MINOR: Tweak upgrade note on KIP-62 to include request.timeout.ms

Author: Jason Gustafson <jason@confluent.io>

Reviewers: Ismael Juma <ismael@juma.me.uk>

Closes #1960 from hachikuji/add-note-on-request-timeout
This commit is contained in:
Jason Gustafson 2016-10-04 10:38:49 -07:00
parent 1a396c9cf2
commit d8d2f3aac4
1 changed files with 7 additions and 1 deletions

View File

@ -64,7 +64,13 @@ Note: Because new protocols are introduced, it is important to upgrade your Kafk
<li> Kafka clusters can now be uniquely identified by a cluster id. It will be automatically generated when a broker is upgraded to 0.10.1.0. The cluster id is available via the kafka.server:type=KafkaServer,name=ClusterId metric and it is part of the Metadata response. Serializers, client interceptors and metric reporters can receive the cluster id by implementing the ClusterResourceListener interface. </li> <li> Kafka clusters can now be uniquely identified by a cluster id. It will be automatically generated when a broker is upgraded to 0.10.1.0. The cluster id is available via the kafka.server:type=KafkaServer,name=ClusterId metric and it is part of the Metadata response. Serializers, client interceptors and metric reporters can receive the cluster id by implementing the ClusterResourceListener interface. </li>
<li> The BrokerState "RunningAsController" (value 4) has been removed. Due to a bug, a broker would only be in this state briefly before transitioning out of it and hence the impact of the removal should be minimal. The recommended way to detect if a given broker is the controller is via the kafka.controller:type=KafkaController,name=ActiveControllerCount metric. </li> <li> The BrokerState "RunningAsController" (value 4) has been removed. Due to a bug, a broker would only be in this state briefly before transitioning out of it and hence the impact of the removal should be minimal. The recommended way to detect if a given broker is the controller is via the kafka.controller:type=KafkaController,name=ActiveControllerCount metric. </li>
<li> The new Java Consumer now allows users to search offsets by timestamp on partitions. </li> <li> The new Java Consumer now allows users to search offsets by timestamp on partitions. </li>
<li> The new Java Consumer now supports heartbeating from a background thread. There is a new configuration <code>max.poll.interval.ms</code> which controls the maximum time between poll invocations before the consumer will proactively leave the group (5 minutes by default). The default value of <code>session.timeout.ms</code> has been adjusted down to 10 seconds, and the default value of <code>max.poll.records</code> has been changed to 500.</li> <li> The new Java Consumer now supports heartbeating from a background thread. There is a new configuration
<code>max.poll.interval.ms</code> which controls the maximum time between poll invocations before the consumer
will proactively leave the group (5 minutes by default). The value of the configuration
<code>request.timeout.ms</code> must always be larger than <code>max.poll.interval.ms</code> because this is the maximum
time that a JoinGroup request can block on the server while the consumer is rebalancing, so we have changed its default
value to just above 5 minutes. Finally, the default value of <code>session.timeout.ms</code> has been adjusted down to
10 seconds, and the default value of <code>max.poll.records</code> has been changed to 500.</li>
<li> When using an Authorizer and a user doesn't have <b>Describe</b> authorization on a topic, the broker will no <li> When using an Authorizer and a user doesn't have <b>Describe</b> authorization on a topic, the broker will no
longer return TOPIC_AUTHORIZATION_FAILED errors to requests since this leaks topic names. Instead, the UNKNOWN_TOPIC_OR_PARTITION longer return TOPIC_AUTHORIZATION_FAILED errors to requests since this leaks topic names. Instead, the UNKNOWN_TOPIC_OR_PARTITION
error code will be returned. This may cause unexpected timeouts or delays when using the producer and consumer since error code will be returned. This may cause unexpected timeouts or delays when using the producer and consumer since