[
https://issues.apache.org/jira/browse/KAFKA-899?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14151847#comment-14151847
]
Jun Rao commented on KAFKA-899:
-------------------------------
Andras,
It seems that your issue is a bit different from this jira. This jira is about
removing the stacktrace in the producer log when the metadata is not available.
Your issue seems to be that the metadata is not propagated as quickly as you
expect. Normally, 100ms should be long enough for a new topic to be created and
its metadata be propagated to all brokers. In your case, it seems that process
takes more than 1 sec. Could you look at the controller and the state-change
log to see where the delay is? For example, is the write to ZK slow or is the
propagation of metadata from the controller to the broker slow?
> LeaderNotAvailableException the first time a new message for a partition is
> processed.
> --------------------------------------------------------------------------------------
>
> Key: KAFKA-899
> URL: https://issues.apache.org/jira/browse/KAFKA-899
> Project: Kafka
> Issue Type: Bug
> Components: core
> Affects Versions: 0.8.0
> Reporter: Jason Rosenberg
> Assignee: Jun Rao
> Fix For: 0.8.2
>
> Attachments: kafka-899.patch, kafka-899_v2.patch, kafka-899_v3.patch
>
>
> I'm porting some unit tests from 0.7.2 to 0.8.0. The test does the
> following, all embedded in the same java process:
> -- spins up a zk instance
> -- spins up a kafka server using a fresh log directory
> -- creates a producer and sends a message
> -- creates a high-level consumer and verifies that it can consume the message
> -- shuts down the consumer
> -- stops the kafka server
> -- stops zk
> The test seems to be working fine now, however, I consistently see the
> following exceptions (which from poking around the mailing list seem to be
> expected?). If these are expected, can we suppress the logging of these
> exceptions, since it clutters the output of tests, and presumably, clutters
> the logs of the running server/consumers, during clean startup and
> shutdown......
> When I call producer.send(), I get:
> 1071 [main] WARN kafka.producer.BrokerPartitionInfo - Error while fetching
> metadata partition 0 leader: none replicas: isr:
> isUnderReplicated: false for topic partition [test-topic,0]: [class
> kafka.common.LeaderNotAvailableException]
> 1081 [main] WARN kafka.producer.async.DefaultEventHandler - Failed to
> collate messages by topic,partition due to
> kafka.common.LeaderNotAvailableException: No leader for any partition
> at
> kafka.producer.async.DefaultEventHandler.kafka$producer$async$DefaultEventHandler$$getPartition(DefaultEventHandler.scala:212)
> at
> kafka.producer.async.DefaultEventHandler$$anonfun$partitionAndCollate$1.apply(DefaultEventHandler.scala:150)
> at
> kafka.producer.async.DefaultEventHandler$$anonfun$partitionAndCollate$1.apply(DefaultEventHandler.scala:148)
> at
> scala.collection.mutable.ResizableArray$class.foreach(ResizableArray.scala:57)
> at scala.collection.mutable.ArrayBuffer.foreach(ArrayBuffer.scala:43)
> at
> kafka.producer.async.DefaultEventHandler.partitionAndCollate(DefaultEventHandler.scala:148)
> at
> kafka.producer.async.DefaultEventHandler.dispatchSerializedData(DefaultEventHandler.scala:94)
> at
> kafka.producer.async.DefaultEventHandler.handle(DefaultEventHandler.scala:72)
> at kafka.producer.Producer.send(Producer.scala:74)
> at kafka.javaapi.producer.Producer.send(Producer.scala:32)
> at
> com.squareup.kafka.server.KafkaServerTest.produceAndConsumeMessage(KafkaServerTest.java:98)
> at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> at
> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
> at
> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
> at java.lang.reflect.Method.invoke(Method.java:597)
> at
> org.junit.runners.model.FrameworkMethod$1.runReflectiveCall(FrameworkMethod.java:44)
> at
> org.junit.internal.runners.model.ReflectiveCallable.run(ReflectiveCallable.java:15)
> at
> org.junit.runners.model.FrameworkMethod.invokeExplosively(FrameworkMethod.java:41)
> at
> org.junit.internal.runners.statements.InvokeMethod.evaluate(InvokeMethod.java:20)
> at
> org.junit.internal.runners.statements.RunBefores.evaluate(RunBefores.java:28)
> at org.junit.runners.ParentRunner.runLeaf(ParentRunner.java:263)
> at
> org.junit.runners.BlockJUnit4ClassRunner.runChild(BlockJUnit4ClassRunner.java:69)
> at
> org.junit.runners.BlockJUnit4ClassRunner.runChild(BlockJUnit4ClassRunner.java:48)
> at org.junit.runners.ParentRunner$3.run(ParentRunner.java:231)
> at org.junit.runners.ParentRunner$1.schedule(ParentRunner.java:60)
> at org.junit.runners.ParentRunner.runChildren(ParentRunner.java:229)
> at org.junit.runners.ParentRunner.access$000(ParentRunner.java:50)
> at org.junit.runners.ParentRunner$2.evaluate(ParentRunner.java:222)
> at org.junit.runners.ParentRunner.run(ParentRunner.java:292)
> at org.junit.runner.JUnitCore.run(JUnitCore.java:157)
> at
> com.intellij.junit4.JUnit4IdeaTestRunner.startRunnerWithArgs(JUnit4IdeaTestRunner.java:77)
> at
> com.intellij.rt.execution.junit.JUnitStarter.prepareStreamsAndStart(JUnitStarter.java:195)
> at
> com.intellij.rt.execution.junit.JUnitStarter.main(JUnitStarter.java:63)
> 1133 [kafka-request-handler-1] WARN kafka.server.HighwaterMarkCheckpoint -
> No highwatermark file is found. Returning 0 as the highwatermark for
> partition [test-topic,0]
> ...
> ...
> It would be great if instead of this exception, it would just log a
> meaningful message, like:
> "No leader was available for partition X, one will now be created"
> Jason
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
