Elasticsearch Troubleshooting and Mitigating Elasticsearch Out of Memory Issues

By Opster Team

Updated: Nov 7, 2023

| 2 min read

Quick links


Elasticsearch is known for its efficient memory management. However, there are instances when Elasticsearch may encounter Out of Memory (OOM) issues. This article delves into the causes, troubleshooting, and mitigation strategies for Elasticsearch Out of Memory issues.

Understanding the Causes

OOM issues in Elasticsearch can be attributed to various factors. One of the most common causes is the Java Virtual Machine (JVM) heap size. Elasticsearch runs on the JVM, and if the heap size is not configured correctly, it can lead to OOM errors. Other factors include excessive shard size, large mapping metadata, and heavy indexing or search operations.

Troubleshooting OOM Issues

1. JVM Heap Size: Check the JVM heap size. If it’s too small, Elasticsearch may not have enough memory to perform operations, leading to OOM errors. Conversely, if it’s too large, it can cause long garbage collection pauses, affecting performance. The recommended heap size is 50% of your total available memory, up to 30.5GB.

2. Shard Size: Large shard sizes can also cause OOM issues. It’s recommended to keep the shard size below 50GB for optimal performance. Use the _cat/shards API to check the shard sizes.

3. Mapping Metadata: Large mapping metadata can consume significant heap space. Use the _mapping API to check the size of your mapping metadata.

4. Indexing and Search Operations: Heavy indexing or search operations can cause memory pressure. Monitor your indexing and search rates using the _cat/indices API.

Mitigating OOM Issues

1. Adjust JVM Heap Size: If the heap size is the issue, adjust it according to the guidelines mentioned above. You can set the heap size in a custom JVM options file to be located in the `config/jvm.options.d` folder.

2. Optimize Shard Size: If the shard size is too large, consider reindexing your data with a smaller shard size. Use the _reindex API for this purpose.

3. Reduce Mapping Metadata: If the mapping metadata is too large, consider reducing it. This can be done by removing unnecessary fields or by using dynamic templates to control the mapping.

4. Throttle Indexing and Search Operations: If heavy indexing or search operations are causing OOM issues, consider throttling these operations. You can use the static `indices.memory.index_buffer_size` setting to control the amount of memory used for indexing, and the search thread pool queue size to control the number of concurrent search requests.

5. Use Circuit Breakers: Elasticsearch provides circuit breakers to prevent operations from consuming too much memory. Ensure that your circuit breaker settings are configured correctly to prevent OOM issues.

6. Monitor Your Cluster: Regularly monitor your cluster using the _cat APIs or the Elasticsearch monitoring features. This can help you identify potential issues before they cause OOM errors.


In conclusion, while Elasticsearch is designed to handle large amounts of data efficiently, it’s not immune to OOM issues. By understanding the causes, regularly monitoring your cluster, and taking appropriate mitigation steps, you can ensure that your Elasticsearch cluster remains healthy and performs optimally.

How helpful was this guide?

We are sorry that this post was not useful for you!

Let us improve this post!

Tell us how we can improve this post?