I am using nifi to build a dataflow with the following setup:
apache nifi 1.14.1
kafka 2.13-2.7.1
confluent schema registry
I am also using the processor ConsumeKafkaRecord_2_6 to process messages from a topic where the key and the value where both serialized using avro - schemas for the key and value are stored in the confluent schema registry. But the processor fails to parse the message because there is not a way - that I can see - to specify that both key and value are avro serialized with schemas stored in the confluent schema registry. The convention for naming the schema is usually [topic name]-value and [topic name]-key. I can read the messages just fine using kcat, formerly kafkacat using:
kcat -b broker1:9092,broker2:9092,broker3:9092 -t mytopic -s avro -r http://schema-registry_url.com -p 0
Is there a way to read such messages or am I supposed to add my own processor to nifi? Here's a trace of the error:
causes: org.apache.nifi.serialization.MalformedRecordException: Error while getting next record. Root cause: org.apache.avro.AvroRuntimeException: Malformed data. Length is negative: negative 62
org.apache.nifi.serialization.MalformedRecordException: Error while getting next record. Root cause: org.apache.avro.AvroRuntimeException: Malformed data. Length is negative: negative 62
at org.apache.nifi.avro.AvroRecordReader.nextRecord(AvroRecordReader.java:52)
at org.apache.nifi.serialization.RecordReader.nextRecord(RecordReader.java:50)
at sun.reflect.GeneratedMethodAccessor559.invoke(Unknown Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)
at org.apache.nifi.controller.service.StandardControllerServiceInvocationHandler.invoke(StandardControllerServiceInvocationHandler.java:254)
at org.apache.nifi.controller.service.StandardControllerServiceInvocationHandler.access$100(StandardControllerServiceInvocationHandler.java:38)
at org.apache.nifi.controller.service.StandardControllerServiceInvocationHandler$ProxiedReturnObjectInvocationHandler.invoke(StandardControllerServiceInvocationHandler.java:240)
at com.sun.proxy.$Proxy192.nextRecord(Unknown Source)
at org.apache.nifi.processors.kafka.pubsub.ConsumerLease.writeRecordData(ConsumerLease.java:549)
at org.apache.nifi.processors.kafka.pubsub.ConsumerLease.lambda$processRecords$3(ConsumerLease.java:342)
at java.util.HashMap$KeySpliterator.forEachRemaining(HashMap.java:1556)
at java.util.stream.ReferencePipeline$Head.forEach(ReferencePipeline.java:647)
at org.apache.nifi.processors.kafka.pubsub.ConsumerLease.processRecords(ConsumerLease.java:329)
at org.apache.nifi.processors.kafka.pubsub.ConsumerLease.poll(ConsumerLease.java:188)
at org.apache.nifi.processors.kafka.pubsub.ConsumeKafkaRecord_2_6.onTrigger(ConsumeKafkaRecord_2_6.java:472)
at org.apache.nifi.processor.AbstractProcessor.onTrigger(AbstractProcessor.java:27)
at org.apache.nifi.controller.StandardProcessorNode.onTrigger(StandardProcessorNode.java:1202)
at org.apache.nifi.controller.tasks.ConnectableTask.invoke(ConnectableTask.java:214)
at org.apache.nifi.controller.scheduling.QuartzSchedulingAgent$2.run(QuartzSchedulingAgent.java:137)
at org.apache.nifi.engine.FlowEngine$2.run(FlowEngine.java:110)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180)
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
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)
Caused by: org.apache.avro.AvroRuntimeException: Malformed data. Length is negative: -62
at org.apache.avro.io.BinaryDecoder.doReadBytes(BinaryDecoder.java:336)
at org.apache.avro.io.BinaryDecoder.readString(BinaryDecoder.java:263)
at org.apache.avro.io.ResolvingDecoder.readString(ResolvingDecoder.java:201)
at org.apache.avro.generic.GenericDatumReader.readString(GenericDatumReader.java:430)
at org.apache.nifi.avro.NonCachingDatumReader.readString(NonCachingDatumReader.java:51)
at org.apache.avro.generic.GenericDatumReader.readMapKey(GenericDatumReader.java:335)
at org.apache.avro.generic.GenericDatumReader.readMap(GenericDatumReader.java:321)
at org.apache.avro.generic.GenericDatumReader.readWithoutConversion(GenericDatumReader.java:177)
at org.apache.avro.generic.GenericDatumReader.read(GenericDatumReader.java:152)
at org.apache.avro.generic.GenericDatumReader.readField(GenericDatumReader.java:240)
at org.apache.avro.generic.GenericDatumReader.readRecord(GenericDatumReader.java:230)
at org.apache.avro.generic.GenericDatumReader.readWithoutConversion(GenericDatumReader.java:174)
at org.apache.avro.generic.GenericDatumReader.read(GenericDatumReader.java:152)
at org.apache.avro.generic.GenericDatumReader.read(GenericDatumReader.java:144)
at org.apache.nifi.avro.AvroReaderWithExplicitSchema.nextAvroRecord(AvroReaderWithExplicitSchema.java:92)
at org.apache.nifi.avro.AvroRecordReader.nextRecord(AvroRecordReader.java:39)
... 27 common frames omitted
I am attaching pictures of the processor
ConsumeKafkaRecord 1 of 2
ConsumeKafkaRecord 2 of 2
AvroReader
SchemaRegistry
If the data is already serialized correctly by some Confluent Serializer, you should prefer using the "Confluent Content-Encoded Schema Reference" option in the AvroReader since the Schema ID is embedded within the record and will get the correct subject/version, accordingly.
Otherwise, using the "Schema Name" or "Schema Text" value will either perform a lookup against the registry or use a literal, however, the deserializer will still expect a certain content-length of the record bytes, and seems to be the cause of the issue Malformed data. Length is negative ...
Related
I am using confluent schema registry and having production issues because of unrecoverable kafka messages. The reason seems to be that the schema that was used by the message producer cannot be found. here is the stack trace:
2021-04-02 13:18:04.947+02:00 ERROR org.apache.spark.executor.Executor:91 - Exception in task 0.3 in stage 0.0 (TID 3)
org.apache.kafka.common.errors.SerializationException: Error deserializing key/value for partition pos_transaction-0 at offset 889990. If needed, please seek past the record to continue consumption.
Caused by: org.apache.kafka.common.errors.SerializationException: Error retrieving Avro schema for id 281
Caused by: io.confluent.kafka.schemaregistry.client.rest.exceptions.RestClientException: Unexpected character ('<' (code 60)): expected a valid value (number, String, array, object, 'true', 'false' or 'null')
at [Source: (sun.net.www.protocol.http.HttpURLConnection$HttpInputStream); line: 1, column: 2]; error code: 50005
at io.confluent.kafka.schemaregistry.client.rest.RestService.sendHttpRequest(RestService.java:202)
at io.confluent.kafka.schemaregistry.client.rest.RestService.httpRequest(RestService.java:229)
at io.confluent.kafka.schemaregistry.client.rest.RestService.getId(RestService.java:409)
at io.confluent.kafka.schemaregistry.client.rest.RestService.getId(RestService.java:402)
at io.confluent.kafka.schemaregistry.client.CachedSchemaRegistryClient.getSchemaByIdFromRegistry(CachedSchemaRegistryClient.java:119)
at io.confluent.kafka.schemaregistry.client.CachedSchemaRegistryClient.getBySubjectAndId(CachedSchemaRegistryClient.java:192)
at io.confluent.kafka.schemaregistry.client.CachedSchemaRegistryClient.getById(CachedSchemaRegistryClient.java:168)
at io.confluent.kafka.serializers.AbstractKafkaAvroDeserializer.deserialize(AbstractKafkaAvroDeserializer.java:121)
at io.confluent.kafka.serializers.AbstractKafkaAvroDeserializer.deserialize(AbstractKafkaAvroDeserializer.java:93)
at io.confluent.kafka.serializers.KafkaAvroDeserializer.deserialize(KafkaAvroDeserializer.java:55)
at org.apache.kafka.common.serialization.ExtendedDeserializer$Wrapper.deserialize(ExtendedDeserializer.java:65)
at org.apache.kafka.common.serialization.ExtendedDeserializer$Wrapper.deserialize(ExtendedDeserializer.java:55)
at org.apache.kafka.clients.consumer.internals.Fetcher.parseRecord(Fetcher.java:1009)
at org.apache.kafka.clients.consumer.internals.Fetcher.access$3400(Fetcher.java:96)
at org.apache.kafka.clients.consumer.internals.Fetcher$PartitionRecords.fetchRecords(Fetcher.java:1186)
at org.apache.kafka.clients.consumer.internals.Fetcher$PartitionRecords.access$1500(Fetcher.java:1035)
at org.apache.kafka.clients.consumer.internals.Fetcher.fetchRecords(Fetcher.java:544)
at org.apache.kafka.clients.consumer.internals.Fetcher.fetchedRecords(Fetcher.java:505)
at org.apache.kafka.clients.consumer.KafkaConsumer.pollForFetches(KafkaConsumer.java:1259)
at org.apache.kafka.clients.consumer.KafkaConsumer.poll(KafkaConsumer.java:1187)
at org.apache.kafka.clients.consumer.KafkaConsumer.poll(KafkaConsumer.java:1115)
at org.apache.spark.streaming.kafka010.CachedKafkaConsumer.poll(CachedKafkaConsumer.scala:136)
at org.apache.spark.streaming.kafka010.CachedKafkaConsumer.get(CachedKafkaConsumer.scala:68)
at org.apache.spark.streaming.kafka010.KafkaRDDIterator.next(KafkaRDD.scala:271)
at org.apache.spark.streaming.kafka010.KafkaRDDIterator.next(KafkaRDD.scala:231)
at scala.collection.Iterator$$anon$11.next(Iterator.scala:409)
at scala.collection.Iterator$$anon$13.hasNext(Iterator.scala:462)
at scala.collection.Iterator$$anon$11.hasNext(Iterator.scala:408)
at org.apache.spark.shuffle.sort.BypassMergeSortShuffleWriter.write(BypassMergeSortShuffleWriter.java:125)
at org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:96)
at org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:53)
at org.apache.spark.scheduler.Task.run(Task.scala:109)
at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:345)
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)
I am using schema registry 4.1.2.
The producer code uses version confluent kafka-avro-serializer version 4.1.2 and avro 1.9.0
The consumer uses the same.
The avro schema has not been actually modified for a long time.
If I go inside the _schemas-0 topic folder in one the kafka servers I see a log file with schema 281 that is supposedly not found.
My questions
How can I find out more about the source of error ?
Is there any way to recover messages stored in kafka with the elusive schema id ?
Thanks!
I'm trying to make my applications work using the schema registry from confluent but at this point I'm not in total control of the producers, you can even see them as legacy applications that simply are not bound to the confluent products.
I was looking at the confluent information and it seems all the messages should include in the payload a Magic Byte and Schema ID
https://docs.confluent.io/3.2.0/schema-registry/docs/serializer-formatter.html
or else when I try to consume it I get an error:
[2020-09-25 13:12:09,008] ERROR WorkerSinkTask{id=s3_parquet_connector-0} Task threw an uncaught and unrecoverable exception (org.apache.kafka.connect.runtime.WorkerTask)
org.apache.kafka.connect.errors.ConnectException: Tolerance exceeded in error handler
at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execAndHandleError(RetryWithToleranceOperator.java:178)
at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execute(RetryWithToleranceOperator.java:104)
at org.apache.kafka.connect.runtime.WorkerSinkTask.convertAndTransformRecord(WorkerSinkTask.java:491)
at org.apache.kafka.connect.runtime.WorkerSinkTask.convertMessages(WorkerSinkTask.java:468)
at org.apache.kafka.connect.runtime.WorkerSinkTask.poll(WorkerSinkTask.java:324)
at org.apache.kafka.connect.runtime.WorkerSinkTask.iteration(WorkerSinkTask.java:228)
at org.apache.kafka.connect.runtime.WorkerSinkTask.execute(WorkerSinkTask.java:200)
at org.apache.kafka.connect.runtime.WorkerTask.doRun(WorkerTask.java:184)
at org.apache.kafka.connect.runtime.WorkerTask.run(WorkerTask.java:234)
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$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
Caused by: org.apache.kafka.connect.errors.DataException: Failed to deserialize data for topic com.obj_pos to Protobuf:
at io.confluent.connect.protobuf.ProtobufConverter.toConnectData(ProtobufConverter.java:123)
at org.apache.kafka.connect.storage.Converter.toConnectData(Converter.java:87)
at org.apache.kafka.connect.runtime.WorkerSinkTask.lambda$convertAndTransformRecord$1(WorkerSinkTask.java:491)
at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execAndRetry(RetryWithToleranceOperator.java:128)
at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execAndHandleError(RetryWithToleranceOperator.java:162)
... 13 more
Caused by: org.apache.kafka.common.errors.SerializationException: Error deserializing Protobuf message for id -1
Caused by: org.apache.kafka.common.errors.SerializationException: Unknown magic byte!
[2020-09-25 13:12:09,010] ERROR WorkerSinkTask{id=s3_parquet_connector-0} Task is being killed and will not recover until manually restarted (org.apache.kafka.connect.runtime.WorkerTask)
my question is, if there is a way of somehow either disable this magic byte check or if I could create a kafka stream that would just append a this 5 bytes to the initial message so that afterwards I could consume it with a consumer that would connect to the schema registry.
What is happening is that the producer is out of my control so I would need somehow to be able to deserialize messages that do not contain those 5 bytes because they are produced by producers that don't rely on the confluent serializers/de-serializers
they are produced by producers that don't rely on the confluent serializers
Then the problem isn't the Registry.
You shouldn't be using the Converters written by Confluent to consume the messages, as those are bound to the Registry, and there is no way to skip it.
You would instead use the BlueApron ones (assuming the data is really protobuf), or write your own Converter classes.
there's something weird happening when we are creating AVRO messages through KSQL and try to consume them by using Kafka Connect. A bit of context:
Source data
A 3rd party provider is producing data on one of our Kafka clusters as JSON (so far, so good). We actually see the data coming in.
Data Transformation
As our internal systems require data to be encoded in AVRO, we created a KSQL cluster that transforms the incoming data into AVRO by creating the following stream in KSQL:
{
"ksql": "
CREATE STREAM src_stream (browser_name VARCHAR)
WITH (KAFKA_TOPIC='json_topic', VALUE_FORMAT='JSON');
CREATE STREAM sink_stream WITH (KAFKA_TOPIC='avro_topic',VALUE_FORMAT='AVRO', PARTITIONS=1, REPLICAS=3) AS
SELECT * FROM src_stream;
",
"streamsProperties": {
"ksql.streams.auto.offset.reset": "earliest"
}
}
(so far, so good)
We see the data being produced from the JSON topic onto the AVRO topic, as the offset increases.
We then create a Kafka connector in a (new) Kafka Connect cluster. As some context, we are using multiple Kafka Connect clusters (with the same properties for those clusters), and as such we have a Kafka Connect cluster running for this data, but an exact copy of the cluster for other AVRO data (1 is for analytics, 1 for our business data).
The sink for this connector is BigQuery, we're using the Wepay BigQuery Sink Connector 1.2.0. Again, so far, so good. Our business cluster is running fine with this connector and the AVRO topics on the business cluster are streaming into BigQuery.
When we try to consume the AVRO topic created by our KSQL statement earlier however, we see an exception being thrown :/
The exception is the following:
org.apache.kafka.connect.errors.ConnectException: Tolerance exceeded in error handler
at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execAndHandleError(RetryWithToleranceOperator.java:178)
at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execute(RetryWithToleranceOperator.java:104)
at org.apache.kafka.connect.runtime.WorkerSinkTask.convertAndTransformRecord(WorkerSinkTask.java:510)
at org.apache.kafka.connect.runtime.WorkerSinkTask.convertMessages(WorkerSinkTask.java:490)
at org.apache.kafka.connect.runtime.WorkerSinkTask.poll(WorkerSinkTask.java:321)
at org.apache.kafka.connect.runtime.WorkerSinkTask.iteration(WorkerSinkTask.java:225)
at org.apache.kafka.connect.runtime.WorkerSinkTask.execute(WorkerSinkTask.java:193)
at org.apache.kafka.connect.runtime.WorkerTask.doRun(WorkerTask.java:175)
at org.apache.kafka.connect.runtime.WorkerTask.run(WorkerTask.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$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
Caused by: org.apache.kafka.connect.errors.DataException: dpt_video_event-created_v2
at io.confluent.connect.avro.AvroConverter.toConnectData(AvroConverter.java:98)
at org.apache.kafka.connect.runtime.WorkerSinkTask.lambda$convertAndTransformRecord$0(WorkerSinkTask.java:510)
at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execAndRetry(RetryWithToleranceOperator.java:128)
at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execAndHandleError(RetryWithToleranceOperator.java:162)
... 13 more
Caused by: org.apache.kafka.common.errors.SerializationException: Error retrieving Avro schema for id 0
Caused by: io.confluent.kafka.schemaregistry.client.rest.exceptions.RestClientException: Schema not found; error code: 40403
at io.confluent.kafka.schemaregistry.client.rest.RestService.sendHttpRequest(RestService.java:209)
at io.confluent.kafka.schemaregistry.client.rest.RestService.httpRequest(RestService.java:235)
at io.confluent.kafka.schemaregistry.client.rest.RestService.getId(RestService.java:415)
at io.confluent.kafka.schemaregistry.client.rest.RestService.getId(RestService.java:408)
at io.confluent.kafka.schemaregistry.client.CachedSchemaRegistryClient.getSchemaByIdFromRegistry(CachedSchemaRegistryClient.java:123)
at io.confluent.kafka.schemaregistry.client.CachedSchemaRegistryClient.getBySubjectAndId(CachedSchemaRegistryClient.java:190)
at io.confluent.kafka.schemaregistry.client.CachedSchemaRegistryClient.getById(CachedSchemaRegistryClient.java:169)
at io.confluent.kafka.serializers.AbstractKafkaAvroDeserializer.deserialize(AbstractKafkaAvroDeserializer.java:121)
at io.confluent.kafka.serializers.AbstractKafkaAvroDeserializer.deserializeWithSchemaAndVersion(AbstractKafkaAvroDeserializer.java:243)
at io.confluent.connect.avro.AvroConverter$Deserializer.deserialize(AvroConverter.java:134)
at io.confluent.connect.avro.AvroConverter.toConnectData(AvroConverter.java:85)
at org.apache.kafka.connect.runtime.WorkerSinkTask.lambda$convertAndTransformRecord$0(WorkerSinkTask.java:510)
at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execAndRetry(RetryWithToleranceOperator.java:128)
at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execAndHandleError(RetryWithToleranceOperator.java:162)
at org.apache.kafka.connect.runtime.errors.RetryWithToleranceOperator.execute(RetryWithToleranceOperator.java:104)
at org.apache.kafka.connect.runtime.WorkerSinkTask.convertAndTransformRecord(WorkerSinkTask.java:510)
at org.apache.kafka.connect.runtime.WorkerSinkTask.convertMessages(WorkerSinkTask.java:490)
at org.apache.kafka.connect.runtime.WorkerSinkTask.poll(WorkerSinkTask.java:321)
at org.apache.kafka.connect.runtime.WorkerSinkTask.iteration(WorkerSinkTask.java:225)
at org.apache.kafka.connect.runtime.WorkerSinkTask.execute(WorkerSinkTask.java:193)
at org.apache.kafka.connect.runtime.WorkerTask.doRun(WorkerTask.java:175)
at org.apache.kafka.connect.runtime.WorkerTask.run(WorkerTask.java:219)
at java.util.concurrent.Executors$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)
Which, to us, indicates that Kafka Connect is reading the message, decodes the AVRO and tries to fetch the schema with ID 0 from the schema registry. Obviously, schema IDs in the schema registry are always > 0.
We're currently stuck in trying to identify the issue here. It looks like KSQL is encoding the message with schema ID 0, but we're unable to find the cause for that :/
Any help is appreciated!
BR,
Patrick
UPDATE:
We have implemented a basic consumer for the AVRO messages and that consumer is correctly identifying the schema in the AVRO messages (ID: 3), so it seems to be rekated to Kafka Connect, instead of the actual KSQL / AVRO messages.
Obviously, schema IDs in the schema registry are always > 0... It looks like KSQL is encoding the message with schema ID 0, but we're unable to find the cause for that
The AvroConverter does a "dumb check" that only looks that the consumed bytes start with a magic byte of 0x0. The next 4 bytes are the ID.
If you are using key.converter=AvroConverter and your keys start like 0x00000 in hex, then the ID would be shown as 0 in the logs, and the lookup would fail.
Last I checked, KSQL doesn't output keys in Avro format, so you will want to check the properties of your connector.
I am trying to use a new feature (https://www.confluent.io/blog/put-several-event-types-kafka-topic/) regarding storing two
different types of events on the same topic. Actually I am using Confluent version 4.1.0 and set these properties below
to make this happen
properties.put(KafkaAvroSerializerConfig.VALUE_SUBJECT_NAME_STRATEGY,TopicRecordNameStrategy.class.getName());
properties.put("value.multi.type", true);
Data are written to topic without issues and can be seen from a Kafka Streams application as Generic Avro Records. Also
on the Kafka Schema registry two new entries are created one for each event hosted on that specific topic.
The problem I am facing is that I cannot export these data from this topic using Kafka Connect. In the simplest case when
I use a File Sink Connector as below
{
"name": "sink-connector",
"config": {
"topics": "source-topic",
"connector.class": "org.apache.kafka.connect.file.FileStreamSinkConnector",
"tasks.max": 1,
"key.converter": "org.apache.kafka.connect.storage.StringConverter",
"key.converter.schema.registry.url":"http://kafka-schema-registry:8081",
"value.converter":"io.confluent.connect.avro.AvroConverter",
"value.converter.schema.registry.url":"http://kafka-schema-registry:8081",
"value.subject.name.strategy":"io.confluent.kafka.serializers.subject.TopicRecordNameStrategy",
"file": "/tmp/sink-file.txt"
}
}
I get an error from the Connector that seems to be some kind of serialization error based on AvroConverter like
the one shown here
org.apache.kafka.connect.errors.DataException: source-topic
at io.confluent.connect.avro.AvroConverter.toConnectData(AvroConverter.java:95)
at org.apache.kafka.connect.runtime.WorkerSinkTask.convertMessages(WorkerSinkTask.java:468)
at org.apache.kafka.connect.runtime.WorkerSinkTask.poll(WorkerSinkTask.java:301)
at org.apache.kafka.connect.runtime.WorkerSinkTask.iteration(WorkerSinkTask.java:205)
at org.apache.kafka.connect.runtime.WorkerSinkTask.execute(WorkerSinkTask.java:173)
at org.apache.kafka.connect.runtime.WorkerTask.doRun(WorkerTask.java:170)
at org.apache.kafka.connect.runtime.WorkerTask.run(WorkerTask.java:214)
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:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)
Caused by: org.apache.kafka.common.errors.SerializationException: Error retrieving Avro schema for id 2
Caused by: io.confluent.kafka.schemaregistry.client.rest.exceptions.RestClientException: Subject not found.; error code: 40401
at io.confluent.kafka.schemaregistry.client.rest.RestService.sendHttpRequest(RestService.java:202)
at io.confluent.kafka.schemaregistry.client.rest.RestService.httpRequest(RestService.java:229)
at io.confluent.kafka.schemaregistry.client.rest.RestService.lookUpSubjectVersion(RestService.java:296)
at io.confluent.kafka.schemaregistry.client.rest.RestService.lookUpSubjectVersion(RestService.java:284)
at io.confluent.kafka.schemaregistry.client.CachedSchemaRegistryClient.getVersionFromRegistry(CachedSchemaRegistryClient.java:125)
at io.confluent.kafka.schemaregistry.client.CachedSchemaRegistryClient.getVersion(CachedSchemaRegistryClient.java:236)
at io.confluent.kafka.serializers.AbstractKafkaAvroDeserializer.deserialize(AbstractKafkaAvroDeserializer.java:152)
at io.confluent.kafka.serializers.AbstractKafkaAvroDeserializer.deserializeWithSchemaAndVersion(AbstractKafkaAvroDeserializer.java:194)
at io.confluent.connect.avro.AvroConverter$Deserializer.deserialize(AvroConverter.java:120)
at io.confluent.connect.avro.AvroConverter.toConnectData(AvroConverter.java:83)
at org.apache.kafka.connect.runtime.WorkerSinkTask.convertMessages(WorkerSinkTask.java:468)
at org.apache.kafka.connect.runtime.WorkerSinkTask.poll(WorkerSinkTask.java:301)
at org.apache.kafka.connect.runtime.WorkerSinkTask.iteration(WorkerSinkTask.java:205)
at org.apache.kafka.connect.runtime.WorkerSinkTask.execute(WorkerSinkTask.java:173)
at org.apache.kafka.connect.runtime.WorkerTask.doRun(WorkerTask.java:170)
at org.apache.kafka.connect.runtime.WorkerTask.run(WorkerTask.java:214)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)
Note that schema registry has an Avro schema with id 2 and another with schema id 3 that describe the two events hosted on the
same topic. Same issues arise when using JDBC connector.
So how do I handle this case in order to export data from my Kafka Cluster
to an external system. Am I missing something on my configuration ? Is it possible to have a topic with multiple type of events
and export them through Kafka Connect ?
Found the solution. My code was passing key as String and value as avro. Hive-sink while reading tried to lookup for avro schema of key and was not able to find it.
Adding the property
key.converter=org.apache.kafka.connect.storage.StringConverter
key.converter.schema.registry.url=http://localhost:8081
helped resolve the issue.
Background: i used wrong avro schema registry while producing to prod topic and as a result the kafka connect went down because of the messages with wrong schema id.So as a recovery plan we wanted to copy the messages in the prod topic to a test topic and then write the good messages to the hdfs.But we are facing issues with certain offsets that have wrong schema id while reading from prod topic.Is there a way to ignore such offsets while writing to another topic.
Exception in thread "StreamThread-1"
org.apache.kafka.streams.errors.StreamsException: Failed to deserialize value
for record. topic=xxxx, partition=9, offset=1259032
Caused by: org.apache.kafka.common.errors.SerializationException: Error
retrieving Avro schema for id 600
Caused by:
io.confluent.kafka.schemaregistry.client.rest.exceptions.RestClientException:
Schema not found io.confluent.rest.exceptions.RestNotFoundException: Schema not found
io.confluent.rest.exceptions.RestNotFoundException: Schema not found
{code}
You can change the deserialization exception handler to skip over those record as describe in the docs: https://docs.confluent.io/current/streams/faq.html#handling-corrupted-records-and-deserialization-errors-poison-pill-records
Ie, you set LogAndContinueExceptionHandler in the config via parameter default.deserialization.exception.handler.