[
https://issues.apache.org/jira/browse/FLUME-2857?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15612227#comment-15612227
]
Denes Arvay commented on FLUME-2857:
------------------------------------
Actually it's not silent, the test fails due to the exception.
{noformat}
kafka.admin.AdminOperationException: replication factor: 1 larger than
available brokers: 0
at kafka.admin.AdminUtils$.assignReplicasToBrokers(AdminUtils.scala:77)
at kafka.admin.AdminUtils$.createTopic(AdminUtils.scala:236)
at kafka.admin.AdminUtils.createTopic(AdminUtils.scala)
at
org.apache.flume.source.kafka.KafkaSourceEmbeddedKafka.createTopic(KafkaSourceEmbeddedKafka.java:134)
at
org.apache.flume.source.kafka.TestKafkaSource.checkDefaultSettingsOnRestart(TestKafkaSource.java:645)
at
org.apache.flume.source.kafka.TestKafkaSource.testDefaultSettingsOnReConfigure(TestKafkaSource.java:632)
...
{noformat}
I agree with Donat, I also think that there is no need for starting the whole
kafka source, calling two {{configure}}s with different context should be
enough, as the context's properties are copied to the {{kafkaProps}} in
{{KafkaSource}}, which is checked by the test. We'd hit two birds with one
stone by removing the unneeded lines: the test would be simpler and the above
mentioned issue wouldn't come up.
> Kafka Source/Channel/Sink does not restore default values when live update
> config
> ---------------------------------------------------------------------------------
>
> Key: FLUME-2857
> URL: https://issues.apache.org/jira/browse/FLUME-2857
> Project: Flume
> Issue Type: Bug
> Components: Channel
> Reporter: Tristan Stevens
> Assignee: Tristan Stevens
> Attachments: FLUME-2857.patch
>
>
> Been using the following config:
> {noformat}
> tier1.channels.channel1.type = org.apache.flume.channel.kafka.KafkaChannel
> tier1.channels.channel1.capacity = 10000
> tier1.channels.channel1.transactionCapacity = 10000
> tier1.channels.channel1.brokerList =
> 10.0.0.64:9092,10.0.0.65:9092,10.0.0.66:9092
> tier1.channels.channel1.topic = flume.aggregator.channel
> tier1.channels.channel1.zookeeperConnect = 10.0.0.64:2181
> tier1.channels.channel1.kafka.producer.type=async
> tier1.channels.channel1.kafka.batch.num.messages=200
> {noformat}
> If I remove the producer.type and batch.num.messages (i.e. restore them to
> default values) when doing a live update config the new values do not take
> effect.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
