The Elasticsearch client is specified as follows:

index.search.backend=elasticsearch

When connecting to Elasticsearch a single or list of hostnames for the Elasticsearch instances must be provided. These are supplied via JanusGraph’s index.[X].hostname key.

index.search.backend=elasticsearch
index.search.hostname=10.0.0.10:9200

Each host or host:port pair specified here will be added to the HTTP client’s round-robin list of request targets. Here’s a minimal configuration that will round-robin over 10.0.0.10 on the default Elasticsearch HTTP port (9200) and 10.0.0.20 on port 7777:

index.search.backend=elasticsearch
index.search.hostname=10.0.0.10, 10.0.0.20:7777

The REST client accepts the index.[X].bulk-refresh option. This option controls when changes are made visible to search. See ?refresh documentation for more information.

SSL support for HTTP can be enabled by setting the index.[X].elasticsearch.ssl.enabled configuration option to true. Note that depending on your configuration you may need to change the value of index.[X].port if your HTTPS port number is different from the default one for the REST API (9200).

When SSL is enabled you may also configure the location and password of the truststore. This can be done as follows:

index.search.elasticsearch.ssl.truststore.location=/path/to/your/truststore.jks
index.search.elasticsearch.ssl.truststore.password=truststorepwd

Note that these settings apply only to Elasticsearch REST client and do not affect any other SSL connections in JanusGraph.

Configuration of the client keystore is also supported:

index.search.elasticsearch.ssl.keystore.location=/path/to/your/keystore.jks
index.search.elasticsearch.ssl.keystore.storepassword=keystorepwd
index.search.elasticsearch.ssl.keystore.keypassword=keypwd

Any of the passwords can be empty.

If needed, the SSL hostname verification can be disabled by setting the index.[X].elasticsearch.ssl.disable-hostname-verification property value to true and the support for self-signed SSL certificates can be enabled by setting index.[X].elasticsearch.ssl.allow-self-signed-certificates property value to true.

	Tip
	It is not recommended to rely on the self-signed SSL certificates or to disable the hostname verification for a production system as it significantly limits the client’s ability to provide the secure communication channel with the Elasticsearch server(s). This may result in leaking the confidential data which may be a part of your JanusGraph index.

REST client supports the following authentication options: Basic HTTP Authentication (username/password) and custom authentication based on the user-provided implementation.

These authentication methods are independent from SSL client authentication described above.

30.2.4.1. REST Client Basic HTTP Authentication

Basic HTTP Authentication is available regardless of the state of SSL support. Optionally, an authentication realm can be specified via index.[X].elasticsearch.http.auth.basic.realm property.

index.search.elasticsearch.http.auth.type=basic
index.search.elasticsearch.http.auth.basic.username=httpuser
index.search.elasticsearch.http.auth.basic.password=httppassword

	Tip
	It is highly recommended to use SSL (e.g. setting index.[X].elasticsearch.ssl.enabled to true) when using this option as the credentials can be intercepted when sent over an unencrypted connection!

index.search.elasticsearch.http.auth.custom.authenticator-class=fully.qualified.class.Name
index.search.elasticsearch.elasticsearch.http.auth.custom.authenticator-args=arg1,arg2,...

import java.io.IOException;
import java.time.LocalDateTime;
import java.time.ZoneOffset;

import org.apache.http.HttpRequestInterceptor;
import org.apache.http.impl.nio.client.HttpAsyncClientBuilder;
import org.janusgraph.diskstorage.es.rest.util.RestClientAuthenticatorBase;

import com.amazonaws.auth.DefaultAWSCredentialsProviderChain;
import com.amazonaws.regions.DefaultAwsRegionProviderChain;
import com.google.common.base.Supplier;

import vc.inreach.aws.request.AWSSigner;
import vc.inreach.aws.request.AWSSigningRequestInterceptor;

/**
 * <p>
 * Elasticsearch REST HTTP(S) client callback implementing AWS request signing.
 * </p>
 * <p>
 * The signer is based on AWS SDK default provider chain, allowing multiple options for providing
 * the caller credentials. See {@link DefaultAWSCredentialsProviderChain} documentation for the details.
 * </p>
 */
public class AWSV4AuthHttpClientConfigCallback extends RestClientAuthenticatorBase {

    private static final String AWS_SERVICE_NAME = "es";
    private HttpRequestInterceptor awsSigningInterceptor;

    public AWSV4AuthHttpClientConfigCallback(final String[] args) {
        // does not require any configuration
    }

    @Override
    public void init() throws IOException {
        DefaultAWSCredentialsProviderChain awsCredentialsProvider = new DefaultAWSCredentialsProviderChain();
        final Supplier<LocalDateTime> clock = () -> LocalDateTime.now(ZoneOffset.UTC);

        // using default region provider chain
        // (https://docs.aws.amazon.com/sdk-for-java/v2/developer-guide/java-dg-region-selection.html)
        DefaultAwsRegionProviderChain regionProviderChain = new DefaultAwsRegionProviderChain();
        final String awsRegion = regionProviderChain.getRegion();

        final AWSSigner awsSigner = new AWSSigner(awsCredentialsProvider, awsRegion, AWS_SERVICE_NAME, clock);
        this.awsSigningInterceptor = new AWSSigningRequestInterceptor(awsSigner);
    }

    @Override
    public HttpAsyncClientBuilder customizeHttpClient(HttpAsyncClientBuilder httpClientBuilder) {
        return httpClientBuilder.addInterceptorLast(awsSigningInterceptor);/
    }
}

If using Elasticsearch 5.0 or higher, a different ingest pipelines can be set for each mixed index. Ingest pipeline can be use to pre-process documents before indexing. A pipeline is composed by a series of processors. Each processor transforms the document in some way. For example date processor can extract a date from a text to a date field. So you can query this date with JanusGraph without it being physically in the primary storage.

See ingest documentation for more information about ingest pipelines and processors documentation for more information about ingest processors.

JanusGraph iterates over all properties prefixed with index.[X].elasticsearch.create.ext., where [X] is an index name such as search. It strips the prefix from each property key. The remainder of the stripped key will be interpreted as an Elasticsearch index creation setting. The value associated with the key is not modified. The stripped key and unmodified value are passed as part of the settings object in the Elasticsearch create index request that JanusGraph issues when bootstrapping on Elasticsearch. This allows embedding arbitrary index creation settings settings in JanusGraph’s properties. Here’s an example configuration fragment that customizes three Elasticsearch index settings using the create.ext config mechanism:

index.search.backend=elasticsearch
index.search.elasticsearch.create.ext.number_of_shards=15
index.search.elasticsearch.create.ext.number_of_replicas=3
index.search.elasticsearch.create.ext.shard.check_on_startup=true

The configuration fragment listed above takes advantage of Elasticsearch’s assumption, implemented server-side, that unqualified create index setting keys have an index. prefix. It’s also possible to spell out the index prefix explicitly. Here’s a JanusGraph config file functionally equivalent to the one listed above, except that the index. prefix before the index creation settings is explicit:

index.search.backend=elasticsearch
index.search.elasticsearch.create.ext.index.number_of_shards=15
index.search.elasticsearch.create.ext.index.number_of_replicas=3
index.search.elasticsearch.create.ext.index.shard.check_on_startup=false

	Tip
	The create.ext mechanism for specifying index creation settings is compatible with JanusGraph’s Elasticsearch configuration.

Check that the Elasticsearch cluster nodes are reachable on the HTTP protocol port from the JanusGraph nodes. Check the node listen port by examining the Elasticsearch node configuration logs or using a general diagnostic utility like netstat. Check the JanusGraph configuration.

For bulk loading or other write-intense applications, consider increasing Elasticsearch’s refresh interval. Refer to this discussion on how to increase the refresh interval and its impact on write performance. Note, that a higher refresh interval means that it takes a longer time for graph mutations to be available in the index.

For additional suggestions on how to increase write performance in Elasticsearch with detailed instructions, please read this blog post.