Elasticsearch Cluster Health Red: How to Fix It
Workaround available Self-hosted, Elastic Cloud Updated November 15, 2026Based on 2 user reportsHow we research
Quick answer
Red means at least one primary shard is unassigned, so some data is unavailable. Use the allocation explain API to see why, and restore the node or a snapshot before forcing anything.
Most likely fix: Diagnose
How to fix it
Try these in order — the most common fix is first.
1. Diagnose
- GET _cluster/health?level=indices to find red indices
- GET _cat/shards?v and look for UNASSIGNED primaries
- GET _cluster/allocation/explain for the failing shard
Did this fix it?2. Bring the missing node back
- Restart or repair the node that left the cluster
- After it rejoins, wait for automatic allocation or run POST _cluster/reroute with no body
Did this fix it?3. Recover from a snapshot or reindex
- If the error is no_valid_shard_copy, restore the index from a snapshot, or reindex from the source system
- Delete the index only if the data is expendable
Did this fix it?4. Last resort: allocate a primary with data loss
- POST _cluster/reroute with allocate_stale_primary or allocate_empty_primary and accept_data_loss: true
- Elastic warns this can overwrite data if the original node later rejoins; use only when recovery is impossible
Did this fix it?
Symptoms
GET _cluster/health shows status red; searches on some indices fail or return partial results; Kibana reports shard failures.
Why it happens
- A node holding the only copy of a primary shard is down
- Disk watermark blocked allocation
- Corrupted or lost shard data (no_valid_shard_copy)
- High JVM memory pressure tripping circuit breakers during recovery
This guide also covers
- elasticsearch cluster health red
- elasticsearch red status fix
- elasticsearch primary shard unassigned
Still not working?
If Elasticsearch keeps failing, you're not alone. See every Elasticsearch problem users report or the alternatives people switch to.
Have an app idea? Our small team builds any software: web and mobile apps, games, desktop, AI tools. Price agreed together.
Talk to usOther Elasticsearch errors
- Elasticsearch Cluster Health Yellow: Causes and Fixes
- Elasticsearch Unassigned Shards: Why and How to Fix
- Elasticsearch circuit_breaking_exception: Data Too Large
- Elasticsearch High JVM Memory Pressure and Out of Memory
- Elasticsearch Disk Watermark: Index Read-Only (FORBIDDEN/12)
- Elasticsearch Limit of Total Fields [1000] Exceeded
- Why Is Elasticsearch So Slow? Fixing Slow Searches
- Elasticsearch Cannot Change Field Mapping: Reindex Guide