Write hotspotting

Write work is backing up on a subset of write-eligible nodes while peer nodes in the same eligible set still have spare capacity. This usually points to uneven shard placement or competing search and indexing work on the busy nodes, which raises ingest latency and leaves available write capacity unused.

Note

For a complete list of insights, refer to AutoOps insights.

Insight details

Field	Value
Component	Elasticsearch
Severity	Medium
Scope	Node
Domains	performance, search, indexing, resource-utilization, shard-allocation

Customization settings

You can customize these settings to adjust when AutoOps detects this event and presents the insight. Refer to AutoOps event settings for details.

The default customization settings are:

Setting	Type	Default
Write queue threshold to count a node as overloaded	Integer	`50`
Consecutive samples meeting the imbalance conditions	Integer	`3`

Tip

Raising these thresholds reduces noise but delays detection. Lowering them triggers the insight sooner but can increase alerts during minor blips.

Example: What you might see in AutoOps

The following is an example of what you might see when this insight is triggered. Real insights use live data and links from your deployment or cluster.

Write work is concentrated on es-data-01

What was detected

The write thread pool queue is elevated on es-data-01 and es-data-02 while other write-eligible nodes in the same set have no writes waiting.

Indices with high indexing activity: logs-prod-000045
Indices with high search activity on the same nodes: logs-prod-000045

Related open insights on these nodes:

Disk watermark flood stage
Deployment not highly available

Inspect open insights

Recommendations

Note

AutoOps shows different recommendations depending on how their conditions match your deployment or cluster.

Resolve open events on the overloaded nodes

Condition: Shown when hotspot-related events are open.

These open events on the overloaded nodes might explain competing work or how far this write hotspot has gone. Open the list and resolve each event, then recheck write queue depth.

[Inspect open events](es-data-01 and es-data-02)

Review active indices on node

Condition: Always shown for this insight.

Review search and indexing load on es-data-01 for the indices listed in this insight. Tune or relocate hot indices if they dominate node activity.

Review query logs

Condition: Shown when search is competing on the overloaded nodes and cluster version is 9.4 or above.

Review query logs on es-data-01 to find expensive searches that fill the queue. Query logs (Elasticsearch 9.4+) give structured per-query timing with less overhead than search slow logs. See Query logs.

Enable and review search slow logs

Condition: Shown when search is competing on the overloaded nodes and cluster version is below 9.4.

Enable search slow logs with the action below, then review the slow log to find expensive queries. See Slow logs for configuration details.

PUT logs-prod-000045/_settings
{
  "index.search.slowlog.threshold.query.warn": "10s",
  "index.search.slowlog.threshold.query.info": "5s",
  "index.search.slowlog.threshold.query.debug": "2s",
  "index.search.slowlog.threshold.query.trace": "500ms",
  "index.search.slowlog.threshold.fetch.warn": "1s",
  "index.search.slowlog.threshold.fetch.info": "800ms",
  "index.search.slowlog.threshold.fetch.debug": "500ms",
  "index.search.slowlog.threshold.fetch.trace": "200ms"
}

Note

Requires the manage index privilege. Requires Elasticsearch 8.0.0 or later. This action changes cluster or index configuration.

Cancel long-running tasks

Condition: Shown when one or more related long-running tasks on the overloaded nodes are cancellable.

Cancel long-running tasks (es-data-01:4521, es-data-02:7788) that are affecting cluster stability. Use the action below.

POST _tasks/es-data-01:4521, es-data-02:7788/_cancel

Note

Requires the manage cluster privilege. Requires Elasticsearch 8.0.0 or later. This action changes cluster or index configuration.

Increase replica count

Condition: Shown when high search activity is present on the overloaded nodes and the affected index has no replica.

Set number_of_replicas to 1 on logs-prod-000045 (currently 2) using the action below.

PUT logs-prod-000045/_settings
{
  "index": {
    "number_of_replicas": 1
  }
}

Note

Requires the manage index privilege. Requires Elasticsearch 8.0.0 or later. This action changes cluster or index configuration.

Rebalance index shards

Condition: Shown when the same hot index appears in both indexing and search activity, or write-heavy shards are not well spread across write-eligible nodes.

Node es-data-01 holds 1000 shards while es-data-02 holds 12. Move shard 0 of index logs-prod-000045 from es-data-01 to es-data-02 using the action below.

POST _cluster/reroute
{
  "commands": [
    {
      "move": {
        "index": "logs-prod-000045",
        "shard": 0,
        "from_node": "es-data-01",
        "to_node": "es-data-02"
      }
    }
  ]
}

Note

Requires the manage cluster privilege. Requires Elasticsearch 8.0.0 or later. This action changes cluster or index configuration.

Rollover indices

Condition: Shown when search is quiet on the overloaded nodes, high indexing activity is detected, and the hot index has fewer primary shards than the available write-eligible data nodes.

If logs-prod-000045 are time-based, roll them over and set 1 primary shards on the new write index so you can avoid a heavy segment merge on the current indices.

Rollover indices by shard size

Condition: Shown when search is quiet on the overloaded nodes, high indexing activity is detected, and primary shard size is above 30GB.

If logs-prod-000045 are time-based, roll them over when average shard size reaches your target.

Review index templates and mappings

Condition: Shown when search is quiet on the overloaded nodes and high indexing activity is concentrated on the indices with the highest indexing latency.

Review templates and mappings for logs-prod-000045; they strongly affect indexing performance. Fix oversized mappings, unnecessary fields, and inefficient index settings.

Review indexing slow logs

Condition: Shown when search is quiet on the overloaded nodes and high indexing activity is detected on the node.

Enable indexing slow logs with the action below on logs-prod-000045, then review the logs to find slow index operations. On Elasticsearch 8.14+, set index.indexing.slowlog.include.user to true to record which user triggered a slow operation.

PUT logs-prod-000045/_settings
{
  "index.indexing.slowlog.threshold.index.warn": "10s",
  "index.indexing.slowlog.threshold.index.info": "5s",
  "index.indexing.slowlog.threshold.index.debug": "2s",
  "index.indexing.slowlog.threshold.index.trace": "500ms",
  "index.indexing.slowlog.source": "1000"
}

Note

Requires the manage index privilege. Requires Elasticsearch 8.0.0 or later. This action changes cluster or index configuration.

Add data node

Condition: Shown when search is quiet on the overloaded nodes, high indexing activity is detected, and the hot index already has at least as many primary shards as the available write-eligible data nodes.

Add a data node to increase capacity and reduce pressure on the existing nodes.

Background and impact

A write hotspot happens when indexing pressure concentrates on a few write-eligible nodes instead of being spread across the tier. This often reflects uneven shard placement, hot shards, or competing search, merge, and indexing work on the same nodes.

If the imbalance persists, ingest latency rises on the busy nodes, write queues can grow into rejections, and searches on those nodes can also slow down. Other write-eligible nodes stay underused, so the problem is distribution of work rather than uniform cluster-wide saturation.

Guide

Hotspotting