[
https://issues.apache.org/jira/browse/KAFKA-2890?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15028420#comment-15028420
]
Alexander Kukushkin edited comment on KAFKA-2890 at 11/26/15 9:50 AM:
----------------------------------------------------------------------
1) I've created cluster of three nodes, with broker ids: 114, 181, 219
2) I've created topics:
bin/kafka-topics.sh --create --zookeeper 172.31.151.192:2181/medium
--replication-factor 3 --partitions 128 --topic test1
bin/kafka-topics.sh --create --zookeeper 172.31.151.192:2181/medium
--replication-factor 3 --partitions 128 --topic test2
bin/kafka-topics.sh --create --zookeeper 172.31.151.192:2181/medium
--replication-factor 3 --partitions 128 --topic test3
bin/kafka-topics.sh --create --zookeeper 172.31.151.192:2181/medium
--replication-factor 3 --partitions 128 --topic test4
bin/kafka-topics.sh --create --zookeeper 172.31.151.192:2181/medium
--replication-factor 3 --partitions 128 --topic test5
bin/kafka-topics.sh --create --zookeeper 172.31.151.192:2181/medium
--replication-factor 3 --partitions 128 --topic test6
3) I've started fourth node, with the broker id: 41
4) I've generated json file for reassignment (see attachment)
5) I've executed reassign command: bin/kafka-reassign-partitions.sh --zookeeper
172.31.151.192:2181/medium --reassignment-json-file reassign.json --execute
6) As a result I've got reassign.out file (see attachment)
Meanwhile in the log files on nodes one could find following lines:
[2015-11-26 09:05:49,315] WARN [ReplicaFetcherThread-3-114], Error in fetch
kafka.server.ReplicaFetcherThread$FetchRequest@574d6b74. Possible cause:
org.apache.kafka.common.protocol.types.SchemaException: Error
reading field 'responses': Error reading field 'topic':
java.nio.BufferUnderflowException (kafka.server.ReplicaFetcherThread)
[2015-11-26 09:06:01,644] WARN [ReplicaFetcherThread-3-181], Error in fetch
kafka.server.ReplicaFetcherThread$FetchRequest@76dc09c7. Possible cause:
org.apache.kafka.common.protocol.types.SchemaException: Error reading field
'responses': Error reading field 'topic': java.nio.BufferUnderflowException
(kafka.server.ReplicaFetcherThread)
[2015-11-26 09:06:31,024] WARN [ReplicaFetcherThread-1-219], Error in fetch
kafka.server.ReplicaFetcherThread$FetchRequest@3fa05c3d. Possible cause:
org.apache.kafka.common.protocol.types.SchemaException: Error
reading field 'responses': Error reading field 'topic':
java.nio.BufferUnderflowException (kafka.server.ReplicaFetcherThread)
was (Author: cyberdemn):
1) I've created cluster of three nodes, with broker ids: 114, 181, 219
2) I've created topics:
bin/kafka-topics.sh --create --zookeeper 172.31.151.192:2181
--replication-factor 3 --partitions 128 --topic test1
bin/kafka-topics.sh --create --zookeeper 172.31.151.192:2181
--replication-factor 3 --partitions 128 --topic test2
bin/kafka-topics.sh --create --zookeeper 172.31.151.192:2181
--replication-factor 3 --partitions 128 --topic test3
bin/kafka-topics.sh --create --zookeeper 172.31.151.192:2181
--replication-factor 3 --partitions 128 --topic test4
bin/kafka-topics.sh --create --zookeeper 172.31.151.192:2181
--replication-factor 3 --partitions 128 --topic test5
bin/kafka-topics.sh --create --zookeeper 172.31.151.192:2181
--replication-factor 3 --partitions 128 --topic test6
3) I've started fourth node, with the broker id: 41
4) I've generated json file for reassignment (see attachment)
5) I've executed reassign command: bin/kafka-reassign-partitions.sh --zookeeper
172.31.151.192:2181 --reassignment-json-file reassign.json --execute
6) As a result I've got reassign.out file (see attachment)
Meanwhile in the log files on nodes one could find following lines:
[2015-11-26 09:05:49,315] WARN [ReplicaFetcherThread-3-114], Error in fetch
kafka.server.ReplicaFetcherThread$FetchRequest@574d6b74. Possible cause:
org.apache.kafka.common.protocol.types.SchemaException: Error
reading field 'responses': Error reading field 'topic':
java.nio.BufferUnderflowException (kafka.server.ReplicaFetcherThread)
[2015-11-26 09:06:01,644] WARN [ReplicaFetcherThread-3-181], Error in fetch
kafka.server.ReplicaFetcherThread$FetchRequest@76dc09c7. Possible cause:
org.apache.kafka.common.protocol.types.SchemaException: Error reading field
'responses': Error reading field 'topic': java.nio.BufferUnderflowException
(kafka.server.ReplicaFetcherThread)
[2015-11-26 09:06:31,024] WARN [ReplicaFetcherThread-1-219], Error in fetch
kafka.server.ReplicaFetcherThread$FetchRequest@3fa05c3d. Possible cause:
org.apache.kafka.common.protocol.types.SchemaException: Error
reading field 'responses': Error reading field 'topic':
java.nio.BufferUnderflowException (kafka.server.ReplicaFetcherThread)
> Strange behaviour during partitions reassignment.
> -------------------------------------------------
>
> Key: KAFKA-2890
> URL: https://issues.apache.org/jira/browse/KAFKA-2890
> Project: Kafka
> Issue Type: Bug
> Affects Versions: 0.9.0.0
> Reporter: Alexander Kukushkin
> Attachments: reassign.json, reassign.out
>
>
> Hi.
> I am playing with the new version of kafka (0.9.0.0).
> Initially I've created cluster of 3 nodes, and created some topics there.
> Later I've added one more node and triggered partitions reassignment. It's
> kind of working, but on the new node in the log file there are strange
> warnings:
> [2015-11-25 14:06:52,998] WARN [ReplicaFetcherThread-1-152], Error in fetch
> kafka.server.ReplicaFetcherThread$FetchRequest@3f442c7b. Possible cause:
> org.apache.kafka.common.protocol.types.SchemaException: Error reading field
> 'responses': Error reading field 'topic': java.nio.BufferUnderflowException
> (kafka.server.ReplicaFetcherThread)
> I've found similar log messages in the following ticket:
> https://issues.apache.org/jira/browse/KAFKA-2756
> But there such messages were related to the replication between different
> versions (0.8 and 0.9).
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
