace the primary node running the instance to become a new primary node running the instance.

11. The consensus method of claim 10 , wherein alternate primary nodes running two different instances are different.

12. The consensus method of claim 10 , wherein alternate primary nodes running the instance comprises different priorities, and wherein an alternate primary node comprising a highest priority is configured to replace the primary node to become the new primary node.

13. A cluster system, comprising:

M backup nodes configured to run an instance, wherein M is a positive integer; and

a first node coupled to the M backup nodes, wherein the first node is a primary node running the instance, and wherein the first node is configured to:

send a service request to the M backup nodes;

receive N pieces of first authentication information from N backup nodes, wherein the N pieces of the first authentication information comprise N signatures of the N backup nodes generating the N pieces of the first authentication information, wherein the N pieces of the first authentication information indicate that the N backup nodes generating the N pieces of the first authentication information successfully authenticated the service request, and wherein N is a positive integer less than or equal to M;

identifying that N exceeds a first preset threshold;

perform the service request in response to identifying that N exceeds the first preset threshold; and

send first acknowledgement information to the M backup nodes, wherein the first acknowledgement information comprises the N signatures, and wherein the first acknowledgement information enables the M backup nodes to perform the service request, and

wherein the M backup nodes are configured to:

receive the service request from the first node;

authenticate the service request to generate the first authentication information;

send the first authentication information to the first node;

receive the first acknowledgement information from the first node; and

perform the service request based on the service request.

14. The cluster system of claim 13 , wherein the first node and the M backup nodes jointly run the instance, wherein a plurality of instances comprise a primary instance and a secondary instance, wherein the first node is the primary node running a target instance of the instances, wherein a second node is one of the M backup nodes running the target instance, wherein the first node is further configured to obtain a consensus with the M backup nodes to replace the primary instance with the target instance as a new primary instance, and wherein a throughput of the target instance is higher than a throughput of the primary instance.