Skip to content


Bootstrap a default version of Amundsen using Docker

The following instructions are for setting up a version of Amundsen using Docker.

  1. Make sure you have at least 3GB of disk space available to Docker. Install docker and docker-compose.
  2. Clone this repo and its submodules by running:
    $ git clone --recursive
  3. Enter the cloned directory and run the command below:
    # For Neo4j Backend
    $ docker-compose -f docker-amundsen.yml up
    # For Atlas
    $ docker-compose -f docker-amundsen-atlas.yml up
    If it’s your first time, you may want to proactively go through troubleshooting steps, especially the first one related to heap memory for ElasticSearch and Docker engine memory allocation (leading to Docker error 137).
  4. Ingest provided sample data into Neo4j by doing the following: (Please skip if you are using Atlas backend)

  5. In a separate terminal window, change directory to databuilder.

  6. sample_data_loader python script included in examples/ directory uses elasticsearch client, pyhocon and other libraries. Install the dependencies in a virtual env and run the script by following the commands below. See Windows Troubleshooting if you encounter an error on python3 install regarding extas_require on windows.
     $ python3 -m venv venv
     $ source venv/bin/activate
     $ pip3 install --upgrade pip
     $ pip3 install -r requirements.txt
     $ python3 install
     $ python3 example/scripts/
  7. View UI at http://localhost:5000 and try to search test, it should return some result.

  8. We could also perform an exact-match search for the table entity. For example: search test_table1 in table field and it’ll return the records that matched.

Atlas Note: Atlas takes some time to boot properly. So you may not be able to see the results immediately after you run the docker-compose up command. Atlas would be ready once you’ll have the following output in the docker output Amundsen Entity Definitions Created...

Verify your setup

  1. You can verify the dummy data has been ingested into Neo4j by by visiting http://localhost:7474/browser/ and run MATCH (n:Table) RETURN n LIMIT 25 in the query box. You should see few tables.
  2. You can verify the data has been loaded into the metadataservice by visiting:
  3. http://localhost:5000/table_detail/gold/hive/test_schema/test_table1
  4. http://localhost:5000/table_detail/gold/dynamo/test_schema/test_table2


  1. If the docker container doesn’t have enough heap memory for Elastic Search, es_amundsen will fail during docker-compose.
  2. docker-compose error: es_amundsen | [1]: max virtual memory areas vm.max_map_count [65530] is too low, increase to at least [262144]
  3. Increase the heap memory detailed instructions here

    1. Edit /etc/sysctl.conf
    2. Make entry vm.max_map_count=262144. Save and exit.
    3. Reload settings $ sysctl -p
    4. Restart docker-compose
  4. If docker-amundsen-local.yml stops because of org.elasticsearch.bootstrap.StartupException: java.lang.IllegalStateException: Failed to create node environment, then es_amundsen cannot write to .local/elasticsearch.

  5. chown -R 1000:1000 .local/elasticsearch
  6. Restart docker-compose
  7. If when running the sample data loader you recieve a connection error related to ElasticSearch or like this for Neo4j:
        Traceback (most recent call last):
          File "/home/ubuntu/amundsen/amundsendatabuilder/venv/lib/python3.6/site-packages/neobolt/", line 831, in _connect
        ConnectionRefusedError: [Errno 111] Connection refused
  8. If elastic search container stops with an error max file descriptors [4096] for elasticsearch process is too low, increase to at least [65535], then add the below code to the file docker-amundsen-local.yml in the elasticsearch definition. ulimits: nofile: soft: 65535 hard: 65535 Then check if all 5 Amundsen related containers are running with docker ps? Can you connect to the Neo4j UI at http://localhost:7474/browser/ and similarly the raw ES API at http://localhost:9200? Does Docker logs reveal any serious issues?

  9. If ES container crashed with Docker error 137 on the first call from the website (http://localhost:5000/), this is because you are using the default Docker engine memory allocation of 2GB. The minimum needed for all the containers to run with the loaded sample data is 3GB. To do this go to your Docker -> Preferences -> Resources -> Advanced and increase the Memory, then restart the Docker engine.

  10. Windows Troubleshooting