Alfresco Search Enterprise


Alfresco Search Enterprise 3.0 consists of Alfresco Content Services, Elasticsearch Server, and the Elasticsearch connector. Use this information to install the Elasticsearch connector, which can be deployed using either JAR files, Docker Compose, or Helm.


  • Alfresco Content Services 7.1 that includes Alfresco ActiveMQ, Alfresco Transform Service, and Database, for more see Install overview.
  • Elasticsearch server 7.10. It may be used as a standard managed service or can be installed with the default configuration, for more see Install Elasticsearch server.
  • Elasticsearch Connector 3.0

See the Supported platforms for more.

Configure Subsystem in Repository

You must first activate and configure the Search Services subsystem in Content Services by using either the TOMCAT_HOME>/shared/classes/ file or the Repository Admin Web Console.

Add the following lines to the configuration file to enable the Elasticsearch Search subsystem.

# Set the Elasticsearch subsystem
# Elasticsearch index properties
# Elasticsearch server properties

These configuration properties are used by Content Services to communicate with the Elasticsearch server. In the example above a plain HTTP connection is configured, but Alfresco Repository also supports communication with Elasticsearch server using Basic Authentication and the HTTPs protocol, for more see Subsystem.

If using the Repository Admin Web Console select Repository Services > Search Service and set the properties from that web page, see below.


Note: In Content Services 7.1, the Test Connection button will fail if your Elasticsearch server does not include the “alfresco” index. You are however still able to Save your configuration and the index will be created automatically.

Install using JAR files

Use this information to install the Elasticsearch connector on the same machine as Content Services using JAR files.

  1. Download the file from Hyland Community and extract it.

  2. Verify all the required services are available:

  • Alfresco ActiveMQ, by default nio://activemq:61616
  • Alfresco Shared FileStore endpoint, by default http://shared-file-store:8099/alfresco/api/-default-/private/sfs/versions/1/file/
  • Alfresco Database using PostgresSQL engine, by default localhost:5432
  • Elasticsearch server, by default http://elasticsearch:9200

Once you have extracted the Elasticsearch connector zip file you install the Alfresco re-indexing app, and then the Alfresco live indexing app.

Alfresco Re-indexing app

The Elasticsearch connector Live Indexing component listens to messages from ActiveMQ. This means some initial information from Alfresco Repository must be indexed using the Re-indexing component. The Re-indexing component can also be used to index a pre-populated Alfresco Repository.

  1. Generate a JSON mapping of namespace to prefix, for your deployed content models.

    To help you build the JSON file you can use Alfresco Model Namespace-Prefix Mapping.

  2. Copy the JAR deployment file for this module to your local Alfresco Repository deployment, for detailed information on how to deploy a Simple JAR Alfresco Repository see Simple Module (JAR).

  3. Once installed, the JSON mapping file can be obtained by using http://localhost:8080/alfresco/s/model/ns-prefix-map, see the Simplified response:

       "prefixUriMap": {
         "": "cmm",
         "": "dl",
         "": "emailserver",
         "": "act",
         "": "sys",
         "": "cmis",
         "": "bpm",
         "": "d",
         "": "lnk",
         "": "imwf",
         "": "inwf",
         "": "cm",
         "": "smf",
         "": "cmiscustom",
         "": "st",
         "": "app",
         "": "imap",
         "": "aos",
         "custom.model": "custom",
         "": ""
  4. Save this content in a new file named reindex.prefixes-file.json.

  5. Run the re-indexing application from the command line by passing the already generated JSON file and details for the Database and Elasticsearch servers.

Note: Due to this application providing default values for Alfresco Repository Database username and password, it’s strongly recommended you set these credentials using the command line. This ensures database credentials won’t be stored in the server filesystem.

java -jar alfresco-elasticsearch-reindexing-3.0.0-app.jar \
--alfresco.reindex.jobName=reindexByIds \ \
--spring.datasource.url=jdbc:postgresql://localhost:5432/alfresco \
--spring.datasource.username=alfresco \
--spring.datasource.password=alfresco \
--alfresco.acceptedContentMediaTypesCache.baseurl=http://localhost:8090/transform/config \

When completed successfully you will see:

o.s.batch.core.step.AbstractStep         : Step: [reindexByIdsStep] executed in 4s952ms
o.a.r.w.ElasticsearchRepoEventItemWriter : Total indexed documents:: 845
o.a.r.listeners.JobLifecycleListener     : Current Status: COMPLETED

Once the program has been executed, the existing Alfresco Repository nodes are available in Search Enterprise.

Note: Additional use cases are be covered in the Administer documentation.

Alfresco Live Indexing app

The Elasticsearch connector Live Indexing app can be started from the command line as a standard Spring Boot application.

  1. Start the Live Indexing app.
java -jar alfresco-elasticsearch-live-indexing-3.0.0-app.jar

If your services are deployed on a different server or port the following parameters can be used.

java -jar alfresco-elasticsearch-live-indexing-3.0.0-app.jar \ \ \
--alfresco.sharedFileStore.baseUrl=http://localhost:8099/alfresco/api/-default-/private/sfs/versions/1/file/ \
--alfresco.acceptedContentMediaTypesCache.baseurl=http://localhost:8090/transform/config \

If required additional memory may be assigned to these services using the default JVM options. For instance, to start the Elasticsearch connector with 2 GB of RAM.

java -Xmx2G -jar alfresco-elasticsearch-live-indexing-3.0.0-app.jar

By default, the Elasticsearch connector is started using port 8080. This port can be changed using the default Spring Boot command line parameter server.port. For instance, to start the Elasticsearch Connector using port 8083.

java -jar alfresco-elasticsearch-live-indexing-3.0.0-app.jar --server.port=8083

Once all services are up and running the Elasticsearch index will be populated and search queries will work as expected when using supported Alfresco applications such as Alfresco Digital Workspace.

Install using Docker Compose

Use this information to quickly start up the Elasticsearch connector using Docker compose. Due to the limited capabilities of Docker compose, this deployment method is only recommended for development and test environments. You can perform the Docker compose deployment using the source code or downloading the distribution zip file. Both of these methods produce the same docker-compose.yaml file needed when deploying Content Services.

Note: The Docker compose file provided is only for test and development purposes.


  • Docker
    • This allows you to run Docker images and Docker compose on a single computer.
  • Docker compose
    • Docker compose is included as part of some Docker installers. If it’s not part of your installation, then install it separately after you’ve installed Docker.

    Note: The Elasticsearch connector Docker images from are only for Enterprise customers. You need credentials to be able to pull these images from Alfresco customers can request their credentials by logging a ticket at Alfresco Support.

Using source code

Create the Docker compose file using the source code.

  1. Retrieve the Elasticsearch connector source code with a Git client using SSH:

     git clone
  2. Move to the folder where you cloned the project and build it using Maven:

     cd alfresco-elasticsearch-connector
     mvn clean install -DskipTests

    The Docker compose file is created in the alfresco-elasticsearch-connector-distribution/src/main/resources/docker-compose folder.

  3. Move to the Docker compose folder in the distribution module:

     cd /alfresco-elasticsearch-connector-distribution/src/main/resources/docker-compose

Using distribution zip

Create the Docker compose file using the distribution zip file.

  1. Download the file from Hyland Community.

  2. Unzip the distribution zip file into a folder:

     unzip alfresco-elasticsearch-connector-distribution-*.zip -d alfresco-elasticsearch-connector-distribution
  3. Move to the Docker compose folder in the distribution folder:

     cd alfresco-elasticsearch-connector-distribution/docker-compose

Deployment steps

Deploy the Docker compose file you created.

  1. Log in to using your credentials:

     $ docker login
  2. Deploy Alfresco Content Services. The docker-compose.yml you generated includes the Repository, ADW, Postgres, Transform Service, ActiveMQ, Alfresco Elasticsearch Connector, Elasticsearch, and Kibana.

     $ docker-compose up --build --force-recreate

The command downloads the images and fetches all the dependencies, then creates each container, and starts the system.

Wait for the logs to show the following message:

alfresco_1 | 05-Sep-2021 13:36:37.893 INFO [main] org.apache.catalina.startup.Catalina.start Server startup in 148870 ms

If you encounter errors whilst the system is starting up:

  • Stop the session (by using CONTROL+C).
  • Try allocating more memory resources, as advised in docker-compose.yml. For example, in Docker, change the memory setting in Preferences (or Settings) > Advanced > Memory, to at least 16 GB.
  • Make sure you restart Docker and wait for the process to finish before continuing.

Open your browser and check everything starts up correctly:

  • Administration and REST APIs http://localhost:8080/alfresco
  • Alfresco Digital Workspace (UI) http://localhost:8080/workspace
  • ActiveMQ Admin Web Console http://localhost:8161/admin
  • Elasticsearch server http://localhost:9200
  • Kibana http://localhost:5601

Log in as the administrator with the default username and password.

Note: Remember to run the Alfresco Re-indexing app as described above in order to add existing Alfresco Repository nodes to the Elasticsearch server.

Alternative deployment

By default, the Docker compose template deploys the Elasticsearch connector services individually:

  • live-indexing-mediation the service manages ActiveMQ messages from Alfresco Repository and Alfresco Transform Service
  • live-indexing-content the service indexes content in Search Enterprise
  • live-indexing-metadata the service indexes metadata in Search Enterprise

The docker-compose.yml file you generated includes:

        SPRING_ELASTICSEARCH_REST_URIS: http://elasticsearch:9200
        SPRING_ACTIVEMQ_BROKERURL: nio://activemq:61616

        SPRING_ELASTICSEARCH_REST_URIS: http://elasticsearch:9200
        SPRING_ACTIVEMQ_BROKERURL: nio://activemq:61616
        ALFRESCO_SHAREDFILESTORE_BASEURL: http://shared-file-store:8099/alfresco/api/-default-/private/sfs/versions/1/file/

        SPRING_ELASTICSEARCH_REST_URIS: http://elasticsearch:9200
        SPRING_ACTIVEMQ_BROKERURL: nio://activemq:61616

Alternatively, you can use the all-in-one Docker image for the Elasticsearch connector named alfresco-elasticsearch-live-indexing that includes every service:

        SPRING_ELASTICSEARCH_REST_URIS: http://elasticsearch:9200
        SPRING_ACTIVEMQ_BROKERURL: nio://activemq:61616
        ALFRESCO_SHAREDFILESTORE_BASEURL: http://shared-file-store:8099/alfresco/api/-default-/private/sfs/versions/1/file/

Note: If the Elasticsearch server is available on your environment, elasticsearch and kibana services can be removed from the docker-compose.yml file. You can adjust the references to the elasticsearch service in your Docker compose file to use your Elasticsearch deployment.

Install using Helm

Use this information to install the the Elasticsearch connector using Helm. The deployment of the Content Services stack for Kubernetes using Helm is available at Alfresco Content Services Containerized Deployment.

Depending on where you want to install Content Services you must follow the appropriate instructions for the Kubernetes cluster, for more see Docker Desktop or AWS EKS.

To replace Search Services with the Elasticsearch Connector you must configure the requirements.yaml file and set the alfresco-elasticsearch-connector.enabled property to true and alfresco-search.enabled to false.

The Elasticsearch Connector will start four new Kubernetes deployments for live indexing:

  • Mediation: must be always a single node. It orchestrates events from Alfresco Repository.
  • Metadata: is responsible for indexing node metadata.
  • Content: indexes content.
  • Path: indexes the path of a node.

Additionally, a Kubernetes job will be started to reindex existing content in Search Enterprise. It is recommended you only run this job at the initial startup. You can enable or disable the setting in the alfresco-elasticsearch-connector.reindexing.enabled property file by using true or false.

To deploy Content Services with the Elasticsearch connector:

 helm install acs alfresco/alfresco-content-services \
 --values esc_values.yaml \
 --set externalPort="80" \
 --set externalProtocol="http" \
 --set externalHost="localhost" \
 --set global.alfrescoRegistryPullSecrets=my-registry-secrets \
 --set repository.replicaCount=1 \
 --set transformrouter.replicaCount=1 \
 --set pdfrenderer.replicaCount=1 \
 --set imagemagick.replicaCount=1 \
 --set libreoffice.replicaCount=1 \
 --set tika.replicaCount=1 \
 --set transformmisc.replicaCount=1 \
 --set postgresql-syncservice.resources.requests.memory="500Mi" \
 --set postgresql-syncservice.resources.limits.memory="500Mi" \
 --set postgresql.resources.requests.memory="500Mi" \
 --set postgresql.resources.limits.memory="500Mi" \
 --set alfresco-search.resources.requests.memory="1000Mi" \
 --set alfresco-search.resources.limits.memory="1000Mi" \
 --set share.resources.limits.memory="1500Mi" \
 --set share.resources.requests.memory="1500Mi" \
 --set repository.resources.limits.memory="2500Mi" \
 --set repository.resources.requests.memory="2500Mi"\
 --timeout 10m0s \

If you are using Docker Desktop locally, you must set antiAffinity to soft and it is recommended you reduce the Elasticsearch server resources:

   enabled: true
   antiAffinity: "soft"

   # Shrink default JVM heap.
   esJavaOpts: "-Xmx128m -Xms128m"

   # Allocate smaller chunks of memory per pod.
       cpu: "100m"
       memory: "512M"
       cpu: "1000m"
       memory: "512M"

   # Request smaller persistent volumes.
     accessModes: [ "ReadWriteOnce" ]
     storageClassName: "hostpath"
         storage: 100M

When the system is up and running, you can access the Kibana console using port forwarding:

 kubectl port-forward service/acs-kibana 5601:5601 -n alfresco

and then you can access the console http://localhost:5601/app/kibana.

If you need access to the Elasticsearch server directly you have to perform the same operation:

 kubectl port-forward service/elasticsearch-master 9200:9200 -n alfresco

and then you can access the server http://localhost:9200/.

More properties that can be used to configure the chart are available here.

Install Elasticsearch server

The Elasticsearch connector uses a standard Elasticsearch 7.10.2 server. No additional plugin is required.

Other alternatives may be selected for your Elasticsearch installation, for more see Installing Elastic Search. Alternatively, a managed service from Elasticsearch or Amazon AWS can be used.

Both Alfresco Repository and the Elasticsearch connector support communication with the Elasticsearch server using HTTP or HTTPs protocol with or without HTTP Basic Authentication.

Note: The Elasticsearch server does not require any additional software from Alfresco in order to be used by Alfresco Search Enterprise 3.0.

Edit this page

Suggest an edit on GitHub
This website uses cookies in order to offer you the most relevant information. Please accept cookies for optimal performance. This documentation is subject to the Alfresco documentation terms.