Remote clusters

You can connect a local cluster to other Elasticsearch clusters, known as remote clusters. Once connected, you can search remote clusters using cross-cluster search. You can also sync data between clusters using cross-cluster replication.

To register a remote cluster, connect the local cluster to nodes in the remote cluster using one of two connection modes:

Your local cluster uses the transport layer to establish communication with remote clusters. The coordinating nodes in the local cluster establish long-lived TCP connections with specific nodes in the remote cluster. Elasticsearch requires these connections to remain open, even if the connections are idle for an extended period.

You can use the remote cluster info API to get information about registered remote clusters.

Sniff mode

In sniff mode, a cluster is created using a name and a list of seed nodes. When a remote cluster is registered, its cluster state is retrieved from one of the seed nodes and up to three gateway nodes are selected as part of remote cluster requests. This mode requires that the gateway node’s publish addresses are accessible by the local cluster.

Sniff mode is the default connection mode.

The gateway nodes selection depends on the following criteria:

  • version: Remote nodes must be compatible with the cluster they are registered to, similar to the rules for Rolling upgrades:

    • Any node can communicate with another node on the same major version. For example, 7.0 can talk to any 7.x node.
    • Only nodes on the last minor version of a certain major version can communicate with nodes on the following major version. In the 6.x series, 6.8 can communicate with any 7.x node, while 6.7 can only communicate with 7.0.
    • Version compatibility is symmetric, meaning that if 6.7 can communicate with 7.0, 7.0 can also communicate with 6.7. The following table depicts version compatibility between local and remote nodes.

      Version compatibility table

      Local cluster

      Remote cluster

      5.0→5.5

      5.6

      6.0→6.6

      6.7

      6.8

      7.0

      7.1→7.x

      5.0→5.5

      Yes

      Yes

      No

      No

      No

      No

      No

      5.6

      Yes

      Yes

      Yes

      Yes

      Yes

      No

      No

      6.0→6.6

      No

      Yes

      Yes

      Yes

      Yes

      No

      No

      6.7

      No

      Yes

      Yes

      Yes

      Yes

      Yes

      No

      6.8

      No

      Yes

      Yes

      Yes

      Yes

      Yes

      Yes

      7.0

      No

      No

      No

      Yes

      Yes

      Yes

      Yes

      7.1→7.x

      No

      No

      No

      No

      Yes

      Yes

      Yes

  • role: Dedicated master nodes are never selected as gateway nodes.
  • attributes: You can tag which nodes should be selected (see Global remote cluster settings), though such tagged nodes still have to satisfy the two above requirements.

Proxy mode

In proxy mode, a cluster is created using a name and a single proxy address. When you register a remote cluster, a configurable number of socket connections are opened to the proxy address. The proxy is required to route those connections to the remote cluster. Proxy mode does not require remote cluster nodes to have accessible publish addresses.

The proxy mode is not the default connection mode and must be configured. Similar to the sniff gateway nodes, the remote connections are subject to the same version compatibility rules as Rolling upgrades.

Configuring remote clusters

You can configure remote clusters settings globally, or configure settings on individual nodes in the elasticsearch.yml file.

Dynamically configure remote clusters

Use the cluster update settings API to dynamically configure remote settings on every node in the cluster. For example:

PUT _cluster/settings
{
  "persistent": {
    "cluster": {
      "remote": {
        "cluster_one": {
          "seeds": [
            "127.0.0.1:9300"
          ],
          "transport.ping_schedule": "30s"
        },
        "cluster_two": {
          "mode": "sniff",
          "seeds": [
            "127.0.0.1:9301"
          ],
          "transport.compress": true,
          "skip_unavailable": true
        },
        "cluster_three": {
          "mode": "proxy",
          "proxy_address": "127.0.0.1:9302"
        }
      }
    }
  }
}

You can dynamically update the compression and ping schedule settings. However, you must re-include seeds or proxy_address in the settings update request. For example:

PUT _cluster/settings
{
  "persistent": {
    "cluster": {
      "remote": {
        "cluster_one": {
          "seeds": [
            "127.0.0.1:9300"
          ],
          "transport.ping_schedule": "60s"
        },
        "cluster_two": {
          "mode": "sniff",
          "seeds": [
            "127.0.0.1:9301"
          ],
          "transport.compress": false
        },
        "cluster_three": {
          "mode": "proxy",
          "proxy_address": "127.0.0.1:9302",
          "transport.compress": true
        }
      }
    }
  }
}

When the compression or ping schedule settings change, all the existing node connections must close and re-open, which can cause in-flight requests to fail.

You can delete a remote cluster from the cluster settings by passing null values for each remote cluster setting:

PUT _cluster/settings
{
  "persistent": {
    "cluster": {
      "remote": {
        "cluster_two": { 
          "mode": null,
          "seeds": null,
          "skip_unavailable": null,
          "transport": {
            "compress": null
          }
        }
      }
    }
  }
}

cluster_two would be removed from the cluster settings, leaving cluster_one and cluster_three intact.

Statically configure remote clusters

If you specify settings in elasticsearch.yml files, only the nodes with those settings can connect to the remote cluster and serve remote cluster requests. For example:

cluster:
    remote:
        cluster_one: 
            seeds: 127.0.0.1:9300 
            transport.ping_schedule: 30s 
        cluster_two: 
            mode: sniff 
            seeds: 127.0.0.1:9301 
            transport.compress: true 
            skip_unavailable: true 
        cluster_three: 
            mode: proxy 
            proxy_address: 127.0.0.1:9302 

cluster_one, cluster_two, and cluster_three are arbitrary cluster aliases representing the connection to each cluster. These names are subsequently used to distinguish between local and remote indices.

The hostname and transport port (default: 9300) of a seed node in the remote cluster.

A keep-alive ping is configured for cluster_one.

The configured connection mode. By default, this is sniff, so the mode is implicit for cluster_one. However, it can be explicitly configured as demonstrated by cluster_two and must be explicitly configured for proxy mode as demonstrated by cluster_three.

Compression is explicitly enabled for requests to cluster_two.

Disconnected remote clusters are optional for cluster_two.

The address for the proxy endpoint used to connect to cluster_three.

Global remote cluster settings

These settings apply to both sniff mode and proxy mode. Sniff mode settings and proxy mode settings are described separately.

cluster.remote.<cluster_alias>.mode
The mode used for a remote cluster connection. The only supported modes are sniff and proxy.
cluster.remote.initial_connect_timeout
The time to wait for remote connections to be established when the node starts. The default is 30s.
node.remote_cluster_client
By default, any node in the cluster can act as a cross-cluster client and connect to remote clusters. The node.remote_cluster_client setting can be set to false (defaults to true) to prevent certain nodes from connecting to remote clusters. Remote cluster requests must be sent to a node that is allowed to act as a cross-cluster client.
cluster.remote.<cluster_alias>.skip_unavailable
Per cluster boolean setting that allows to skip specific clusters when no nodes belonging to them are available and they are the target of a remote cluster request. Default is false, meaning that all clusters are mandatory by default, but they can selectively be made optional by setting this setting to true.
cluster.remote.<cluster_alias>.transport.ping_schedule
Sets the time interval between regular application-level ping messages that are sent to ensure that transport connections to nodes belonging to remote clusters are kept alive. If set to -1, application-level ping messages to this remote cluster are not sent. If unset, application-level ping messages are sent according to the global transport.ping_schedule setting, which defaults to -1 meaning that pings are not sent.
cluster.remote.<cluster_alias>.transport.compress
Per cluster boolean setting that enables you to configure compression for requests to a specific remote cluster. This setting impacts only requests sent to the remote cluster. If the inbound request is compressed, Elasticsearch compresses the response. If unset, the global transport.compress is used as the fallback setting.

Sniff mode remote cluster settings

cluster.remote.<cluster_alias>.seeds
The list of seed nodes used to sniff the remote cluster state.
cluster.remote.<cluster_alias>.node_connections
The number of gateway nodes to connect to for this remote cluster. The default is 3.
cluster.remote.node.attr
A node attribute to filter out nodes that are eligible as a gateway node in the remote cluster. For instance a node can have a node attribute node.attr.gateway: true such that only nodes with this attribute will be connected to if cluster.remote.node.attr is set to gateway.

Proxy mode remote cluster settings

cluster.remote.<cluster_alias>.proxy_address
The address used for all remote connections.
cluster.remote.<cluster_alias>.proxy_socket_connections
The number of socket connections to open per remote cluster. The default is 18.
cluster.remote.<cluster_alias>.server_name
An optional hostname string which is sent in the server_name field of the TLS Server Name Indication extension if TLS is enabled. The TLS transport will fail to open remote connections if this field is not a valid hostname as defined by the TLS SNI specification.