BlancoByte

Connector registration failed

Symptom: The pipeline status shows an error immediately after clicking Start.

Check the connector-init logs:

docker logs bb-connector-init

Check the Debezium REST API directly:

curl http://localhost:8083/connectors
curl http://localhost:8083/connectors/pg-postgres-sourcedb/status

Common cause: The source database is not yet healthy when the connector tries to register. Try stopping and restarting the pipeline from the UI after 30 seconds.

No events arriving in ClickHouse

Symptom: The pipeline shows as running but the Monitor page shows 0 events.

Check Kafka topics exist:

docker exec bb-kafka kafka-topics.sh --bootstrap-server localhost:9092 --list

Check the sink consumer is running:

docker logs bb-sink --tail 50

Check the connector is actually running in Debezium:

curl http://localhost:8083/connectors/pg-postgres-sourcedb/status | python3 -m json.tool

Look for "state": "RUNNING". If it shows "state": "FAILED", check docker logs bb-debezium --tail 50.

TypeMismatch error on insert

Symptom: Errors like TypeMismatchError: Column id: required argument is not an integer in the sink logs.

Cause: The column type mapping between the source schema and the ClickHouse table does not match. This can happen if a table was created by a previous pipeline with a different schema.

Fix: Drop the affected ClickHouse table and restart the pipeline. The table will be recreated with the correct schema.

-- Run in ClickHouse (via Query Editor or HTTP API)
DROP TABLE IF EXISTS cdc.your_table;

Then stop and restart the pipeline from the UI.

MariaDB connector not registering

Symptom: MariaDB pipeline fails but PostgreSQL and MySQL work fine.

Check MariaDB is healthy:

docker logs bb-mariadb --tail 20

Check the connector-init output:

docker logs bb-connector-init | grep -i mariadb

Verify binlog is enabled:

docker exec bb-mariadb mariadb -u root -prootpass -e "SHOW VARIABLES LIKE 'log_bin';"

Full reset

If anything is in a broken state, a full reset clears all data and starts fresh.

docker-compose down -v
docker-compose up -d

This removes all Kafka topics, ClickHouse data, and pipeline state. The demo databases will be re-seeded automatically.