OpenSearch Index Read-Only: Disk Watermark and ClusterBlockException

How to fix it

// try these in order, the most common fix is first

1. FIX 1Free disk space

   Copy steps
   1. 01GET _cat/allocation?v&s=disk.percent:desc
   2. 02Delete old indices or add storage/nodes
   Did this fix it? Yes No

2. FIX 2Clear the block

   Copy steps
   1. 01PUT _all/_settings {"index.blocks.read_only_allow_delete": null}
   2. 02Some sources report the block clears itself on newer versions; verify with GET my-index/_settings
   Did this fix it? Yes No

3. FIX 3Prevent recurrence

   Copy steps
   1. 01Create alerts on free storage
   2. 02Use ISM to delete or roll over old indices
   Did this fix it? Yes No

4. FIX 4Do not just raise the watermark

   1. 01Setting flood-stage to 99% or disabling thresholds only hides the problem
   Did this fix it? Yes No

Symptoms

Writes fail with ClusterBlockException or 'index read-only / allow delete (api)'; new indexes cannot be created.

Why it happens

• C-01Disk usage above the flood-stage watermark
• C-02On managed domains less than 20% or 20 GiB free
• C-03JVM memory pressure above 92% triggers write block on Amazon's service

// this guide also covers

• opensearch index read only
• opensearch clusterblockexception
• opensearch flood stage watermark
• opensearch read-only allow delete

The same problem in other apps

All plan limits, paywalls & watermarks guides →

• Elasticsearch Disk Watermark: Index Read-Only (FORBIDDEN/12) (Elasticsearch)
• Algolia Record Quota Exceeded or Limit Reached (Algolia)

Other OpenSearch errors

• OpenSearch Cluster Status Red
• OpenSearch Cluster Status Yellow
• OpenSearch Unassigned Shards
• OpenSearch High JVM Memory Pressure
• OpenSearch OutOfMemoryError and Circuit Breaker
• OpenSearch Exceeded Maximum Shard Limit
• OpenSearch 429 Too Many Requests
• OpenSearch Docker Container Won't Start