[
https://issues.apache.org/jira/browse/STORM-2494?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16129179#comment-16129179
]
Stig Rohde Døssing commented on STORM-2494:
-------------------------------------------
[~hmclouro] Okay, I just wanted to make sure you didn't spend time debugging to
find out why this happens :)
Regarding fixing this, I think it might not be easy. We'd need to make sure
that the spout always calls poll before commit in nextTuple, so it can trigger
rebalance in case the spout has been idle for longer than the session timeout,
and we'd probably also need to change the rebalance listener logic. It
currently assumes that when onPartitionsRevoked runs, the spout was assigned
the partitions it has offset managers for.
Users who don't want to upgrade storm-kafka-client for some reason can still
work around this by setting a higher session timeout, and newer Kafka versions
allow you to set it pretty high without losing fast detection of dead consumers
due to KIP-62.
> KafkaSpout does not handle CommitFailedException
> ------------------------------------------------
>
> Key: STORM-2494
> URL: https://issues.apache.org/jira/browse/STORM-2494
> Project: Apache Storm
> Issue Type: Bug
> Components: storm-kafka-client
> Affects Versions: 1.1.0
> Reporter: Yuri Barseghyan
> Assignee: Hugo Louro
>
> In situations when tuple processing takes longer than session timeout, we get
> CommitFailedException and instead of recovering from it Storm worker dies.
> {code}
> 2017-04-26 11:07:04.902 o.a.s.util [ERROR] Async loop died!
> org.apache.kafka.clients.consumer.CommitFailedException: Commit cannot be
> completed since the group has already rebalanced and assigned the partitions
> to another member. This means that the time between subsequent calls to
> poll() was longer than the configured session.timeout.ms, which typically
> implies that the poll loop is spending too much time message processing. You
> can address this either by increasing the session timeout or by reducing the
> maximum size of batches returned in poll() with max.poll.records.
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerCoordinator$OffsetCommitResponseHandler.handle(ConsumerCoordinator.java:578)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerCoordinator$OffsetCommitResponseHandler.handle(ConsumerCoordinator.java:519)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.AbstractCoordinator$CoordinatorResponseHandler.onSuccess(AbstractCoordinator.java:679)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.AbstractCoordinator$CoordinatorResponseHandler.onSuccess(AbstractCoordinator.java:658)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.RequestFuture$1.onSuccess(RequestFuture.java:167)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.RequestFuture.fireSuccess(RequestFuture.java:133)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.RequestFuture.complete(RequestFuture.java:107)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient$RequestFutureCompletionHandler.onComplete(ConsumerNetworkClient.java:426)
> ~[stormjar.jar:3.0.2]
> \tat org.apache.kafka.clients.NetworkClient.poll(NetworkClient.java:278)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.clientPoll(ConsumerNetworkClient.java:360)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.poll(ConsumerNetworkClient.java:224)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.poll(ConsumerNetworkClient.java:192)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.poll(ConsumerNetworkClient.java:163)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerCoordinator.commitOffsetsSync(ConsumerCoordinator.java:404)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.KafkaConsumer.commitSync(KafkaConsumer.java:1058)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.storm.kafka.spout.KafkaSpout.commitOffsetsForAckedTuples(KafkaSpout.java:384)
> ~[stormjar.jar:3.0.2]
> \tat org.apache.storm.kafka.spout.KafkaSpout.nextTuple(KafkaSpout.java:219)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.storm.daemon.executor$fn__4976$fn__4991$fn__5022.invoke(executor.clj:644)
> ~[storm-core-1.1.0.jar:1.1.0]
> \tat org.apache.storm.util$async_loop$fn__557.invoke(util.clj:484)
> [storm-core-1.1.0.jar:1.1.0]
> \tat clojure.lang.AFn.run(AFn.java:22) [clojure-1.7.0.jar:?]
> \tat java.lang.Thread.run(Thread.java:748) [?:1.8.0_131]
> 2017-04-26 11:07:04.909 o.a.s.d.executor [ERROR]
> org.apache.kafka.clients.consumer.CommitFailedException: Commit cannot be
> completed since the group has already rebalanced and assigned the partitions
> to another member. This means that the time between subsequent calls to
> poll() was longer than the configured session.timeout.ms, which typically
> implies that the poll loop is spending too much time message processing. You
> can address this either by increasing the session timeout or by reducing the
> maximum size of batches returned in poll() with max.poll.records.
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerCoordinator$OffsetCommitResponseHandler.handle(ConsumerCoordinator.java:578)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerCoordinator$OffsetCommitResponseHandler.handle(ConsumerCoordinator.java:519)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.AbstractCoordinator$CoordinatorResponseHandler.onSuccess(AbstractCoordinator.java:679)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.AbstractCoordinator$CoordinatorResponseHandler.onSuccess(AbstractCoordinator.java:658)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.RequestFuture$1.onSuccess(RequestFuture.java:167)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.RequestFuture.fireSuccess(RequestFuture.java:133)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.RequestFuture.complete(RequestFuture.java:107)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient$RequestFutureCompletionHandler.onComplete(ConsumerNetworkClient.java:426)
> ~[stormjar.jar:3.0.2]
> \tat org.apache.kafka.clients.NetworkClient.poll(NetworkClient.java:278)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.clientPoll(ConsumerNetworkClient.java:360)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.poll(ConsumerNetworkClient.java:224)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.poll(ConsumerNetworkClient.java:192)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.poll(ConsumerNetworkClient.java:163)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.KafkaConsumer.commitSync(KafkaConsumer.java:1058)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.kafka.clients.consumer.internals.ConsumerCoordinator.commitOffsetsSync(ConsumerCoordinator.java:404)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.storm.kafka.spout.KafkaSpout.commitOffsetsForAckedTuples(KafkaSpout.java:384)
> ~[stormjar.jar:3.0.2]
> \tat org.apache.storm.kafka.spout.KafkaSpout.nextTuple(KafkaSpout.java:219)
> ~[stormjar.jar:3.0.2]
> \tat
> org.apache.storm.daemon.executor$fn__4976$fn__4991$fn__5022.invoke(executor.clj:644)
> ~[storm-core-1.1.0.jar:1.1.0]
> \tat org.apache.storm.util$async_loop$fn__557.invoke(util.clj:484)
> [storm-core-1.1.0.jar:1.1.0]
> \tat clojure.lang.AFn.run(AFn.java:22) [clojure-1.7.0.jar:?]
> \tat java.lang.Thread.run(Thread.java:748) [?:1.8.0_131]
> 2017-04-26 11:07:04.953 o.a.s.util [ERROR] Halting process: (\"Worker died\")
> java.lang.RuntimeException: (\"Worker died\")
> \tat org.apache.storm.util$exit_process_BANG_.doInvoke(util.clj:341)
> [storm-core-1.1.0.jar:1.1.0]
> \tat clojure.lang.RestFn.invoke(RestFn.java:423) [clojure-1.7.0.jar:?]
> \tat org.apache.storm.daemon.worker$fn__5646$fn__5647.invoke(worker.clj:763)
> [storm-core-1.1.0.jar:1.1.0]
> \tat
> org.apache.storm.daemon.executor$mk_executor_data$fn__4863$fn__4864.invoke(executor.clj:274)
> [storm-core-1.1.0.jar:1.1.0]
> \tat org.apache.storm.util$async_loop$fn__557.invoke(util.clj:494)
> [storm-core-1.1.0.jar:1.1.0]
> \tat clojure.lang.AFn.run(AFn.java:22) [clojure-1.7.0.jar:?]
> \tat java.lang.Thread.run(Thread.java:748) [?:1.8.0_131]
> 2017-04-26 11:07:44.507 o.a.s.k.s.KafkaSpoutRetryExponentialBackoff [DEBUG]
> Instantiated KafkaSpoutRetryExponentialBackoff{delay=TimeInterval{length=0,
> timeUnit=SECONDS}, ratio=TimeInterval{length=2, timeUnit=MILLISECONDS},
> maxRetries=2147483647, maxRetryDelay=TimeInterval{length=10,
> timeUnit=SECONDS}}
> 2017-04-26 11:07:44.516 o.a.s.k.s.KafkaSpoutRetryExponentialBackoff [DEBUG]
> Instantiated KafkaSpoutRetryExponentialBackoff{delay=TimeInterval{length=0,
> timeUnit=SECONDS}, ratio=TimeInterval{length=0, timeUnit=MILLISECONDS},
> maxRetries=2147483647, maxRetryDelay=TimeInterval{length=0,
> timeUnit=MILLISECONDS}}
> 2017-04-26 11:07:45.048 o.a.s.k.s.KafkaSpout [INFO] Kafka Spout opened with
> the following configuration:
> KafkaSpoutConfig{kafkaProps={enable.auto.commit=false,
> request.timeout.ms=30000, group.id=Group1,
> bootstrap.servers=192.168.1.143:9092, session.timeout.ms=20000},
> key=org.apache.kafka.common.serialization.StringDeserializer@1b5080fd,
> value=org.apache.kafka.common.serialization.StringDeserializer@2720873b,
> pollTimeoutMs=200, offsetCommitPeriodMs=5000, maxUncommittedOffsets=1000,
> firstPollOffsetStrategy=UNCOMMITTED_EARLIEST,
> subscription=org.apache.storm.kafka.spout.NamedSubscription@7f068c1f,
> translator=org.apache.storm.kafka.spout.SimpleRecordTranslator@1f1ca6a2,
> retryService=KafkaSpoutRetryExponentialBackoff{delay=TimeInterval{length=0,
> timeUnit=SECONDS}, ratio=TimeInterval{length=2, timeUnit=MILLISECONDS},
> maxRetries=2147483647, maxRetryDelay=TimeInterval{length=10,
> timeUnit=SECONDS}}}
> 2017-04-26 11:07:45.111 o.a.s.k.s.KafkaSpout [INFO] Kafka Spout opened with
> the following configuration:
> KafkaSpoutConfig{kafkaProps={enable.auto.commit=false,
> request.timeout.ms=30000, group.id=Group2,
> bootstrap.servers=192.168.1.143:9092, session.timeout.ms=20000},
> key=org.apache.kafka.common.serialization.StringDeserializer@45ffa954,
> value=org.apache.kafka.common.serialization.StringDeserializer@4b384f9b,
> pollTimeoutMs=200, offsetCommitPeriodMs=5000, maxUncommittedOffsets=1000,
> firstPollOffsetStrategy=UNCOMMITTED_EARLIEST,
> subscription=org.apache.storm.kafka.spout.NamedSubscription@4f07c224,
> translator=org.apache.storm.kafka.spout.SimpleRecordTranslator@a0545a0,
> retryService=KafkaSpoutRetryExponentialBackoff{delay=TimeInterval{length=0,
> timeUnit=SECONDS}, ratio=TimeInterval{length=2, timeUnit=MILLISECONDS},
> maxRetries=2147483647, maxRetryDelay=TimeInterval{length=10,
> timeUnit=SECONDS}}}
> 2017-04-26 11:07:45.297 o.a.s.k.s.NamedSubscription [INFO] Kafka consumer
> subscribed topics [topic-1]
> 2017-04-26 11:07:45.302 o.a.s.k.s.NamedSubscription [INFO] Kafka consumer
> subscribed topics [topic-2]
> 2017-04-26 11:07:45.456 o.a.s.k.s.KafkaSpout [INFO] Partitions revoked.
> [consumer-group=Group1,
> consumer=org.apache.kafka.clients.consumer.KafkaConsumer@32cbdbb0,
> topic-partitions=[]]
> 2017-04-26 11:07:45.463 o.a.s.k.s.KafkaSpout [INFO] Partitions revoked.
> [consumer-group=Group1,
> consumer=org.apache.kafka.clients.consumer.KafkaConsumer@275d5222,
> topic-partitions=[]]
> 2017-04-26 11:07:45.545 o.a.s.k.s.KafkaSpout [INFO] Partitions reassignment.
> [consumer-group=Group1,
> consumer=org.apache.kafka.clients.consumer.KafkaConsumer@275d5222,
> topic-partitions=[topic-1]]
> 2017-04-26 11:07:45.546 o.a.s.k.s.KafkaSpout [INFO] Partitions reassignment.
> [consumer-group=Group1,
> consumer=org.apache.kafka.clients.consumer.KafkaConsumer@32cbdbb0,
> topic-partitions=[topic-2]]
> 2017-04-26 11:07:45.551 o.a.s.k.s.i.OffsetManager [DEBUG] Instantiated
> OffsetManager{topic-partition=topic-1, fetchOffset=11803,
> committedOffset=11802, ackedMsgs=[]}
> 2017-04-26 11:07:45.551 o.a.s.k.s.i.OffsetManager [DEBUG] Instantiated
> OffsetManager{topic-partition=topic-2, fetchOffset=11801,
> committedOffset=11800, ackedMsgs=[]}
> 2017-04-26 11:07:45.552 o.a.s.k.s.KafkaSpout [INFO] Initialization complete
> 2017-04-26 11:07:45.552 o.a.s.k.s.KafkaSpout [INFO] Initialization complete
> {code}
> I think expected behaviour would be that KafkaSpout would recover from
> exception (client will reconnect and get partitions reassigned) without
> worker getting killed.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
