Shulou(Shulou.com)05/31 Report--

How to analyze the principle of hadoop2.x namenode ha architecture, I believe that many inexperienced people do not know what to do about it. Therefore, this paper summarizes the causes and solutions of the problem. Through this article, I hope you can solve this problem.

In the architecture diagram of version 2.x of ha

You can see that in order to seamlessly switch between the two NN, the DN needs to send the Block message to both NN at the same time, and what the NN needs to do is to synchronize the metadata information.

This involves the problem of synchronization, shared storage.

In terms of reliability, DFSZKFailoverController is launched on each NN to monitor the running status of the local NN in order to make a corresponding judgment and processing.

And zookeeper is to achieve a distributed lock, to prevent the emergence of two active NN at the same time, as well as some data sharing.

We can look at this basic architecture diagram of comparative principles.

You can see that on each NN, there are HADeamon processes, timing and DFSZKFailoverController HealthMonitor for heartbeat detection.

DFSZKFailoverController contains ActiveStandbyElector to elect active and standby.

Metadata information is updated in real time through QJM.

We can regard standby NN as a NN under safemode, which does not provide external services, but can update a series of operations, such as metadata information modification, block data reporting and updating, etc.

After reading the above, have you mastered the method of how to analyze the principle of hadoop2.x namenode ha architecture? If you want to learn more skills or want to know more about it, you are welcome to follow the industry information channel, thank you for reading!

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.