Kafka sink connector failed

[amotl]

Hi again,

@amotl

@M_Duyen_L_u

Can you suggest me a [working] config file of CrateDB sink ?

That’s a good idea. Let us check our resources if we can find one. It’s an excellent opportunity to improve our documentation if it’s not easy to discover.

@M_Duyen_L_u

Failed to process record: Failed to process a sink record [io.debezium.connector.jdbc.JdbcSinkConnectorTask]

So we’ve figured out you might be using Debezium for the job? Maybe ingredients from the other tutorial at Replicating data to CrateDB with Debezium and Kafka can support your struggles for a better outcome?

I am very sorry I am not a Debezium/Kafka expert, but as far as I can correlate your configuration shared here, in particular that fragment…

@M_Duyen_L_u

    "auto.create": "true",
    "auto.evolve": "true",
    "insert.mode": "upsert",
    "pk.mode": "record_key",
    "pk.fields": "_id",
    "delete.enabled": "true",

with the error message in focus…

Caused by: org.apache.kafka.connect.errors.ConnectException: Configured primary key mode 'record_value' cannot have null schema

and then compare it against this fragment from the other tutorial…

@hernanc

        "auto.create": "false",
        "auto.evolve": "false",

		"insert.mode": "upsert",
		"pk.fields": "id",
		"pk.mode": "record_value",	

… we can see a deviation in the pk.mode configuration to refer to record_key instead of record_value like in the example blueprint? Is it possible that the cause for your problems are in that area of the destination connector configuration?

With kind regards,
Andreas.