Destination MongoDB #

The extracted replicant-cli will be referred to as the $REPLICANT_HOME directory.

I. Set up Connection Configuration #

  1. From $REPLICANT_HOME, navigate to the connection configuration file:

    vi conf/conn/mongodb_dst.yaml
  2. For connecting to the MongoDB server, you can choose between the following methods for an authenticated connection:

    Connect using MongoDB connection string URI #

    To connect to MongoDB using connection strings URI, specify your credentials in plain text in the connection configuration file like the following sample:

    type: MONGODB
    url: "mongodb://localhost:27019/?w=majority"
    max-connections: 30
        url: "mongodb://localhost:27017/?w=majority&replicaSet=mongors1"
        url: "mongodb://localhost:27027/?w=majority&replicaSet=mongors2"

    For multiple replica-sets, specify all of them under replica-sets according to the preceding format.

    Replicant monitors the replica-sets for oplog entries to carry out real-time replication. Each url of a MongoDB replica set must represent the host:port belonging to the replica set. url must contain the option replicaSet=<replicaSet_name> to represent the URL as a replica set.

    You can specify additional connection configurations in the url string according to the MongoDB syntax. For example, you can specify number of connections, Read Concern Options, Write Concern Options, etc. For more information, see Connection String Options.

    Connect using SSL #

    To connect to MongoDB using SSL, specify the SSL connection parameters in the ssl section of the connection configuration file:

        path: 'PATH_TO_KEYSTORE'
        password: 'KEYSTORE_PASSWORD'
        path: 'PATH_TO_TRUST_STORE'
        password: 'TRUSTSTORE_PASSWORD'

    Replcate the following:

    • PATH_TO_KEYSTORE: Path to your KeyStore file.
    • KEYSTORE_PASSWORD: Your KeyStore password.
    • PATH_TO_TRUST_STORE: Path to your TrustStore file.
    • TRUSTSTORE_PASSWORD: Your TrustStore password.

    Kerberos authentication #

    Arcion Replicant supports replication from Kerberized MongoDB clusters. For Kerberos authentication, you have the following options available:

    In this method, you can lay out the connection configuration file in the following manner:

    1. Specify the hostname and port number of your Kerberized MongoDB cluster using the host and port parameters respectively. If you have replica sets, specify them in the format we discuss in the Connect using connection strings URI section. The hostname must be a fully qualified domain name (FQDN) instead of an IP address.

    2. Under kerberos section of the connection configuration file, specify the following:

      User principal name follows this format:


    The following shows a sample connection configuration:

    type: MONGODB
    host: "" 
    port: 27017
    max-connections: 3
        host: ""
        port: 27017
        host: ""
        port: 27017
      kerberosConfigFilePath: /etc/krb5.conf
      user-principal: "replicant%40REPLICANT.IO" 

    In this method, you specify the necessary connection parameters inside the MongoDB connection string URI and set it as the url value in the connection configuration file. This applies to both the Kerberized MongoDB cluster URI and the individual URIs of the replica sets.

    For more information on different components of MongoDB connection string URI, see Connection string URI components.

    The connection string follows this format:


    The following shows a sample configuration using this method:

    type: MONGODB
    url: "mongodb://$external&authMechanism=GSSAPI"
    max-connections: 3
        url: "mongodb://$external&authMechanism=GSSAPI"
        url: "mongodb://$external&authMechanism=GSSAPI"
    max-retries: 1
    retry-wait-duration-ms: 1000

II. Set up Applier Configuration #

  1. From $REPLICANT_HOME, navigate to the Applier configuration file:

    vi conf/dst/mongodb.yaml
  2. The configuration file has two parts:

    • Parameters related to snapshot mode.
    • Parameters related to realtime mode.

    For snapshot mode, make the necessary changes as follows:

      threads: 16
      batch-size-rows: 5000
      txn-size-rows: 5000
    #  map-key-to-id: false
    #  skip-tables-on-failures : false
        enable: false
        type: FILE   # PIPE, FILE
      handle-failed-opers: true
      initIndexesPostSnapshot: true
    #   denormalize:
    #     enable: true
    #  user-role:
    #    init-user-roles: true
    #  init-system-tables: true

    For operating in realtime mode, use the realtime section to specify your configuration. Below is a sample config:

      threads: 8
      batch-size-rows: 1000
      txn-size-rows: 2_0000
      handle-failed-opers: true
    #  map-key-to-id: false
    #  skip-tables-on-failures : false
    #   perTableConfig:
    #   - schema: tpch
    #     tables:
    #       CUSTOMER:
    #         skip-upto-cursor: '{"extractorId":0,"replicaSetName":"mongors1","resumeToken":6868517489379115009,"seqNum":3,"v":1,"timestamp":1599201348000}'
    # Transactional mode config
    # realtime:
    #   threads: 1
    #   batch-size-rows: 1000
    #   replay-consistency: GLOBAL #allowed values are GLOBAL/EVENTUAL
    #   txn-group-count: 100
    #   skip-upto-cursors: ['{"extractorId":0,"replicaSetName":"mongors1","resumeToken":6868517489379115009,"seqNum":3,"v":1,"timestamp":1599201348000}']

For a detailed explanation of configuration parameters in the Applier file, read Applier Reference.