Distributed consistency transaction execution method based on retry mechanism

Abstract

The invention discloses a distributed consistency transaction execution method based on a retry mechanism, which comprises the following steps: S1, when an exception occurs, reporting retry data at an exception point; writing the received retry data into a retry queue; automatically scanning the retry data needing to be retry in the retry queue to perform automatic retry execution; and S2, when the automatic retry execution is successful, retry data dequeue is carried out, and when the automatic retry execution is failed, the next retry time is updated. According to the invention, a relatively intelligent retry strategy is realized, so that the self-recovery capability of the service system is supplemented, and the stability of the service system can be improved; a visual operation interface is provided for abnormal data needing to be manually confirmed, and the problem processing cost is reduced; fine monitoring and alarming are carried out at a retry point; and the place where the abnormity occurs can be basically corresponded through re-commissioning, so that the troubleshooting efficiency can be improved.

Description

technical field [0001] The invention relates to the field of computer software, in particular to a distributed consistent transaction execution method based on a retry mechanism. Background technique [0002] In a distributed environment, most of our systems do not exist alone, and interdependence between systems is the norm. [0003] From a macro perspective, distributed "dependence" not only increases the overall complexity of the system, but also increases the probability of system problems. For example: when placing an order, the call to the payment system is successful, but the final addition of goods times out; when a serious system problem occurs in the downstream service that the business system depends on, the business is affected for a period of time. System problems will eventually be reflected in data consistency. [0004] There are two main technical solutions currently used: the first one is to call other system services first and then call the local service ...

AI Technical Summary

Problems solved by technology

[0005] The main disadvantages of the first solution are: 1. It is only applicable to the situation where there is only one other system service call, and it is not enough to call multiple other system services; 2. It is not applicable when the service call times out

[0006] The main disadvantages of the second solution are: 1. Strong intrusion to the application

Each branch of the business logic needs to implement three operations: try, confirm, and cancel. The application is highly intrusive and the transformation cost is high.

2. It is more difficult to realize

Images