You can start the materialized
binary with:
./materialized_(linux|mac) --dev --w=1
And then launch into the SQL interface (e.g. psql -h 0.0.0.0 -p 6875 -d materialize
)
And here is the part I'm having trouble testing: actually connecting to a Kerberized Kafka cluster:
CREATE SOURCE quotes
FROM KAFKA BROKER '<broker host>' TOPIC '<topic>'
WITH (
security_protocol = 'sasl_plaintext',
sasl_kerberos_keytab = '<path to keytab>',
sasl_kerberos_principal = '<principal val>',
sasl_kerberos_kinit_cmd = '<kinit cmd>',
sasl_kerberos_min_time_before_relogin = '<val>'
)
FORMAT AVRO
USING CONFLUENT SCHEMA REGISTRY '<schema registry>';
I assume you're using a Schema Registry; if not, you could just change to FORMAT BYTES
since all I'm trying to see is if this will connect to the Kafka cluster.
You can also specify sasl_mechanisms
and sasl_kerberos_service_name
but I believe those properly default through rdkafka
; if not here's another statement to try:
CREATE SOURCE quotes
FROM KAFKA BROKER '<broker host>' TOPIC '<topic>'
WITH (
security_protocol = 'sasl_plaintext',
sasl_mechanisms = 'GSSAPI',
sasl_kerberos_service_name = 'kafka',
sasl_kerberos_keytab = '<path_to_keytab>',
sasl_kerberos_principal = '<principal val>',
sasl_kerberos_kinit_cmd = '<kinit cmd>',
sasl_kerberos_min_time_before_relogin = '<val>'
)
FORMAT AVRO USING CONFLUENT SCHEMA REGISTRY '<schema registry>';