Help Center/ MapReduce Service/ Component Operation Guide (LTS) (Ankara Region)/ Using Elasticsearch/ Common Issues About Elasticsearch/ What Can I Do If the Speed of Writing Data into Elasticsearch Is Slow Due to Fault on an Elasticsearch Node?
Updated on 2024-11-29 GMT+08:00
View PDF
Share

What Can I Do If the Speed of Writing Data into Elasticsearch Is Slow Due to Fault on an Elasticsearch Node?

Symptom

In a large cluster at a site, 246 EsNode instances are deployed. If a node is faulty, the overall data writing speed of the Elasticsearch cluster becomes slow.

Cause Analysis

Based on the analysis of specific services and onsite fault locating, it is found that the faults occur on one of the nodes in the cluster. The number of index shards is set to 246, and the Elasticsearch shards are evenly distributed to each EsNode instance. When data is written in batches, the writing speed of data on one node is slow. As a result, the overall data writing speed is slow.

Solution

Set a proper number of shards. When a fault occurs on a node, migrate the hot shards on the node to other nodes.