Heartbeat basic principle and logical architecture of Linux high-availability cluster scheme

In this article we mainly learn the basic principles and logical architecture of heartbeat high-availability clusters, as well as the simple configuration of Heartbeat

650) this.width=650; "src=" Http://www.heartbeat.com/wp-content/uploads/2016/05/heartbeat-logo-300x300.png "alt=" Heartbeat "/>

ll This article navigation

· The basic principle of heartbeat

· Heartbeat Cluster Components

· Heartbeat Heartbeat Connection

· Heartbeat's brain fissure (resource contention, resource isolation)

· Heartbeat 's configuration file

· Heartbeat to highly available cluster configuration

ll requirements

Mastering the relevant components and simple configuration of heartbeat high availability cluster

the basic principle of heartbeat  

Heartbeat is a cluster software, it mainly consists of heartbeat information detection and resource management of two core components.

In a cluster built by heartbeat, each server sends heartbeat information (messages) to other cluster nodes and collects and analyzes them to determine the status of the node, and thus considers the node to be valid. When the server does not detect the heartbeat information of other nodes in the specified time or is unable to connect through the network, the other node is considered invalid, at this point, the server needs to start the Resource takeover module to take over the services and resources on the failed node.

Heartbeat can only complete heartbeat information detection and resource monitoring, and will not monitor other resources and application services. To monitor whether other resources and app services are available, you need to install third-party plug-ins, such as Ipfail, Ldirectord, and so on.

Similarly, the problem with the operating system itself heartbeat also cannot be monitored. If the primary node is unable to send the heartbeat information to the standby node due to operating system problems, the standby node cannot receive information from the master node, thus assuming that the primary node is invalid, and the resource Takeover module is initiated to take over the services and resources of the master node. While the primary node resources and services are not released, resource contention occurs at the primary and standby nodes, which can cause data corruption of the shared resource or crash of the file system in severe cases. For Linux systems, to solve this problem, you need to open the watchdog module in the kernel, after the module is turned on, watchdog will periodically write to the/dev/watchdog device files to determine whether the system is functioning properly. If watchdog thinks the kernel hangs, it will automatically restart the system, freeing up the node's services and resources.

Heartbeat cluster Components

650) this.width=650; "Src=" https://s2.51cto.com/wyfs02/M00/93/4E/wKioL1kJ9UThKh33AACbPOgq4BY134.jpg-wh_500x0-wm_ 3-wmp_4-s_1243312774.jpg "title=" heartbeat-cluster component. jpg "alt=" wkiol1kj9uthkh33aacbpogq4by134.jpg-wh_50 "/>

This article from "Ccschan" blog, reproduced please contact the author!

Heartbeat basic principle and logical architecture of Linux high-availability cluster scheme