Arcion Self-hosted

Arcion self-hosted quickstart #

I. Host machine requirements #

In order to run Replicant, we recommend that the machine Replicant runs on meet a set of minimum requirements. Failing to meet these requirements might impact performance or might stop Replicant from being able to run at all.

Please ensure your host machine or server where Replicant runs on meets the following minimum hardware and software requirements.

Minimum hardware requirements #

  • 16 CPU cores and 32GB of memory
  • 100GB-200GB of SSD/NVMe storage space

Minimum system requirements #

  • Linux (CentOS/Ubuntu/Redhat)
  • JDK 8 (JRE or JDK installation)
Note: The storage requirements may increase depending on your source-target pair and method of replication. For more information, check out the documentation for each specific source and target you need to use or reach out to our support team.

II. Download Replicant and create REPLICANT_HOME #

Replicant is available through a ZIP file that you must download to the machine or server where you want to host Replicant. If you do not already have the ZIP file and/or an Arcion license, contact our team to get access.

Once you have access to the ZIP file, download it to the directory of your choice on the host machine or server and unzip it:

unzip replicant-cli-<version>.zip

Unzipping the archive creates a directory replicant-cli. We refer this directory as $REPLICANT_HOME throughout the documentation and Arcion website.

III. Adding a license #

Before running Replicant, you need to add a license file to the REPLICANT_HOME directory. You must name the license file replicant.lic so that Replicant can detect the license when Replicant starts.

If you require a license file, contact our team to get one. We will happily provide users with a 30-day free trial license to try out the product or run a POC.

To add the license file, follow these steps:

  1. Obtain an Arcion license file.
  2. Rename the license file to replicant.lic.
  3. Move or copy the replicant.lic file into the REPLICANT_HOME directory. You must copy the replicant.lic file into the root directory, not in the licenses folder of Replicant.

IV. Set up a source database #

After getting Replicant downloaded and adding the license, you can create your source connector. The source connector will be the database where data will be migrated or replicated from.

For configuring a source connector, a few configuration files are required. These files include:

  • connection configuration file
  • filter configuration file

To set up a source connector, you’ll need to do the following:

  1. From REPLICANT_HOME navigate to the sample connection configuration file of the source database:

    vi conf/conn/<database-name>.yaml

    Make the necessary changes as shown below:

    type: <database-name>

host: <host> #Enter the hostname or IP address to connect to the source
             #database
port: <port> #Enter the port on which the source database server is running

username: <database-username> #Enter the source database username
password: <user-password>  #Enter the source database password

max-connections: 30

max-retries: 10
retry-wait-duration-ms: 1000

    Please note that certain databases might have additional configuration parameters. For example, Oracle has the additional parameter service-name.

    For further database specific examples, please refer to one of our Source database setup pages.

  2. From REPLICANT_HOME, navigate to the sample filter file of the source database:

    vi filter/<source-database-name>_filter.yaml

    Make the necessary changes as shown below:

    allow:
- catalog: "catalog_name" #Enter your database name if applicable
  schema : "schema_name" #Enter your schema name if applicable
  types: [TABLE,VIEW] #Database object types to replicate
  allow:
    "Table_Name1": #Enter the name of your table
    "Table_Name2": #Enter the name of your table

V. Set up target database #

  1. From REPLICANT_HOME navigate to the sample connection configuration file of the target database:

    vi conf/conn/<database-name>.yaml

    Make the necessary changes as shown below:

    type: <database-name>

host: <host> #Enter the hostname or IP address to connect to the target
             #database
port: <port> #Enter the port on which the target database server is running

username: <database-username> #Enter the target database username
password: <user-password>  #Enter the target database password

max-connections: 30

max-retries: 10
retry-wait-duration-ms: 1000

    For database specific examples, please refer to one of our Target database setup pages.

VI. Run Replicant in snapshot mode #

Replicant is now ready to run in snapshot mode. The snapshot will only transfer existing data from the source database to the target database. If you would like to transfer real-time changes in addition to the snapshot, skip step VI and proceed to steps VII and VIII to run Replicant in full mode.

  1. Execute the following command from REPLICANT_HOME to run Replicant in snapshot mode:

    ./bin/replicant snapshot conf/conn/<source-database-name>.yaml \
conf/conn/<destination-database-name>.yaml \
--extractor conf/src/<source-database-name>.yaml \
--applier conf/dst/<destination-database-name>.yaml \
--filter filter/<source-database-name>_filter.yaml

The proceeding steps are only required if you intend to run Replicant in real-time mode.

VII. Set up heartbeat table #

  1. Create a heartbeat table in the catalog/schema you are going to replicate with the following DDL:

    CREATE TABLE <catalog>.<schema>.replicate_io_cdc_heartbeat( \
timestamp <data_type_equivalent_to_long>)

  2. Grant INSERT, UPDATE, and DELETE privileges to the user configured for Replicant.

  3. From REPLICANT_HOME, navigate to the heartbeat table’s configuration.

    vi conf/src/<source-database-name>.yaml

  4. Under the Realtime Section, make the necessary changes as follows

    heartbeat:
  enable: true
  catalog: <catalog_name> #if the source database supports catalog, change the catalogue name accordingly
  schema: <schema_name> #if the source database supports schema, change the schema name accordingly
  interval-ms: 10000

VIII. Run Replicant in full mode #

  1. From REPLICANT_HOME, enter the following to run Replicant in full mode:

    ./bin/replicant full conf/conn/<source-database-name>.yaml \
conf/conn/<destination-database-name>.yaml \
--extractor conf/src/<source-database-name>.yaml \
--applier conf/dst/<destination-database-name>.yaml \
--filter filter/<source-database-name>_filter.yaml \

Upgrade Arcion Replicant #

We recommend that you always use the latest version of Replicant. That way, you can enjoy the latest features and various quality improvmements.

You can grab the latest version of Replicant from our Arcion Self-hosted page.

To get the most out of your upgrades, follow the tips below:

  • Keep the configuration files isolated from the replicant-cli directory tree. During version upgrades, they might get overwritten.
  • After upgrade, copy your replicant.lic license file to the new REPLICANT_HOME.
  • Copy the contents of your older $REPLICANT_HOME/lib directory to the new version lib directory $REPLICANT_HOME/lib.

Database Specific Setup Overview #

Different source and target databases might have slightly more specific and different setup instructions than the general guidelines provided in this quickstart guide. Follow these six steps for a pipeline-specific setup for Replicant:

  1. Setting up the source database.
  2. Setting up the target database.
  3. Running Replicant.
  4. Performance enhancing and troubleshoot.