Linux server cluster unified peripheral event monitoring method based on netlink

Abstract

The invention discloses a Linux server cluster unified peripheral event monitoring method based on netlink; the method comprises the steps that: after a monitor request from a monitoring server end is received by the service process of a monitored server end, subprocess verification login account information is created to permit a subprocess to open a relevant peripheral event handling process to monitor a specific netlink multicast group and wait for peripheral events sent by a kernel; a peripheral event kernel receiving module receives the peripheral events sent by a device drive, and sends the same to the user space netlink multicast group according to sequence and priority of the events; a peripheral event handling process transmits the received peripheral events to the service process of a monitored server through the subprocess and then transmitted to the monitoring server end through the network, and thereby the purpose of unified monitoring of the peripheral events is finally achieved. The Linux server cluster unified peripheral event monitoring method based on netlink has the advantages that multiple terminals can be simultaneously monitored, peripheral events in the server can be received in a unified and real-time manner, and the programming method is simple and easy to implement.

Description

technical field [0001] The invention relates to the field of computers, in particular to a netlink-based unified peripheral event monitoring method for Linux server clusters. Background technique [0002] In a linux server cluster, there are usually multiple server terminals, and in a large server architecture, many server terminals are distributed in different geographical locations. Peripheral asynchronous events such as events and unapproved external storage device partition mount events lack an effective and real-time remote and unified monitoring method. [0003] At present, there is no unified peripheral event monitoring method within the Linux server cluster in the prior art. [0004] Moreover, for the monitoring of peripheral events inside a single server, it mainly relies on the device file system daemon udevd in user space to collect peripheral events to dynamically change the device node information under the device file system directory / dev, and there is no oth...

AI Technical Summary

Problems solved by technology

[0003] At present, there is no unified peripheral event monitoring method in the linux server cluster in the prior art

[0004] Moreover, for the monitoring of peripheral events inside a single server, it mainly relies on the device file system daemon udevd in user space to collect peripheral events to dynamically change the device node information under the device file system directory / dev, and there is no other It is a separate and independent peripheral event user space receiving program, and the device file system daemon udevd is poorly customizable, and it is difficult to implement a method of uniformly monitoring all important peripheral events in the server cluster by changing the existing udevd daemon

Images