MSSQL doesn't work with incremeting or timestamp source connector - sql-server

Question

I'm trying to create source connector for my SQL Server tables, here's an example:
{
"name": "test",
"config": {
"tasks.max":"1",
"connector.class": "io.confluent.connect.jdbc.JdbcSourceConnector",
"connection.url": "jdbc:sqlserver://REDACTED;databaseName=REDACTED",
"connection.user": "REDACTED",
"connection.password": "REDACTED",
"topic.prefix": "test",
"poll.interval.ms" : 2000,
"table.whitelist": "dbo.Client",
"mode":"incrementing",
"incrementing.column.name": "Id"
}
}
I'm getting following error, which doesn't say anything...
java.lang.IllegalArgumentException: Number of groups must be positive.
at org.apache.kafka.connect.util.ConnectorUtils.groupPartitions(ConnectorUtils.java:41)
at io.confluent.connect.jdbc.JdbcSourceConnector.taskConfigs(JdbcSourceConnector.java:148)
at org.apache.kafka.connect.runtime.Worker.connectorTaskConfigs(Worker.java:305)
at org.apache.kafka.connect.runtime.distributed.DistributedHerder.reconfigureConnector(DistributedHerder.java:997)
at org.apache.kafka.connect.runtime.distributed.DistributedHerder.reconfigureConnectorTasksWithRetry(DistributedHerder.java:950)
at org.apache.kafka.connect.runtime.distributed.DistributedHerder.access$900(DistributedHerder.java:110)
at org.apache.kafka.connect.runtime.distributed.DistributedHerder$17$1.call(DistributedHerder.java:963)
at org.apache.kafka.connect.runtime.distributed.DistributedHerder$17$1.call(DistributedHerder.java:960)
at org.apache.kafka.connect.runtime.distributed.DistributedHerder.tick(DistributedHerder.java:270)
at org.apache.kafka.connect.runtime.distributed.DistributedHerder.run(DistributedHerder.java:219)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
The same happens when I use mode timestamp with timestamp.column.name (column Modified with type datetimeoffset).
When I use mode bulk with a custom query, it works fine.
Any ideas?