A container log collection method and device

Abstract

The embodiment of the invention discloses a container log collection method and device, which are applied to computational nodes in an elastic calculating system. Corresponding storage paths of container log files corresponding to application programs are previously stored in a preset container log catalog. The method comprises the steps of searching for log files matched with user collection instructions according to the preset container log catalogs and the user collection instructions every a preset time period or every preset time point, wherein the user collection instructions include one or several combinations of identification information of application programs corresponding to container logs to be collected, identification information of containers corresponding to the container logs to be collected and identification information of log files corresponding to container logs to be collected; collecting logs recorded in the log files. By means of the technical scheme provided by the embodiment of the invention, log collection efficiency can be improved.

Description

technical field [0001] The invention relates to the technical field of the Internet, in particular to a container log collection method and device. Background technique [0002] Elastic computing can create and destroy containers at any time according to the business needs of the application, without the need to purchase investment in advance. It is a cloud computing service that is simpler and more efficient than physical server or virtual server management. [0003] An elastic computing system usually includes multiple elastic computing nodes, and each container is assigned to an elastic computing node, where an elastic computing node represents a physical server. In addition, in an elastic computing system, a container can only correspond to one application, and an application can run in multiple containers, and these containers may be allocated on the same elastic computing node, or may be allocated on different elastic computing nodes Specifically, which node each cont...

AI Technical Summary

Problems solved by technology

Assume, through manual determination, that application A is running on containers 1, 2, and 3 of elastic computing nodes a, b, and c respectively, then it is necessary to obtain the paths d1, d2 of the log files corresponding to containers 1, 2, and 3 respectively , d3, and then configure the collection paths d1, d2, and d3 on the log collection end to collect logs; however, if the container of application A changes before the next collection, container 1 on elastic computing node a is destroyed, and the original container The program running in 1 is running in the new container 4 on the elastic computing node b. Then, if only the log files corresponding to d1, d2, and d3 are still collected, the log collection of application A may be incomplete, because d1 The logs that need to be collected in the log files corresponding to , d2, and d3 may only be part of the container before the change. In order to ensure the integrity of the log collection, at this time, it is necessary to apply the above-mentioned method again to re-determine all the containers of application A. Log path, collect the logs in the log file corresponding to the new log path

Images