Link Search Menu Expand Document Documentation Menu
Documentation

Availability and recovery settings

Availability and recovery settings include settings for the following:

To learn more about static and dynamic settings, see Configuring OpenSearch.

General recovery settings

OpenSearch supports the following general recovery settings:

  • indices.recovery.chunk_size (Dynamic, byte unit): Controls the chunk size used when transferring data during index recovery operations. This setting affects the amount of data transferred in each network request during shard recovery. Larger chunk sizes can improve recovery speed but may increase memory usage. Default is 512kb.

  • indices.recovery.recovery_activity_timeout (Dynamic, time unit): Sets the timeout for individual recovery activities during shard recovery operations. If a recovery activity (such as transferring a file chunk) takes longer than this timeout, the recovery operation is considered failed and will be retried. Default is 30m.

Snapshot settings

OpenSearch supports the following snapshot settings:

  • snapshot.max_concurrent_operations (Dynamic, integer): The maximum number of concurrent snapshot operations. Default is 1000.

  • snapshot.repository_data.cache.threshold (Static, byte size value or percentage): The maximum size of repository metadata that can be cached in memory. This setting improves snapshot operation performance by reducing the need to repeatedly download metadata during clone, restore, and status check operations. You can specify this value as an absolute size (for example, 2gb or 500mb) or as a percentage of heap memory (for example, 3% or 1%). Metadata that exceeds this threshold is not cached. Under heap memory pressure, cached data may be garbage collected automatically because it is stored using soft references. Default is 500 kb or 1% of heap memory, whichever is higher.

For security-related snapshot settings, see Security settings.

Shared file system

For information about using a shared file system, see Shared file system.

Amazon S3 settings

For information about Amazon S3 repository settings, see Amazon S3.

Cluster manager task throttling settings

For information about cluster manager task throttling settings, see Setting throttling limits.

Remote-backed storage settings

OpenSearch supports the following cluster-level remote-backed storage settings:

  • cluster.remote_store.translog.buffer_interval (Dynamic, time unit): The default value of the translog buffer interval used when performing periodic translog updates. This setting is only effective when the index setting index.remote_store.translog.buffer_interval is not present.

For more remote-backed storage settings, see Remote-backed storage and Configuring remote-backed storage.

For remote segment backpressure settings, see Remote segment backpressure settings.

For remote segment warmer settings, see Remote segment warmer settings.

Search backpressure settings

Search backpressure is a mechanism used to identify resource-intensive search requests and cancel them when the node is under duress. For more information, see Search backpressure settings.

Shard indexing backpressure settings

Shard indexing backpressure is a smart rejection mechanism at a per-shard level that dynamically rejects indexing requests when your cluster is under strain. For more information, see shard indexing backpressure settings.

Segment replication settings

For information about segment replication settings, see Segment replication.

For information about segment replication backpressure settings, see Segment replication backpressure.

Cross-cluster replication settings

For information about cross-cluster replication settings, see Replication settings.

350 characters left

Have a question? .

Want to contribute? or .