Cluster log centralized collection method applied to platform development

Abstract

The invention discloses a cluster log centralized collection method applied to platform development. The method involves an application system, a monitor, a log collector, a configuration module interface, an interface configuration module and a log storage module, wherein the application system, the monitor and the log collector are in electric communication connection sequentially; the log collector is in electric communication connection with the interface configuration module and the log storage module; and the interface configuration module is in electric communication connection with the configuration module interface. According to the method, log collection can be performed for a specified interface, and centralized collection is realized, so that developers and operation and maintenance personnel can track and analyze logs in real time more conveniently; when not needed, the log collection is cancelled by using configuration, so that server resources can be saved to the maximum extent; and the use is more convenient.

Description

technical field [0001] The invention relates to the field of platform development and operation, in particular to a centralized collection method for cluster logs applied in platform development. Background technique [0002] With more and more smart TV users, users who use personalized recommendation systems experience a blowout growth, and the recommendation platform is always accepting challenges. At this time, it is necessary to pay attention to the operation status of the recommendation platform in real time. Log, by analyzing the log to determine whether the platform is normal, whether it can withstand the pressure, whether the recommendation effect is qualified, etc., which is convenient for platform maintenance. [0003] There are generally two types of existing log collectors: (1) Embed log collection code in the code to print the log of a single server. This kind of log is scattered on each server in the cluster. It is difficult to track and analyze the log. It is...

AI Technical Summary

Problems solved by technology

[0003] There are generally two types of existing log collectors: (1) Embed log collection code in the code to print the log of a single server. This kind of log is scattered on each server in the cluster. It is difficult to track and analyze the log. It is not conducive to real-time tracking and analysis, and in the case of no need to print logs, it is necessary to modify the application configuration file and restart the application, causing instability to the business; (2) Use a unified log collector, although this kind of log can be collected uniformly Go to one place, store in the database or message queue or log file, but collect a large number of logs at the same time. This kind of collector is used in the big data analysis platform, but it is used in some small-scale or non-data-oriented products, which will consume server resources. Wasteful, it cannot collect logs in a targeted manner, and it is difficult to track and analyze logs in real time

Images