Cross-cluster replication API

Use these replication operations to programmatically manage cross-cluster replication.

Start replication

Introduced 1.1

Initiate replication of an index from the leader cluster to the follower cluster. Send this request to the follower cluster.

Request

  1. PUT /_plugins/_replication/<follower-index>/_start
  2. {
  3.    "leader_alias":"<connection-alias-name>",
  4.    "leader_index":"<index-name>",
  5.    "use_roles":{
  6.       "leader_cluster_role":"<role-name>",
  7.       "follower_cluster_role":"<role-name>"
  8.    }
  9. }

Specify the following options:

OptionsDescriptionTypeRequired
leader_aliasThe name of the cross-cluster connection. You define this alias when you set up a cross-cluster connection.stringYes
leader_indexThe index on the leader cluster that you want to replicate.stringYes
use_rolesThe roles to use for all subsequent backend replication tasks between the indexes. Specify a leader_cluster_role and follower_cluster_role. See Map the leader and follower cluster roles.stringIf security plugin is enabled

Sample response

  1. {
  2.    "acknowledged": true
  3. }

Stop replication

Introduced 1.1

Terminates replication and converts the follower index to a standard index. Send this request to the follower cluster.

Request

  1. POST /_plugins/_replication/<follower-index>/_stop
  2. {}

Sample response

  1. {
  2.    "acknowledged": true
  3. }

Pause replication

Introduced 1.1

Pauses replication of the leader index. Send this request to the follower cluster.

Request

  1. POST /_plugins/_replication/<follower-index>/_pause 
  2. {}

You can’t resume replication after it’s been paused for more than 12 hours. You must stop replication, delete the follower index, and restart replication of the leader.

Sample response

  1. {
  2.    "acknowledged": true
  3. }

Resume replication

Introduced 1.1

Resumes replication of the leader index. Send this request to the follower cluster.

Request

  1. POST /_plugins/_replication/<follower-index>/_resume
  2. {}

Sample response

  1. {
  2.    "acknowledged": true
  3. }

Get replication status

Introduced 1.1

Gets the status of index replication. Possible statuses are SYNCING, BOOTSTRAPING, PAUSED, and REPLICATION NOT IN PROGRESS. Use the syncing details to measure replication lag. Send this request to the follower cluster.

Request

  1. GET /_plugins/_replication/<follower-index>/_status

Sample response

  1. {
  2.   "status" : "SYNCING",
  3.   "reason" : "User initiated",
  4.   "leader_alias" : "my-connection-name",
  5.   "leader_index" : "leader-01",
  6.   "follower_index" : "follower-01",
  7.   "syncing_details" : {
  8.     "leader_checkpoint" : 19,
  9.     "follower_checkpoint" : 19,
  10.     "seq_no" : 0
  11.   }
  12. }

To include shard replication details in the response, add the &verbose=true parameter.

The leader and follower checkpoint values begin as negative integers and reflect the shard count (-1 for one shard, -5 for five shards, and so on). The values increment toward positive integers with each change that you make. For example, when you make a change on the leader index, the leader_checkpoint becomes 0. The follower_checkpoint is initially still -1 until the follower index pulls the change from the leader, at which point it increments to 0. If the values are the same, it means the indexes are fully synced.

Get leader cluster stats

Introduced 1.1

Gets information about replicated leader indexes on a specified cluster.

Request

  1. GET /_plugins/_replication/leader_stats

Sample response

  1. {
  2.    "num_replicated_indices": 2,
  3.    "operations_read": 15,
  4.    "translog_size_bytes": 1355,
  5.    "operations_read_lucene": 0,
  6.    "operations_read_translog": 15,
  7.    "total_read_time_lucene_millis": 0,
  8.    "total_read_time_translog_millis": 659,
  9.    "bytes_read": 1000,
  10.    "index_stats":{
  11.       "leader-index-1":{
  12.          "operations_read": 7,
  13.          "translog_size_bytes": 639,
  14.          "operations_read_lucene": 0,
  15.          "operations_read_translog": 7,
  16.          "total_read_time_lucene_millis": 0,
  17.          "total_read_time_translog_millis": 353,
  18.          "bytes_read":466
  19.       },
  20.       "leader-index-2":{
  21.          "operations_read": 8,
  22.          "translog_size_bytes": 716,
  23.          "operations_read_lucene": 0,
  24.          "operations_read_translog": 8,
  25.          "total_read_time_lucene_millis": 0,
  26.          "total_read_time_translog_millis": 306,
  27.          "bytes_read": 534
  28.       }
  29.    }
  30. }

Get follower cluster stats

Introduced 1.1

Gets information about follower (syncing) indexes on a specified cluster.

Request

  1. GET /_plugins/_replication/follower_stats

Sample response

  1. {
  2.    "num_syncing_indices": 2,
  3.    "num_bootstrapping_indices": 0,
  4.    "num_paused_indices": 0,
  5.    "num_failed_indices": 0,
  6.    "num_shard_tasks": 2,
  7.    "num_index_tasks": 2,
  8.    "operations_written": 3,
  9.    "operations_read": 3,
  10.    "failed_read_requests": 0,
  11.    "throttled_read_requests": 0,
  12.    "failed_write_requests": 0,
  13.    "throttled_write_requests": 0,
  14.    "follower_checkpoint": 1,
  15.    "leader_checkpoint": 1,
  16.    "total_write_time_millis": 2290,
  17.    "index_stats":{
  18.       "follower-index-1":{
  19.          "operations_written": 2,
  20.          "operations_read": 2,
  21.          "failed_read_requests": 0,
  22.          "throttled_read_requests": 0,
  23.          "failed_write_requests": 0,
  24.          "throttled_write_requests": 0,
  25.          "follower_checkpoint": 1,
  26.          "leader_checkpoint": 1,
  27.          "total_write_time_millis": 1355
  28.       },
  29.       "follower-index-2":{
  30.          "operations_written": 1,
  31.          "operations_read": 1,
  32.          "failed_read_requests": 0,
  33.          "throttled_read_requests": 0,
  34.          "failed_write_requests": 0,
  35.          "throttled_write_requests": 0,
  36.          "follower_checkpoint": 0,
  37.          "leader_checkpoint": 0,
  38.          "total_write_time_millis": 935
  39.       }
  40.    }
  41. }

Get auto-follow stats

Introduced 1.1

Gets information about auto-follow activity and any replication rules configured on the specified cluster.

Request

  1. GET /_plugins/_replication/autofollow_stats

Sample response

  1. {
  2.    "num_success_start_replication": 2,
  3.    "num_failed_start_replication": 0,
  4.    "num_failed_leader_calls": 0,
  5.    "failed_indices":[
  7.    ],
  8.    "autofollow_stats":[
  9.       {
  10.          "name":"my-replication-rule",
  11.          "pattern":"movies*",
  12.          "num_success_start_replication": 2,
  13.          "num_failed_start_replication": 0,
  14.          "num_failed_leader_calls": 0,
  15.          "failed_indices":[
  17.          ]
  18.       }
  19.    ]
  20. }

Update settings

Introduced 1.1

Updates settings on the follower index.

Request

  1. PUT /_plugins/_replication/<follower-index>/_update
  2. {
  3.    "settings":{
  4.       "index.number_of_shards": 4,
  5.       "index.number_of_replicas": 2
  6.    }
  7. }

Sample response

  1. {
  2.    "acknowledged": true
  3. }

Create replication rule

Introduced 1.1

Automatically starts replication on indexes matching a specified pattern. If a new index on the leader cluster matches the pattern, OpenSearch automatically creates a follower index and begins replication. You can also use this API to update existing replication rules.

Send this request to the follower cluster.

Make sure to note the names of all auto-follow patterns after you create them. The replication plugin currently does not include an API operation to retrieve a list of existing patterns.

Request

  1. POST /_plugins/_replication/_autofollow
  2. {
  3.    "leader_alias" : "<connection-alias-name>",
  4.    "name": "<auto-follow-pattern-name>",
  5.    "pattern": "<pattern>",
  6.    "use_roles":{
  7.       "leader_cluster_role": "<role-name>",
  8.       "follower_cluster_role": "<role-name>"
  9.    }
  10. }

Specify the following options:

OptionsDescriptionTypeRequired
leader_aliasThe name of the cross-cluster connection. You define this alias when you set up a cross-cluster connection.stringYes
nameA name for the auto-follow pattern.stringYes
patternAn array of index patterns to match against indexes in the specified leader cluster. Supports wildcard characters. For example, leader-*.stringYes
use_rolesThe roles to use for all subsequent backend replication tasks between the indexes. Specify a leader_cluster_role and follower_cluster_role. See Map the leader and follower cluster roles.stringIf security plugin is enabled

Sample response

  1. {
  2.    "acknowledged": true
  3. }

Delete replication rule

Introduced 1.1

Deletes the specified replication rule. This operation prevents any new indexes from being replicated but does not stop existing replication that the rule has already initiated. Replicated indexes remain read-only until you stop replication.

Send this request to the follower cluster.

Request

  1. DELETE /_plugins/_replication/_autofollow
  2. {
  3.    "leader_alias" : "<connection-alias-name>",
  4.    "name": "<auto-follow-pattern-name>",
  5. }

Specify the following options:

OptionsDescriptionTypeRequired
leader_aliasThe name of the cross-cluster connection. You define this alias when you set up a cross-cluster connection.stringYes
nameThe name of the pattern.stringYes

Sample response

  1. {
  2.    "acknowledged": true
  3. }