debezium and confluent giving LEADER_NOT_AVAILABLE after confluent load

[igue...@rocketlawyer.com]

Hi,

got this error after debezium connector load in confluent,

WARN [Producer clientId=producer-4] Error while fetching metadata with correlation id 1 : {test_server.ig_test.test_dbz=LEADER_NOT_AVAILABLE}

I believe that I've set all host.names properties to localhost

does anybody have an idea what im missing in configuration ?

[Jiri Pechanec]

Could you please check your advertised hostname?

J.

[igue...@rocketlawyer.com]

here extraction from kafka log

        advertised.host.name = localhost
        advertised.listeners = PLAINTEXT://localhost:9092
        advertised.port = 9092
        alter.config.policy.class.name = null
        authorizer.class.name =
        auto.create.topics.enable = true
        auto.leader.rebalance.enable = true
        background.threads = 10
        broker.id = 0
        broker.id.generation.enable = true
        broker.interceptor.class = class org.apache.kafka.server.interceptor.DefaultBrokerInterceptor
        broker.rack = null
        compression.type = producer
        connections.max.idle.ms = 600000
        controlled.shutdown.enable = true
        controlled.shutdown.max.retries = 3
        controlled.shutdown.retry.backoff.ms = 5000
        controller.socket.timeout.ms = 30000
        create.topic.policy.class.name = null
        default.replication.factor = 1
        delete.records.purgatory.purge.interval.requests = 1
        delete.topic.enable = true
        fetch.purgatory.purge.interval.requests = 1000
        group.initial.rebalance.delay.ms = 0
        group.max.session.timeout.ms = 300000
        group.min.session.timeout.ms = 6000
        host.name =
        inter.broker.listener.name = null
        inter.broker.protocol.version = 1.0-IV0
        leader.imbalance.check.interval.seconds = 300
        leader.imbalance.per.broker.percentage = 10
        listener.security.protocol.map = PLAINTEXT:PLAINTEXT,SSL:SSL,SASL_PLAINTEXT:SASL_PLAINTEXT,SASL_SSL:SASL_SSL
        listeners = PLAINTEXT://localhost:9092
        log.cleaner.backoff.ms = 15000
        log.cleaner.dedupe.buffer.size = 134217728
        log.cleaner.delete.retention.ms = 86400000
        log.cleaner.enable = true
        log.cleaner.io.buffer.load.factor = 0.9
        log.cleaner.io.buffer.size = 524288
        log.cleaner.io.max.bytes.per.second = 1.7976931348623157E308
        log.cleaner.min.cleanable.ratio = 0.5
        log.cleaner.min.compaction.lag.ms = 0
        log.cleaner.threads = 1
        log.cleanup.policy = [delete]

[Jiri Pechanec]

Do you use a single Kafka instance or multiple instances - i.e. cluster? If the latter one then advertised.* parameters must be set to the hostnam/IP on which each instance is running.

J.

[igue...@rocketlawyer.com]

it is a single instance, but connect running in distributed mode,

I am using confluent so the setting is little bit different than canonical kafka 

[Jiri Pechanec]

Let me ask, is it broken for you. I forgot to point you out to - http://debezium.io/docs/tutorial/ - this warnings are a normal nothing to worry about as it is just about to create new topics - look at the document.

J.