Process monitoring method and system in a virtualized environment

Abstract

The invention discloses a process monitoring method and system in a virtualization environment. The process monitoring method comprises the steps of (1) obtaining the process descriptor of a process to be monitored, (2) detecting whether the signal queue of the process to be monitored in the process descriptor includes signals which cause the process to withdraw abnormally before an operation system processes the signal queue of the process to be monitored, and (3) collecting the current operating information of the process to be monitored, generating corresponding warning information and feeding the warning information back to a terminal if the signal queue of the process to be monitored in the process descriptor includes the signals which cause the process to withdraw abnormally. With the process monitoring method, the processes in any namespace can be monitored, the process to be monitored does not need to be changed, and the warning information is guaranteed to be fed back timely and accurately.

Description

technical field [0001] The invention relates to the technical field of computer applications, in particular to a process monitoring method and system in a virtualized environment. Background technique [0002] With the development and maturity of virtualization technology, more and more enterprises choose to introduce virtualization technology on the server to maximize the use of server hardware resources, reduce hardware procurement costs, and reduce energy consumption. As a lightweight virtualization technology, container virtualization has the characteristics of simple deployment and low virtualization overhead, so it is increasingly used in server virtualization. By introducing container virtualization technology on the server, we can deploy multiple containers on a physical machine to run multiple instances of an application, and can also run multiple different applications in different containers to ensure that they are not mutually exclusive. influences. Therefore, ...

AI Technical Summary

Problems solved by technology

Therefore, higher requirements are put forward for the operation and maintenance of the server. When a process in a certain container exits abnormally, the traditional detection method needs to make adaptive modifications to the monitored process, so that the monitored process periodically reports to the The monitoring system actively reports the heartbeat information, or the monitoring system queries whether there is a corresponding process ID directory under the / proc file system by periodically querying the information of the process to be monitored. Both monitoring methods require the process to be monitored and the monitoring system Binding together requires changes to the monitoring process. At the same time, neither of these two methods can monitor the abnormal exit of the process in a timely and accurate manner, so the corresponding alarm information cannot be issued in a timely and accurate manner; corresponding measures cannot be taken to Corresponding exception recovery processing cannot guarantee the high reliability and self-recovery of products and systems

[0003] In addition, the currently commonly used detection methods are only applicable to Linux operating systems that have not introduced namespaces. After introducing namespaces in Linux operating systems, the monitoring process and the monitored process may run in different namespaces. The monitoring process requires The processing scenarios are more complex, and the traditional process monitoring method does not involve the processing in this scenario, and the process monitoring across namespaces cannot be completed

Images