KafkaServer启动
在startup方法里，会实例化KafkaController
class KafkaServer {
def startup() {
.......
kafkaController = new KafkaController(config, zkUtils, time, metrics, threadNamePrefix)
kafkaController.startup()
......
}
}
KafkaController重要属性
- activeControllerId：获取controller的主broker的id
- eventManager: ControllerEventManager实例，负责处理事件
ControllerEventManager
ControllerEventManager实例化一个线程，用来单独处理KafkaController发送的事件。
class ControllerEventManager(rateAndTimeMetrics: Map[ControllerState, KafkaTimer],
eventProcessedListener: ControllerEvent => Unit) {
// eventProcessedListener是event的回调函数
@volatile private var _state: ControllerState = ControllerState.Idle
// event队列
private val queue = new LinkedBlockingQueue[ControllerEvent]
private val thread = new ControllerEventThread("controller-event-thread")
def state: ControllerState = _state
def start(): Unit = thread.start()
def close(): Unit = thread.shutdown()
// KafkaController调用put方法，添加事件
def put(event: ControllerEvent): Unit = queue.put(event)
class ControllerEventThread(name: String) extends ShutdownableThread(name = name) {
override def doWork(): Unit = {
// 从queue取出event
val controllerEvent = queue.take()
// 更新状态为event的状态
_state = controllerEvent.state
try {
// event自身包含process方法，定义了如何处理event
rateAndTimeMetrics(state).time {
controllerEvent.process()
}
} catch {
case e: Throwable => error(s"Error processing event $controllerEvent", e)
}
// 回调
try eventProcessedListener(controllerEvent)
catch {
case e: Throwable => error(s"Error while invoking listener for processed event $controllerEvent", e)
}
// 更新状态为Idle
_state = ControllerState.Idle
}
}
}
事件基类
state表示处理event时，controller的状态。
process表示处理event的程序
sealed trait ControllerEvent {
def state: ControllerState
def process(): Unit
}
ControllerState
ControllerState目前有10种
object ControllerState {
case object Idle extends ControllerState {
def value = 0
override protected def hasRateAndTimeMetric: Boolean = false
}
case object ControllerChange extends ControllerState {
def value = 1
}
case object BrokerChange extends ControllerState {
def value = 2
override def rateAndTimeMetricName = Some("LeaderElectionRateAndTimeMs")
}
case object TopicChange extends ControllerState {
def value = 3
}
case object TopicDeletion extends ControllerState {
def value = 4
}
case object PartitionReassignment extends ControllerState {
def value = 5
}
case object AutoLeaderBalance extends ControllerState {
def value = 6
}
case object ManualLeaderBalance extends ControllerState {
def value = 7
}
case object ControlledShutdown extends ControllerState {
def value = 8
}
case object IsrChange extends ControllerState {
def value = 9
}
Startup事件
KafkaController在启动的时候，会生成Startup事件，添加到EventManage
class KafkaController {
def startup() = {
eventManager.put(Startup)
eventManager.start()
}
}
Startup定义
case object Startup extends ControllerEvent {
def state = ControllerState.ControllerChange
override def process(): Unit = {
// 注册zookeeper的连接事件
registerSessionExpirationListener()
// 注册"/controller"节点的数据变化事件(节点存储主controller的id，数据变化意味着主controller发送变化)
registerControllerChangeListener()
// 选举，尝试争取controller
elect()
}
private def registerSessionExpirationListener() = {
zkUtils.zkClient.subscribeStateChanges(new SessionExpirationListener(this, eventManager))
}
class SessionExpirationListener(controller: KafkaController, eventManager: ControllerEventManager) extends IZkStateListener with Logging {
@throws[Exception]
override def handleNewSession(): Unit = {
// 当重新连接到zookeeper时，发送Reelect事件
eventManager.put(controller.Reelect)
}
}
private def registerControllerChangeListener() = {
zkUtils.zkClient.subscribeDataChanges(ZkUtils.ControllerPath, new ControllerChangeListener(this, eventManager))
}
}
#Reelect事件 Reelect表明竞选主controller
case object Reelect extends ControllerEvent {
def state = ControllerState.ControllerChange
override def process(): Unit = {
val wasActiveBeforeChange = isActive
// 获取主controller的id
activeControllerId = getControllerID()
// 先前是主controller，现在是从controller
if (wasActiveBeforeChange && !isActive) {
// 关闭先前zookeeper事件的监听
onControllerResignation()
}
// 尝试竞选主controller
elect()
}
}
def elect(): Unit = {
val timestamp = time.milliseconds
// 节点数据，为brokerId + timestamp
val electString = ZkUtils.controllerZkData(config.brokerId, timestamp)
activeControllerId = getControllerID()
// activeControllerId为-1，表示主controller还没有
if (activeControllerId != -1) {
debug("Broker %d has been elected as the controller, so stopping the election process.".format(activeControllerId))
return
}
try {
// 竞争创建临时节点。如果未成功，会抛出ZkNodeExistsException异常
val zkCheckedEphemeral = new ZKCheckedEphemeral(ZkUtils.ControllerPath,
electString,
controllerContext.zkUtils.zkConnection.getZookeeper,
controllerContext.zkUtils.isSecure)
zkCheckedEphemeral.create()
info(config.brokerId + " successfully elected as the controller")
activeControllerId = config.brokerId
// 注册监听事件，这些都是主controller负责
onControllerFailover()
} catch {
case _: ZkNodeExistsException =>
activeControllerId = getControllerID
if (activeControllerId != -1)
debug("Broker %d was elected as controller instead of broker %d".format(activeControllerId, config.brokerId))
else
warn("A controller has been elected but just resigned, this will result in another round of election")
case e2: Throwable =>
error("Error while electing or becoming controller on broker %d".format(config.brokerId), e2)
triggerControllerMove()
}
}
def onControllerFailover() {
info("Broker %d starting become controller state transition".format(config.brokerId))
readControllerEpochFromZookeeper()
incrementControllerEpoch()
// 分区分配事件
registerPartitionReassignmentListener()
// 通知事件
registerIsrChangeNotificationListener()
// replica选举事件
registerPreferredReplicaElectionListener()
// topic变化事件
registerTopicChangeListener()
// topic删除事件
registerTopicDeletionListener()
// broker事件
registerBrokerChangeListener()
// 初始化context
initializeControllerContext()
val (topicsToBeDeleted, topicsIneligibleForDeletion) = fetchTopicDeletionsInProgress()
topicDeletionManager.init(topicsToBeDeleted, topicsIneligibleForDeletion)
sendUpdateMetadataRequest(controllerContext.liveOrShuttingDownBrokerIds.toSeq)
// 启动副本状态机
replicaStateMachine.startup()
// 启动分片状态机
partitionStateMachine.startup()
// 监听partition改变事件
controllerContext.allTopics.foreach(topic => registerPartitionModificationsListener(topic))
info("Broker %d is ready to serve as the new controller with epoch %d".format(config.brokerId, epoch))
maybeTriggerPartitionReassignment()
topicDeletionManager.tryTopicDeletion()
val pendingPreferredReplicaElections = fetchPendingPreferredReplicaElections()
onPreferredReplicaElection(pendingPreferredReplicaElections)
info("starting the controller scheduler")
kafkaScheduler.startup()
if (config.autoLeaderRebalanceEnable) {
scheduleAutoLeaderRebalanceTask(delay = 5, unit = TimeUnit.SECONDS)
}
}
概括
KafkaController负责集群中的一些事件，比如topic的新增和删除。
KafkaController在kafka集群中，扮演着很重要的角色。所以为了保证controller的高可用，集群的每个节点，都会运行KafkaController服务，通过zookeeper的选举，保证任何时刻只有一个主的controller。
KafkaController在启动的时候，就会尝试竞争主controller。如果竞选成功，就会在zookeeper中注册监听事件。