Shulou(Shulou.com)06/03 Report--

In the previous article, it has been clearly stated that DKM as the management platform of the fast release DKhadoop, its four major functions are: management function, monitoring function, diagnosis function and integration function. The management function has listed some instructions for you. Today, we will share and analyze the monitoring function of the DKM platform in detail.

DKM provides many monitoring functions to detect the health of the cluster, the performance of components, and the performance and resource requirements of jobs running in the cluster. Specifically, it includes the following monitoring functions:

1. Service monitoring:

Viewing the results of service and role instance-level health checks and displaying them on a chart can help you diagnose the problem. If a health check finds that the status of a component requires special attention or even has a problem, the system will make recommendations on the actions that the administrator should take. At the same time, the system administrator can view the execution history of operations on the service or role, as well as the audit log of configuration changes.

2. Host monitoring

Monitor information about all hosts in the cluster, including which hosts are online or offline, the memory currently consumed on the host, the role instance allocation running on the host, the host allocation on different racks, and so on. The summary view shows all hosts in the cluster, and you can further view the rich details of a single host, including a visual chart showing the key metrics of the host.

3. Behavior monitoring:

DKM provides lists and diagrams to view the activities taking place on the cluster, to show the behavior of tasks currently being executed, and to view historical activities. At the same time, it provides a lot of statistics of the resources used by each job, and the system administrator can diagnose performance problems or behavior problems by comparing the different performance data of similar tasks and comparing the performance data of different execution in the same task.

4. Call the police:

You can generate alerts for specified events by configuring DKM.

5. Log:

Describes how to access various ways in the log that take into account the way you are viewing the current context. For example, when monitoring a service, you can easily click on a link to view log entries for related specific services through the same user interface. When viewing information about the activity of the user, you can easily view the relevant log entries that occur on the host of the job under which the job is running.

6. Real-time monitoring of clusters and services:

DKM monitors the services of the cluster in real time. If an exception occurs in the service, an alarm will be generated. At the same time, monitoring the performance data of the cluster, such as CPU, memory, network and so on, on the one hand, can monitor the performance status of the cluster, on the other hand, it can help to tune the performance.

7. HDFS service monitoring:

Take HDFS service as an example, the HDFS service page displays summary information of service status, including HDFS summary, total number of HDFS file system blocks, total size, total number of files, and so on.

8. Node-level monitoring

DKM can view the service status of a single node

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.