apache-spark - 查询开始时使用结构化流从 Kafka 主题的开头读取

Question

我使用结构化流从kafka主题中读取，使用spark 2.4和scala 2.12

我使用检查点使我的查询容错。

但是，每次我开始查询时，它都会跳转到当前偏移量，而不会在连接到主题之前读取现有数据。

我缺少 kafka 流的配置吗？

读：

 val df = spark.readStream
    .format("kafka")
    .option("kafka.bootstrap.servers", "localhost:9092")
    .option("subscribe", "test")
    .option("maxOffsetsPerTrigger","1")
    .option("startingOffset","earliest")
    .option("auto.offset.reset","earliest")
    .load()
val msg = df.select($"value" cast "string", $"topic", $"partition", $"offset")

写：

val query= msg.writeStream
    .foreachBatch(
      (dfbatch: Dataset[Row], batchid: Long) =>
      {
      println(s"IM AT BATCH ID: $batchid")
      dfbatch.show()
      dfbatch.write.csv(s"s3a://abucket/$param")
      }
    )
      .option("checkpointLocation","s3a://checkpoint/")
      .trigger(Trigger.ProcessingTime("10 seconds"))
      .format("console")
      .start()

 query.awaitTermination()

编辑：

这是我清空检查点目录后的日志：

0/07/11 18:15:16 INFO CheckpointFileManager: Writing atomically to s3a://checkpoint/metadata using temp file s3a://checkpoint/.metadata.304a751a-68b7-4b8d-858c-3aa5df272db4.tmp
20/07/11 18:15:17 INFO CheckpointFileManager: Renamed temp file s3a://checkpoint/.metadata.304a751a-68b7-4b8d-858c-3aa5df272db4.tmp to s3a://checkpoint/metadata
20/07/11 18:15:17 INFO MicroBatchExecution: Starting [id = e83c6066-9611-4e9b-97d5-d02421b2d1d6, runId = e77896e3-ce76-488b-8345-7a29cc0d7d0b]. Use s3a://checkpoint/ to store the query checkpoint.
20/07/11 18:15:17 INFO MicroBatchExecution: Using MicroBatchReader [KafkaV2[Subscribe[test]]] from DataSourceV2 named 'kafka' [org.apache.spark.sql.kafka010.KafkaSourceProvider@2375c472]
20/07/11 18:15:17 INFO MicroBatchExecution: Starting new streaming query.
20/07/11 18:15:17 INFO MicroBatchExecution: Stream started from {}
20/07/11 18:15:18 INFO ConsumerConfig: ConsumerConfig values: 
    auto.commit.interval.ms = 5000
    auto.offset.reset = earliest
    bootstrap.servers = [localhost:9092]
    check.crcs = true
    client.id = 
    connections.max.idle.ms = 540000
    default.api.timeout.ms = 60000
    enable.auto.commit = false
    exclude.internal.topics = true
    fetch.max.bytes = 52428800
    fetch.max.wait.ms = 500
    fetch.min.bytes = 1
    group.id = spark-kafka-source-1fa35d7f-b356-4806-9ee7-658ef48c837d--2088528104-driver-0
    heartbeat.interval.ms = 3000
    interceptor.classes = []
    internal.leave.group.on.close = true
    isolation.level = read_uncommitted
    key.deserializer = class org.apache.kafka.common.serialization.ByteArrayDeserializer
    max.partition.fetch.bytes = 1048576
    max.poll.interval.ms = 300000
    max.poll.records = 1
    metadata.max.age.ms = 300000
    metric.reporters = []
    metrics.num.samples = 2
    metrics.recording.level = INFO
    metrics.sample.window.ms = 30000
    partition.assignment.strategy = [class org.apache.kafka.clients.consumer.RangeAssignor]
    receive.buffer.bytes = 65536
    reconnect.backoff.max.ms = 1000
    reconnect.backoff.ms = 50
    request.timeout.ms = 30000
    retry.backoff.ms = 100
    sasl.client.callback.handler.class = null
    sasl.jaas.config = null
    sasl.kerberos.kinit.cmd = /usr/bin/kinit
    sasl.kerberos.min.time.before.relogin = 60000
    sasl.kerberos.service.name = null
    sasl.kerberos.ticket.renew.jitter = 0.05
    sasl.kerberos.ticket.renew.window.factor = 0.8
    sasl.login.callback.handler.class = null
    sasl.login.class = null
    sasl.login.refresh.buffer.seconds = 300
    sasl.login.refresh.min.period.seconds = 60
    sasl.login.refresh.window.factor = 0.8
    sasl.login.refresh.window.jitter = 0.05
    sasl.mechanism = GSSAPI
    security.protocol = PLAINTEXT
    send.buffer.bytes = 131072
    session.timeout.ms = 10000
    ssl.cipher.suites = null
    ssl.enabled.protocols = [TLSv1.2, TLSv1.1, TLSv1]
    ssl.endpoint.identification.algorithm = https
    ssl.key.password = null
    ssl.keymanager.algorithm = SunX509
    ssl.keystore.location = null
    ssl.keystore.password = null
    ssl.keystore.type = JKS
    ssl.protocol = TLS
    ssl.provider = null
    ssl.secure.random.implementation = null
    ssl.trustmanager.algorithm = PKIX
    ssl.truststore.location = null
    ssl.truststore.password = null
    ssl.truststore.type = JKS
    value.deserializer = class org.apache.kafka.common.serialization.ByteArrayDeserializer

20/07/11 18:15:18 INFO AppInfoParser: Kafka version : 2.0.0
20/07/11 18:15:18 INFO AppInfoParser: Kafka commitId : 3402a8361b734732
20/07/11 18:15:18 INFO Metadata: Cluster ID: X8K8aVFyRi6OcUDs1zXOhQ
20/07/11 18:15:18 INFO AbstractCoordinator: [Consumer clientId=consumer-1, groupId=spark-kafka-source-1fa35d7f-b356-4806-9ee7-658ef48c837d--2088528104-driver-0] Discovered group coordinator Myuser-PC:9092 (id: 2147483647 rack: null)
20/07/11 18:15:18 INFO ConsumerCoordinator: [Consumer clientId=consumer-1, groupId=spark-kafka-source-1fa35d7f-b356-4806-9ee7-658ef48c837d--2088528104-driver-0] Revoking previously assigned partitions []
20/07/11 18:15:18 INFO AbstractCoordinator: [Consumer clientId=consumer-1, groupId=spark-kafka-source-1fa35d7f-b356-4806-9ee7-658ef48c837d--2088528104-driver-0] (Re-)joining group
20/07/11 18:15:18 INFO AbstractCoordinator: [Consumer clientId=consumer-1, groupId=spark-kafka-source-1fa35d7f-b356-4806-9ee7-658ef48c837d--2088528104-driver-0] Successfully joined group with generation 1
20/07/11 18:15:18 INFO ConsumerCoordinator: [Consumer clientId=consumer-1, groupId=spark-kafka-source-1fa35d7f-b356-4806-9ee7-658ef48c837d--2088528104-driver-0] Setting newly assigned partitions [test-0]
20/07/11 18:15:18 INFO Fetcher: [Consumer clientId=consumer-1, groupId=spark-kafka-source-1fa35d7f-b356-4806-9ee7-658ef48c837d--2088528104-driver-0] Resetting offset for partition test-0 to offset 0.
20/07/11 18:15:18 INFO Fetcher: [Consumer clientId=consumer-1, groupId=spark-kafka-source-1fa35d7f-b356-4806-9ee7-658ef48c837d--2088528104-driver-0] Resetting offset for partition test-0 to offset 18.
20/07/11 18:15:18 INFO deprecation: io.bytes.per.checksum is deprecated. Instead, use dfs.bytes-per-checksum
20/07/11 18:15:18 INFO CheckpointFileManager: Writing atomically to s3a://checkpoint/sources/0/0 using temp file s3a://checkpoint/sources/0/.0.e900c0dd-dbb5-4ca7-ada2-e3a84e892c5f.tmp
20/07/11 18:15:19 INFO CheckpointFileManager: Renamed temp file s3a://checkpoint/sources/0/.0.e900c0dd-dbb5-4ca7-ada2-e3a84e892c5f.tmp to s3a://checkpoint/sources/0/0
20/07/11 18:15:19 INFO KafkaMicroBatchReader: Initial offsets: {"test":{"0":18}}
20/07/11 18:15:19 INFO Fetcher: [Consumer clientId=consumer-1, groupId=spark-kafka-source-1fa35d7f-b356-4806-9ee7-658ef48c837d--2088528104-driver-0] Resetting offset for partition test-0 to offset 18.
20/07/11 18:15:19 INFO CheckpointFileManager: Writing atomically to s3a://checkpoint/offsets/0 using temp file s3a://checkpoint/offsets/.0.a7a4e7f3-7e4a-433f-8532-23d6179c3b98.tmp
20/07/11 18:15:19 INFO CheckpointFileManager: Renamed temp file s3a://checkpoint/offsets/.0.a7a4e7f3-7e4a-433f-8532-23d6179c3b98.tmp to s3a://checkpoint/offsets/0
20/07/11 18:15:19 INFO MicroBatchExecution: Committed offsets for batch 0. Metadata OffsetSeqMetadata(0,1594480519101,Map(spark.sql.streaming.stateStore.providerClass -> org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider, spark.sql.streaming.flatMapGroupsWithState.stateFormatVersion -> 2, spark.sql.streaming.multipleWatermarkPolicy -> min, spark.sql.streaming.aggregation.stateFormatVersion -> 2, spark.sql.shuffle.partitions -> 200))
20/07/11 18:15:20 INFO KafkaMicroBatchReader: Partitions added: Map()
20/07/11 18:15:20 INFO CodeGenerator: Code generated in 171.85213 ms
20/07/11 18:15:20 INFO CodeGenerator: Code generated in 23.189288 ms

我进入检查点目录并修改了数据，它从头开始处理！必须有一个相同的配置..

标签： apache-sparkapache-kafkaspark-structured-streaming

Answer 1

太烦人了……我拼错了这个选项startingOffset

正确的拼写方式是：

   .option("startingOffsets","earliest")

现在它可以工作了。