Loading

Disk watermark high

Disk usage on one or more nodes has crossed the high watermark. Elasticsearch begins relocating shards away from those nodes.

Note

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

Field Value
Component Elasticsearch
Severity High
Scope Node
Domains cluster-health, disk-storage, stability

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
Early warning margin before the high watermark Percentage 0

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.

The high disk watermark has been exceeded by 12 GB.

Affected nodes: es-data-01 and es-data-02.

Affected tier(s): hot, warm.

Remaining space before the flood-stage watermark: 48 GB.

Largest indices on the affected node(s): logs-prod-000045.

Note

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

When the high watermark is crossed, Elasticsearch actively relocates shards away from the affected nodes to relieve disk pressure. Writes to existing shards on those nodes continue until the flood-stage watermark is reached, at which point indices are made read-only. Act promptly. On hot nodes, high disk usage directly threatens ingest throughput. On warm or cold nodes, ILM transitions or data deletion are typically the right levers. On any tier, if relocation is not enough, expand disk capacity or apply index retention policies before the flood stage is reached.

Fix watermark errors