本文是以kafka0.8.2.2为例讲解
一，如何删除一个topic
删除一个topic有两个关键点：
1，配置删除参数
delete.topic.enable这个Broker参数配置为True。
2，执行
bin/kafka-topics.sh --zookeeper zk_host:port/chroot --delete --topic my_topic_name
假如不配置删除参数为true的话，topic其实并没有被清除，只是被标记为删除。此时，估计一般人的做法是删除topic在Zookeeper的信息和日志，其实这个操作并不会清除kafkaBroker内存的topic数据。所以，此时最佳的策略是配置删除参数为true然后，重启kafka。
二，重要的类介绍
1，PartitionStateMachine
该类代表分区的状态机。决定者分区的当前状态，和状态转移。四种状态
NonExistentPartition
NewPartition
OnlinePartition
OfflinePartition
2，ReplicaManager
负责管理当前机器的所有副本，处理读写、删除等具体动作。
读写：写获取partition对象，再获取Replica对象，再获取Log对象，采用其管理的Segment对象将数据写入、读出。
3，ReplicaStateMachine
副本的状态机。决定者副本的当前状态和状态之间的转移。一个副本总共可以处于一下几种状态的一种
NewReplica：Crontroller在分区重分配的时候可以创建一个新的副本。只能接受变为follower的请求。前状态可以是NonExistentReplica
OnlineReplica：新启动的分区，能接受变为leader或者follower请求。前状态可以是NewReplica, OnlineReplica or OfflineReplica
OfflineReplica：死亡的副本处于这种状态。前状态可以是NewReplica, OnlineReplica
ReplicaDeletionStarted：分本删除开始的时候处于这种状态，前状态是OfflineReplica
ReplicaDeletionSuccessful：副本删除成功。前状态是ReplicaDeletionStarted
ReplicaDeletionIneligible：删除失败的时候处于这种状态。前状态是ReplicaDeletionStarted
NonExistentReplica：副本成功删除之后处于这种状态，前状态是ReplicaDeletionSuccessful
4，TopicDeletionManager
该类管理着topic删除的状态机
1)，TopicCommand通过创建/admin/delete_topics/<topic>，来发布topic删除命令。
2)，Controller监听/admin/delete_topic子节点变动，开始分别删除topic
3)，Controller有个后台线程负责删除Topic
三，源码彻底解析topic的删除过程
此处会分四个部分：
A),客户端执行删除命令作用
B),不配置delete.topic.enable整个流水的源码
C),配置了delete.topic.enable整个流水的源码
D),手动删除zk上topic信息和磁盘数据
1，客户端执行删除命令
bin/kafka-topics.sh --zookeeper zk_host:port/chroot --delete --topic my_topic_name
进入kafka-topics.sh我们会看到
exec $(dirname $0)/kafka-run-class.sh kafka.admin.TopicCommand $@
进入TopicCommand里面，main方法里面
else if(opts.options.has(opts.deleteOpt))
deleteTopic(zkClient, opts)
实际内容是
val topics = getTopics(zkClient, opts)
if (topics.length == 0) {
println("Topic %s does not exist".format(opts.options.valueOf(opts.topicOpt)))
}
topics.foreach { topic =>
try {
ZkUtils.createPersistentPath(zkClient, ZkUtils.getDeleteTopicPath(topic))
在"/admin/delete_topics"目录下创建了一个topicName的节点。
2，假如不配置delete.topic.enable整个流水是
总共有两处listener会响应：
A),TopicChangeListener
B),DeleteTopicsListener
使用topic的删除命令删除一个topic的话，指挥触发DeleteTopicListener。
var topicsToBeDeleted = {
import JavaConversions._
(children: Buffer[String]).toSet
}
val nonExistentTopics = topicsToBeDeleted.filter(t => !controllerContext.allTopics.contains(t))
topicsToBeDeleted --= nonExistentTopics
if(topicsToBeDeleted.size > 0) {
info("Starting topic deletion for topics " + topicsToBeDeleted.mkString(","))
// mark topic ineligible for deletion if other state changes are in progress
topicsToBeDeleted.foreach { topic =>
val preferredReplicaElectionInProgress =
controllerContext.partitionsUndergoingPreferredReplicaElection.map(_.topic).contains(topic)
val partitionReassignmentInProgress =
controllerContext.partitionsBeingReassigned.keySet.map(_.topic).contains(topic)
if(preferredReplicaElectionInProgress || partitionReassignmentInProgress)
controller.deleteTopicManager.markTopicIneligibleForDeletion(Set(topic))
}
// add topic to deletion list
controller.deleteTopicManager.enqueueTopicsForDeletion(topicsToBeDeleted)
}
由于都会判断delete.topic.enable是否为true，假如不为true就不会执行，为true就进入执行
controller.deleteTopicManager.markTopicIneligibleForDeletion(Set(topic))
controller.deleteTopicManager.enqueueTopicsForDeletion(topicsToBeDeleted)
3，delete.topic.enable配置为true
此处与步骤2的区别，就是那两个处理函数。
controller.deleteTopicManager.markTopicIneligibleForDeletion(Set(topic))
controller.deleteTopicManager.enqueueTopicsForDeletion(topicsToBeDeleted)
markTopicIneligibleForDeletion函数的处理为
if(isDeleteTopicEnabled) {
val newTopicsToHaltDeletion = topicsToBeDeleted & topics
topicsIneligibleForDeletion ++= newTopicsToHaltDeletion
if(newTopicsToHaltDeletion.size > 0)
info("Halted deletion of topics %s".format(newTopicsToHaltDeletion.mkString(",")))
}
主要是停止删除topic，假如存储以下三种情况
* Halt delete topic if -
* 1. replicas being down
* 2. partition reassignment in progress for some partitions of the topic
* 3. preferred replica election in progress for some partitions of the topic
enqueueTopicsForDeletion主要作用是更新删除topic的集合，并激活TopicDeleteThread
def enqueueTopicsForDeletion(topics: Set[String]) {
if(isDeleteTopicEnabled) {
topicsToBeDeleted ++= topics
partitionsToBeDeleted ++= topics.flatMap(controllerContext.partitionsForTopic)
resumeTopicDeletionThread()
}
}
在删除线程DeleteTopicsThread的doWork方法中
topicsQueuedForDeletion.foreach { topic =>
// if all replicas are marked as deleted successfully, then topic deletion is done
if(controller.replicaStateMachine.areAllReplicasForTopicDeleted(topic)) {
// clear up all state for this topic from controller cache and zookeeper
completeDeleteTopic(topic)
info("Deletion of topic %s successfully completed".format(topic))
}
进入completeDeleteTopic方法中
// deregister partition change listener on the deleted topic. This is to prevent the partition change listener
// firing before the new topic listener when a deleted topic gets auto created
partitionStateMachine.deregisterPartitionChangeListener(topic)
val replicasForDeletedTopic = controller.replicaStateMachine.replicasInState(topic, ReplicaDeletionSuccessful)
// controller will remove this replica from the state machine as well as its partition assignment cache
replicaStateMachine.handleStateChanges(replicasForDeletedTopic, NonExistentReplica)
val partitionsForDeletedTopic = controllerContext.partitionsForTopic(topic)
// move respective partition to OfflinePartition and NonExistentPartition state
partitionStateMachine.handleStateChanges(partitionsForDeletedTopic, OfflinePartition)
partitionStateMachine.handleStateChanges(partitionsForDeletedTopic, NonExistentPartition)
topicsToBeDeleted -= topic
partitionsToBeDeleted.retain(_.topic != topic)
controllerContext.zkClient.deleteRecursive(ZkUtils.getTopicPath(topic))
controllerContext.zkClient.deleteRecursive(ZkUtils.getTopicConfigPath(topic))
controllerContext.zkClient.delete(ZkUtils.getDeleteTopicPath(topic))
controllerContext.removeTopic(topic)
主要作用是解除掉监控分区变动的listener，删除Zookeeper具体节点信息，删除磁盘数据，更新内存数据结构，比如从副本状态机里面移除分区的具体信息。
其实，最终要的是我们的副本磁盘数据是如何删除的。我们重点介绍这个部分。
首次清除的话，在删除线程DeleteTopicsThread的doWork方法中
{
// if you come here, then no replica is in TopicDeletionStarted and all replicas are not in
// TopicDeletionSuccessful. That means, that either given topic haven't initiated deletion
// or there is at least one failed replica (which means topic deletion should be retried).
if(controller.replicaStateMachine.isAnyReplicaInState(topic, ReplicaDeletionIneligible)) {
// mark topic for deletion retry
markTopicForDeletionRetry(topic)
}
进入markTopicForDeletionRetry
val failedReplicas = controller.replicaStateMachine.replicasInState(topic, ReplicaDeletionIneligible)
info("Retrying delete topic for topic %s since replicas %s were not successfully deleted"
.format(topic, failedReplicas.mkString(",")))
controller.replicaStateMachine.handleStateChanges(failedReplicas, OfflineReplica)
在ReplicaStateMachine的handleStateChanges方法中，调用了handleStateChange，处理OfflineReplica
// send stop replica command to the replica so that it stops fetching from the leader
brokerRequestBatch.addStopReplicaRequestForBrokers(List(replicaId), topic, partition, deletePartition = false)
接着在handleStateChanges中
brokerRequestBatch.sendRequestsToBrokers(controller.epoch, controllerContext.correlationId.getAndIncrement)
给副本数据存储节点发送StopReplicaKey副本指令，并开始删除数据
stopReplicaRequestMap foreach { case(broker, replicaInfoList) =>
val stopReplicaWithDelete = replicaInfoList.filter(p => p.deletePartition == true).map(i => i.replica).toSet
val stopReplicaWithoutDelete = replicaInfoList.filter(p => p.deletePartition == false).map(i => i.replica).toSet
debug("The stop replica request (delete = true) sent to broker %d is %s"
.format(broker, stopReplicaWithDelete.mkString(",")))
debug("The stop replica request (delete = false) sent to broker %d is %s"
.format(broker, stopReplicaWithoutDelete.mkString(",")))
replicaInfoList.foreach { r =>
val stopReplicaRequest = new StopReplicaRequest(r.deletePartition,
Set(TopicAndPartition(r.replica.topic, r.replica.partition)), controllerId, controllerEpoch, correlationId)
controller.sendRequest(broker, stopReplicaRequest, r.callback)
}
}
stopReplicaRequestMap.clear()
Broker的KafkaApis的Handle方法在接受到指令后
case RequestKeys.StopReplicaKey => handleStopReplicaRequest(request)
val (response, error) = replicaManager.stopReplicas(stopReplicaRequest)
接着是在stopReplicas方法中
{
controllerEpoch = stopReplicaRequest.controllerEpoch
// First stop fetchers for all partitions, then stop the corresponding replicas
replicaFetcherManager.removeFetcherForPartitions(stopReplicaRequest.partitions.map(r => TopicAndPartition(r.topic, r.partition)))
for(topicAndPartition <- stopReplicaRequest.partitions){
val errorCode = stopReplica(topicAndPartition.topic, topicAndPartition.partition, stopReplicaRequest.deletePartitions)
responseMap.put(topicAndPartition, errorCode)
}
(responseMap, ErrorMapping.NoError)
}
进一步进入stopReplica方法，正式进入日志删除
getPartition(topic, partitionId) match {
case Some(partition) =>
if(deletePartition) {
val removedPartition = allPartitions.remove((topic, partitionId))
if (removedPartition != null)
removedPartition.delete() // this will delete the local log
}
以上就是kafka的整个日志删除流水。
4，手动删除zk上topic信息和磁盘数据
TopicChangeListener会监听处理，但是处理很简单，只是更新了
val deletedTopics = controllerContext.allTopics -- currentChildren
controllerContext.allTopics = currentChildren
val addedPartitionReplicaAssignment = ZkUtils.getReplicaAssignmentForTopics(zkClient, newTopics.toSeq)
controllerContext.partitionReplicaAssignment = controllerContext.partitionReplicaAssignment.filter(p =>
四，总结
Kafka的topic的删除过程，实际上就是基于Zookeeper做了一个订阅发布系统。Zookeeper的客户端创建一个节点/admin/delete_topics/<topic>，由kafka Controller监听到事件之后正式触发topic的删除：解除Partition变更监听的listener，清除内存数据结构，删除副本数据，删除topic的相关Zookeeper节点。
delete.topic.enable配置该参数为false的情况下执行了topic的删除命令，实际上未做任何动作。我们此时要彻底删除topic建议修改该参数为true，重启kafka，这样topic信息会被彻底删除，已经测试。
一般流行的做法是手动删除Zookeeper的topic相关信息及磁盘数据但是这样的话会造成部分内存数据未清除。至于是否会有隐患，未测试。
