Made a Tool to Streams Changes from Microsoft SQL Server to Apache Kafka

Hacker News Top Tools

Summary

Athena is a lightweight Change Data Capture (CDC) tool that streams changes from Microsoft SQL Server to Apache Kafka. Built in Golang, it offers simpler setup than Debezium and automatically manages CDC configurations.

No content available
Original Article
View Cached Full Text

Cached at: 06/02/26, 09:34 PM

Niyko/Athena

Source: https://github.com/Niyko/Athena

Athena logo Athena is a lightweight Change Data Capture (CDC) solution that streams changes from Microsoft SQL Server to Apache Kafka. Built in Golang, it supports SASL-authenticated Kafka brokers and provides a straightforward setup experience. Unlike alternatives such as Debezium, which can be complex to configure and manage. Athena offers greater simplicity and operational ease. It automatically manages CDC setups, publishes database changes to a single Kafka topic, and delivers a clean, intuitive event format that is easy for downstream consumers to understand and process.

:zap: How things work

  • Creates a message for changes like create, update, delete for rows in MSSQL database tables to a single Kafka topic.
  • Athena only creates messages for all new table changes. Existing ones are ignored.
  • Kafka topic have to be created before hand. Unlike Debezium, Athena will not create the topic own it’s own.
  • All CDC setups in MSSQL is automatically done by Athena when setup command is run.
  • By default, Athena will poll for changes for all tables, you can use the skippedTables option in the config.json to ignore any tables.

:cyclone: Simple Installation

You can download the pre-compiled binaries from the Github releases page and copy them to the desired location. After that you can follow the below steps in order.

Create a config.json file in the root folder where you but the Athena binary. Here is the format of the JSON file. Fill all the credentials also.

You can find more details about the paramters in config file in below sections.

{
    "dbHost": "127.0.0.1",
    "dbPort": 1433,
    "dbUser": "",
    "dbPassword": "",
    "dbName": "",

    "kafkaHost": "",
    "kafkaEnableTLS": false,
    "kafkaTopic": "",

    "kafkaSASLMechanisms": "NONE",
    "kafkaSASLUsername": "",
    "kafkaSASLPassword": "",

    "pollInterval": 10,
    "fetchLimit": 50,
    "skippedTables": [],

    // If you want to collect logs in clickhouse
    "clickHouse": false,
    "clickHouseHost": "<host>:<port>",
    "clickHouseUsername": "",
    "clickHousePassword": "",
    "clickHouseDatabase": "",
    "clickHouseTableName": "",
    "clickHouseTableTTL": 12
}

Create topic with name given in config.json without scheme and with required partitions in you Kafka broker.

Run the setup command in order to create the CDC in database and other required setup.

./athena setup

Setup a service for running Athena in the background. Setting this up will different for Windows and Linux. Below given are the steps to create them on a Linux distro.

Create a service file called athena_mssql_kafka.service in the directory /etc/systemd/system using the following commands.

cd /etc/systemd/system
nano athena_mssql_kafka.service

Copy and paste the below contents to the above created service file athena_mssql_kafka.service.

[Unit]
Description=Athena MSSQL Kafka Service
After=network.target

[Service]
Type=simple
ExecStart=athena run

[Install]
WantedBy=multi-user.target

Please note that path in ExecStart needs to change while creating the service file.

Now you can start the service and also check the status of the service.

systemctl start athena_mmsql_kafka.service
systemctl status athena_mmsql_kafka.service

:gear: Configuring Athena

Athena can be configured using the config.json file created on the root the Athena binary. Here are the details of the configuration keys and what they do in table format.

OptionDescriptionExample
dbHostDatabase host of MSSQL127.0.0.1
dbPortDatabase port of MSSQL1433
dbUserUsername for the MSSQL database
dbPasswordPassword for the MSSQL database
dbNameDatabase name of MSSQL
kafkaHostHost with port for the Kafka server127.0.0.1:9092
kafkaTopicKafka topic that you created for table changes to show
kafkaEnableTLSEnables TLS for Kafka connectiontrue, false
kafkaSASLMechanismsSASL mechanism that need to be used for Kafka connectionNONE, SASL-PLAIN, SASL-SCRAM-SHA-256, SASL-SCRAM-SHA-512
kafkaSASLUsernameSASL user name of the Kafka server
kafkaSASLPasswordSASL password of the Kafka server
pollIntervalInterval where next polling to the database is made. It’s given in seconds format.10
fetchLimitNumber of CDC changes rows that will be pulled from the table at once.50
skippedTablesArray of tables that needs to skipped while taking CDC changes.[“table1”, “table2”]
clickHouseEnable Clickhouse logs. Table and struture for Clickhouse is automatically created by Athena when setup command is runtrue, false
clickHouseHostHost with port for the Clickhouse server127.0.0.1:8123
clickHouseUsernameUser name of Clickhouse server
clickHousePasswordPassword of Clickhouse server
clickHouseDatabaseDasebase name of Clickhouse server
clickHouseTableNameTable name of Clickhouse server
clickHouseTableTTLTime to live for each record in hours24

:mushroom: Helper options in Athena

Athena executable have some other helper functions apart from setup or run which are explained below. These can be run like eg: ./athena uninstall

OptionDescription
uninstallWill disable CDC in MSSQL database and remove the SQlite database
add-cdcWill run CDC setup in the MSSQL database
remove-cdcWill disable CDC in MSSQL database
clear-cdc-historyClear CDC history or changes that Athena didn’t process yet from the MSSQL database
recreate-clickhouseRerun the Clickhouse migration
recreate-sqliteRecreate the SQlite database and rerun the migration
helpTo view all the options available

:triangular_ruler: Development

For setting up development environment, there is a docker file in the folder dev. It will create all necessary services like MSSQL with sample database, Kafka etc. This same environment can be used for running integration tests.

  • Install latest version of Go from here.
  • Clone that project from Github.
  • Run go mod download command to install all mods.
  • Then run the commands below as needed.
cd dev
docker composer up -d
cd ..
set GORUN=true # Used for identifying if script is run from go run command to choose correct path for reading config.json or db.sqlite
go run . setup
go run . run

Please note that the docker-compose.yml in the dev folder should only be used for development purpose.

:cactus: How to run tests

Before running the tests, make sure you have setup the development environment and also config.json is setup correctly.

cd tests
go test -v -run TestIntegration

:hammer_and_wrench: How to build

You can build the binaries or do development of Athena by following the below steps. Athena is build fully on Golang. So you should install latest version of Go from here. Do note that building binaries are managed with the Goreleaser.

  • Clone that project from Github.
  • Run go mod download command to install all mods.
  • Run SET GORUN=true command to set gorun variable.
  • Run the command goreleaser release --snapshot --clean for building the binaries.

:page_with_curl: License

Athena is licensed under the MIT License.

Similar Articles

SQLite is all you need for durable workflows

Hacker News Top

This blog post argues that SQLite, combined with Litestream for async backups, provides a simple and effective approach to durable execution for many workflow systems, especially AI agents, without needing a separate orchestration tier or network database.

Basedash MCP server

Product Hunt

Basedash is a new MCP server that functions as a data analyst integration for various AI tools.