Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance. Project price only 1 $
You can buy this project and download/modify it how often you want.
/**
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kafka.log
import java.io._
import java.nio.file.Files
import java.util.concurrent._
import com.yammer.metrics.core.Gauge
import kafka.metrics.KafkaMetricsGroup
import kafka.server.checkpoints.OffsetCheckpointFile
import kafka.server.{BrokerState, RecoveringFromUncleanShutdown, _}
import kafka.utils._
import kafka.zk.KafkaZkClient
import org.apache.kafka.common.{KafkaException, TopicPartition}
import org.apache.kafka.common.utils.Time
import org.apache.kafka.common.errors.{KafkaStorageException, LogDirNotFoundException}
import scala.collection.JavaConverters._
import scala.collection._
import scala.collection.mutable.ArrayBuffer
import scala.util.{Failure, Success, Try}
/**
* The entry point to the kafka log management subsystem. The log manager is responsible for log creation, retrieval, and cleaning.
* All read and write operations are delegated to the individual log instances.
*
* The log manager maintains logs in one or more directories. New logs are created in the data directory
* with the fewest logs. No attempt is made to move partitions after the fact or balance based on
* size or I/O rate.
*
* A background thread handles log retention by periodically truncating excess log segments.
*/
@threadsafe
class LogManager(logDirs: Seq[File],
initialOfflineDirs: Seq[File],
val topicConfigs: Map[String, LogConfig], // note that this doesn't get updated after creation
val initialDefaultConfig: LogConfig,
val cleanerConfig: CleanerConfig,
recoveryThreadsPerDataDir: Int,
val flushCheckMs: Long,
val flushRecoveryOffsetCheckpointMs: Long,
val flushStartOffsetCheckpointMs: Long,
val retentionCheckMs: Long,
val maxPidExpirationMs: Int,
scheduler: Scheduler,
val brokerState: BrokerState,
brokerTopicStats: BrokerTopicStats,
logDirFailureChannel: LogDirFailureChannel,
time: Time) extends Logging with KafkaMetricsGroup {
import LogManager._
val LockFile = ".lock"
val InitialTaskDelayMs = 30 * 1000
private val logCreationOrDeletionLock = new Object
private val currentLogs = new Pool[TopicPartition, Log]()
// Future logs are put in the directory with "-future" suffix. Future log is created when user wants to move replica
// from one log directory to another log directory on the same broker. The directory of the future log will be renamed
// to replace the current log of the partition after the future log catches up with the current log
private val futureLogs = new Pool[TopicPartition, Log]()
// Each element in the queue contains the log object to be deleted and the time it is scheduled for deletion.
private val logsToBeDeleted = new LinkedBlockingQueue[(Log, Long)]()
private val _liveLogDirs: ConcurrentLinkedQueue[File] = createAndValidateLogDirs(logDirs, initialOfflineDirs)
@volatile private var _currentDefaultConfig = initialDefaultConfig
@volatile private var numRecoveryThreadsPerDataDir = recoveryThreadsPerDataDir
// This map contains all partitions whose logs are getting loaded and initialized. If log configuration
// of these partitions get updated at the same time, the corresponding entry in this map is set to "true",
// which triggers a config reload after initialization is finished (to get the latest config value).
// See KAFKA-8813 for more detail on the race condition
// Visible for testing
private[log] val partitionsInitializing = new ConcurrentHashMap[TopicPartition, Boolean]().asScala
def reconfigureDefaultLogConfig(logConfig: LogConfig): Unit = {
this._currentDefaultConfig = logConfig
}
def currentDefaultConfig: LogConfig = _currentDefaultConfig
def liveLogDirs: Seq[File] = {
if (_liveLogDirs.size == logDirs.size)
logDirs
else
_liveLogDirs.asScala.toBuffer
}
private val dirLocks = lockLogDirs(liveLogDirs)
@volatile private var recoveryPointCheckpoints = liveLogDirs.map(dir =>
(dir, new OffsetCheckpointFile(new File(dir, RecoveryPointCheckpointFile), logDirFailureChannel))).toMap
@volatile private var logStartOffsetCheckpoints = liveLogDirs.map(dir =>
(dir, new OffsetCheckpointFile(new File(dir, LogStartOffsetCheckpointFile), logDirFailureChannel))).toMap
private val preferredLogDirs = new ConcurrentHashMap[TopicPartition, String]()
private def offlineLogDirs: Iterable[File] = {
val logDirsSet = mutable.Set[File]() ++= logDirs
_liveLogDirs.asScala.foreach(logDirsSet -=)
logDirsSet
}
loadLogs()
// public, so we can access this from kafka.admin.DeleteTopicTest
val cleaner: LogCleaner =
if(cleanerConfig.enableCleaner)
new LogCleaner(cleanerConfig, liveLogDirs, currentLogs, logDirFailureChannel, time = time)
else
null
val offlineLogDirectoryCount = newGauge(
"OfflineLogDirectoryCount",
new Gauge[Int] {
def value = offlineLogDirs.size
}
)
for (dir <- logDirs) {
newGauge(
"LogDirectoryOffline",
new Gauge[Int] {
def value = if (_liveLogDirs.contains(dir)) 0 else 1
},
Map("logDirectory" -> dir.getAbsolutePath)
)
}
/**
* Create and check validity of the given directories that are not in the given offline directories, specifically:
*
*
Ensure that there are no duplicates in the directory list
*
Create each directory if it doesn't exist
*
Check that each path is a readable directory
*
*/
private def createAndValidateLogDirs(dirs: Seq[File], initialOfflineDirs: Seq[File]): ConcurrentLinkedQueue[File] = {
val liveLogDirs = new ConcurrentLinkedQueue[File]()
val canonicalPaths = mutable.HashSet.empty[String]
for (dir <- dirs) {
try {
if (initialOfflineDirs.contains(dir))
throw new IOException(s"Failed to load ${dir.getAbsolutePath} during broker startup")
if (!dir.exists) {
info(s"Log directory ${dir.getAbsolutePath} not found, creating it.")
val created = dir.mkdirs()
if (!created)
throw new IOException(s"Failed to create data directory ${dir.getAbsolutePath}")
}
if (!dir.isDirectory || !dir.canRead)
throw new IOException(s"${dir.getAbsolutePath} is not a readable log directory.")
// getCanonicalPath() throws IOException if a file system query fails or if the path is invalid (e.g. contains
// the Nul character). Since there's no easy way to distinguish between the two cases, we treat them the same
// and mark the log directory as offline.
if (!canonicalPaths.add(dir.getCanonicalPath))
throw new KafkaException(s"Duplicate log directory found: ${dirs.mkString(", ")}")
liveLogDirs.add(dir)
} catch {
case e: IOException =>
logDirFailureChannel.maybeAddOfflineLogDir(dir.getAbsolutePath, s"Failed to create or validate data directory ${dir.getAbsolutePath}", e)
}
}
if (liveLogDirs.isEmpty) {
fatal(s"Shutdown broker because none of the specified log dirs from ${dirs.mkString(", ")} can be created or validated")
Exit.halt(1)
}
liveLogDirs
}
def resizeRecoveryThreadPool(newSize: Int): Unit = {
info(s"Resizing recovery thread pool size for each data dir from $numRecoveryThreadsPerDataDir to $newSize")
numRecoveryThreadsPerDataDir = newSize
}
// dir should be an absolute path
def handleLogDirFailure(dir: String): Unit = {
info(s"Stopping serving logs in dir $dir")
logCreationOrDeletionLock synchronized {
_liveLogDirs.remove(new File(dir))
if (_liveLogDirs.isEmpty) {
fatal(s"Shutdown broker because all log dirs in ${logDirs.mkString(", ")} have failed")
Exit.halt(1)
}
recoveryPointCheckpoints = recoveryPointCheckpoints.filter { case (file, _) => file.getAbsolutePath != dir }
logStartOffsetCheckpoints = logStartOffsetCheckpoints.filter { case (file, _) => file.getAbsolutePath != dir }
if (cleaner != null)
cleaner.handleLogDirFailure(dir)
val offlineCurrentTopicPartitions = currentLogs.collect {
case (tp, log) if log.dir.getParent == dir => tp
}
offlineCurrentTopicPartitions.foreach { topicPartition => {
val removedLog = currentLogs.remove(topicPartition)
if (removedLog != null) {
removedLog.closeHandlers()
removedLog.removeLogMetrics()
}
}}
val offlineFutureTopicPartitions = futureLogs.collect {
case (tp, log) if log.dir.getParent == dir => tp
}
offlineFutureTopicPartitions.foreach { topicPartition => {
val removedLog = futureLogs.remove(topicPartition)
if (removedLog != null) {
removedLog.closeHandlers()
removedLog.removeLogMetrics()
}
}}
info(s"Logs for partitions ${offlineCurrentTopicPartitions.mkString(",")} are offline and " +
s"logs for future partitions ${offlineFutureTopicPartitions.mkString(",")} are offline due to failure on log directory $dir")
dirLocks.filter(_.file.getParent == dir).foreach(dir => CoreUtils.swallow(dir.destroy(), this))
}
}
/**
* Lock all the given directories
*/
private def lockLogDirs(dirs: Seq[File]): Seq[FileLock] = {
dirs.flatMap { dir =>
try {
val lock = new FileLock(new File(dir, LockFile))
if (!lock.tryLock())
throw new KafkaException("Failed to acquire lock on file .lock in " + lock.file.getParent +
". A Kafka instance in another process or thread is using this directory.")
Some(lock)
} catch {
case e: IOException =>
logDirFailureChannel.maybeAddOfflineLogDir(dir.getAbsolutePath, s"Disk error while locking directory $dir", e)
None
}
}
}
private def addLogToBeDeleted(log: Log): Unit = {
this.logsToBeDeleted.add((log, time.milliseconds()))
}
// Only for testing
private[log] def hasLogsToBeDeleted: Boolean = !logsToBeDeleted.isEmpty
private def loadLog(logDir: File, recoveryPoints: Map[TopicPartition, Long], logStartOffsets: Map[TopicPartition, Long]): Unit = {
debug(s"Loading log '${logDir.getName}'")
val topicPartition = Log.parseTopicPartitionName(logDir)
val config = topicConfigs.getOrElse(topicPartition.topic, currentDefaultConfig)
val logRecoveryPoint = recoveryPoints.getOrElse(topicPartition, 0L)
val logStartOffset = logStartOffsets.getOrElse(topicPartition, 0L)
val log = Log(
dir = logDir,
config = config,
logStartOffset = logStartOffset,
recoveryPoint = logRecoveryPoint,
maxProducerIdExpirationMs = maxPidExpirationMs,
producerIdExpirationCheckIntervalMs = LogManager.ProducerIdExpirationCheckIntervalMs,
scheduler = scheduler,
time = time,
brokerTopicStats = brokerTopicStats,
logDirFailureChannel = logDirFailureChannel)
if (logDir.getName.endsWith(Log.DeleteDirSuffix)) {
addLogToBeDeleted(log)
} else {
val previous = {
if (log.isFuture)
this.futureLogs.put(topicPartition, log)
else
this.currentLogs.put(topicPartition, log)
}
if (previous != null) {
if (log.isFuture)
throw new IllegalStateException("Duplicate log directories found: %s, %s!".format(log.dir.getAbsolutePath, previous.dir.getAbsolutePath))
else
throw new IllegalStateException(s"Duplicate log directories for $topicPartition are found in both ${log.dir.getAbsolutePath} " +
s"and ${previous.dir.getAbsolutePath}. It is likely because log directory failure happened while broker was " +
s"replacing current replica with future replica. Recover broker from this failure by manually deleting one of the two directories " +
s"for this partition. It is recommended to delete the partition in the log directory that is known to have failed recently.")
}
}
}
/**
* Recover and load all logs in the given data directories
*/
private def loadLogs(): Unit = {
info("Loading logs.")
val startMs = time.milliseconds
val threadPools = ArrayBuffer.empty[ExecutorService]
val offlineDirs = mutable.Set.empty[(String, IOException)]
val jobs = mutable.Map.empty[File, Seq[Future[_]]]
for (dir <- liveLogDirs) {
try {
val pool = Executors.newFixedThreadPool(numRecoveryThreadsPerDataDir)
threadPools.append(pool)
val cleanShutdownFile = new File(dir, Log.CleanShutdownFile)
if (cleanShutdownFile.exists) {
debug(s"Found clean shutdown file. Skipping recovery for all logs in data directory: ${dir.getAbsolutePath}")
} else {
// log recovery itself is being performed by `Log` class during initialization
brokerState.newState(RecoveringFromUncleanShutdown)
}
var recoveryPoints = Map[TopicPartition, Long]()
try {
recoveryPoints = this.recoveryPointCheckpoints(dir).read
} catch {
case e: Exception =>
warn(s"Error occurred while reading recovery-point-offset-checkpoint file of directory $dir", e)
warn("Resetting the recovery checkpoint to 0")
}
var logStartOffsets = Map[TopicPartition, Long]()
try {
logStartOffsets = this.logStartOffsetCheckpoints(dir).read
} catch {
case e: Exception =>
warn(s"Error occurred while reading log-start-offset-checkpoint file of directory $dir", e)
}
val jobsForDir = for {
dirContent <- Option(dir.listFiles).toList
logDir <- dirContent if logDir.isDirectory
} yield {
CoreUtils.runnable {
try {
loadLog(logDir, recoveryPoints, logStartOffsets)
} catch {
case e: IOException =>
offlineDirs.add((dir.getAbsolutePath, e))
error(s"Error while loading log dir ${dir.getAbsolutePath}", e)
}
}
}
jobs(cleanShutdownFile) = jobsForDir.map(pool.submit)
} catch {
case e: IOException =>
offlineDirs.add((dir.getAbsolutePath, e))
error(s"Error while loading log dir ${dir.getAbsolutePath}", e)
}
}
try {
for ((cleanShutdownFile, dirJobs) <- jobs) {
dirJobs.foreach(_.get)
try {
cleanShutdownFile.delete()
} catch {
case e: IOException =>
offlineDirs.add((cleanShutdownFile.getParent, e))
error(s"Error while deleting the clean shutdown file $cleanShutdownFile", e)
}
}
offlineDirs.foreach { case (dir, e) =>
logDirFailureChannel.maybeAddOfflineLogDir(dir, s"Error while deleting the clean shutdown file in dir $dir", e)
}
} catch {
case e: ExecutionException =>
error(s"There was an error in one of the threads during logs loading: ${e.getCause}")
throw e.getCause
} finally {
threadPools.foreach(_.shutdown())
}
info(s"Logs loading complete in ${time.milliseconds - startMs} ms.")
}
/**
* Start the background threads to flush logs and do log cleanup
*/
def startup(): Unit = {
/* Schedule the cleanup task to delete old logs */
if (scheduler != null) {
info("Starting log cleanup with a period of %d ms.".format(retentionCheckMs))
scheduler.schedule("kafka-log-retention",
cleanupLogs _,
delay = InitialTaskDelayMs,
period = retentionCheckMs,
TimeUnit.MILLISECONDS)
info("Starting log flusher with a default period of %d ms.".format(flushCheckMs))
scheduler.schedule("kafka-log-flusher",
flushDirtyLogs _,
delay = InitialTaskDelayMs,
period = flushCheckMs,
TimeUnit.MILLISECONDS)
scheduler.schedule("kafka-recovery-point-checkpoint",
checkpointLogRecoveryOffsets _,
delay = InitialTaskDelayMs,
period = flushRecoveryOffsetCheckpointMs,
TimeUnit.MILLISECONDS)
scheduler.schedule("kafka-log-start-offset-checkpoint",
checkpointLogStartOffsets _,
delay = InitialTaskDelayMs,
period = flushStartOffsetCheckpointMs,
TimeUnit.MILLISECONDS)
scheduler.schedule("kafka-delete-logs", // will be rescheduled after each delete logs with a dynamic period
deleteLogs _,
delay = InitialTaskDelayMs,
unit = TimeUnit.MILLISECONDS)
}
if (cleanerConfig.enableCleaner)
cleaner.startup()
}
/**
* Close all the logs
*/
def shutdown(): Unit = {
info("Shutting down.")
removeMetric("OfflineLogDirectoryCount")
for (dir <- logDirs) {
removeMetric("LogDirectoryOffline", Map("logDirectory" -> dir.getAbsolutePath))
}
val threadPools = ArrayBuffer.empty[ExecutorService]
val jobs = mutable.Map.empty[File, Seq[Future[_]]]
// stop the cleaner first
if (cleaner != null) {
CoreUtils.swallow(cleaner.shutdown(), this)
}
val localLogsByDir = logsByDir
// close logs in each dir
for (dir <- liveLogDirs) {
debug(s"Flushing and closing logs at $dir")
val pool = Executors.newFixedThreadPool(numRecoveryThreadsPerDataDir)
threadPools.append(pool)
val logsInDir = localLogsByDir.getOrElse(dir.toString, Map()).values
val jobsForDir = logsInDir map { log =>
CoreUtils.runnable {
// flush the log to ensure latest possible recovery point
log.flush()
log.close()
}
}
jobs(dir) = jobsForDir.map(pool.submit).toSeq
}
try {
for ((dir, dirJobs) <- jobs) {
dirJobs.foreach(_.get)
// update the last flush point
debug(s"Updating recovery points at $dir")
checkpointRecoveryOffsetsAndCleanSnapshot(dir, localLogsByDir.getOrElse(dir.toString, Map()).values.toSeq)
debug(s"Updating log start offsets at $dir")
checkpointLogStartOffsetsInDir(dir)
// mark that the shutdown was clean by creating marker file
debug(s"Writing clean shutdown marker at $dir")
CoreUtils.swallow(Files.createFile(new File(dir, Log.CleanShutdownFile).toPath), this)
}
} catch {
case e: ExecutionException =>
error(s"There was an error in one of the threads during LogManager shutdown: ${e.getCause}")
throw e.getCause
} finally {
threadPools.foreach(_.shutdown())
// regardless of whether the close succeeded, we need to unlock the data directories
dirLocks.foreach(_.destroy())
}
info("Shutdown complete.")
}
/**
* Truncate the partition logs to the specified offsets and checkpoint the recovery point to this offset
*
* @param partitionOffsets Partition logs that need to be truncated
* @param isFuture True iff the truncation should be performed on the future log of the specified partitions
*/
def truncateTo(partitionOffsets: Map[TopicPartition, Long], isFuture: Boolean): Unit = {
val affectedLogs = ArrayBuffer.empty[Log]
for ((topicPartition, truncateOffset) <- partitionOffsets) {
val log = {
if (isFuture)
futureLogs.get(topicPartition)
else
currentLogs.get(topicPartition)
}
// If the log does not exist, skip it
if (log != null) {
//May need to abort and pause the cleaning of the log, and resume after truncation is done.
val needToStopCleaner = cleaner != null && truncateOffset < log.activeSegment.baseOffset
if (needToStopCleaner && !isFuture)
cleaner.abortAndPauseCleaning(topicPartition)
try {
if (log.truncateTo(truncateOffset))
affectedLogs += log
if (needToStopCleaner && !isFuture)
cleaner.maybeTruncateCheckpoint(log.dir.getParentFile, topicPartition, log.activeSegment.baseOffset)
} finally {
if (needToStopCleaner && !isFuture) {
cleaner.resumeCleaning(Seq(topicPartition))
info(s"Compaction for partition $topicPartition is resumed")
}
}
}
}
for ((dir, logs) <- affectedLogs.groupBy(_.dir.getParentFile)) {
checkpointRecoveryOffsetsAndCleanSnapshot(dir, logs)
}
}
/**
* Delete all data in a partition and start the log at the new offset
*
* @param topicPartition The partition whose log needs to be truncated
* @param newOffset The new offset to start the log with
* @param isFuture True iff the truncation should be performed on the future log of the specified partition
*/
def truncateFullyAndStartAt(topicPartition: TopicPartition, newOffset: Long, isFuture: Boolean): Unit = {
val log = {
if (isFuture)
futureLogs.get(topicPartition)
else
currentLogs.get(topicPartition)
}
// If the log does not exist, skip it
if (log != null) {
//Abort and pause the cleaning of the log, and resume after truncation is done.
if (cleaner != null && !isFuture)
cleaner.abortAndPauseCleaning(topicPartition)
try {
log.truncateFullyAndStartAt(newOffset)
if (cleaner != null && !isFuture) {
cleaner.maybeTruncateCheckpoint(log.dir.getParentFile, topicPartition, log.activeSegment.baseOffset)
}
} finally {
if (cleaner != null && !isFuture) {
cleaner.resumeCleaning(Seq(topicPartition))
info(s"Compaction for partition $topicPartition is resumed")
}
}
checkpointRecoveryOffsetsAndCleanSnapshot(log.dir.getParentFile, Seq(log))
}
}
/**
* Write out the current recovery point for all logs to a text file in the log directory
* to avoid recovering the whole log on startup.
*/
def checkpointLogRecoveryOffsets(): Unit = {
logsByDir.foreach { case (dir, partitionToLogMap) =>
liveLogDirs.find(_.getAbsolutePath.equals(dir)).foreach { f =>
checkpointRecoveryOffsetsAndCleanSnapshot(f, partitionToLogMap.values.toSeq)
}
}
}
/**
* Write out the current log start offset for all logs to a text file in the log directory
* to avoid exposing data that have been deleted by DeleteRecordsRequest
*/
def checkpointLogStartOffsets(): Unit = {
liveLogDirs.foreach(checkpointLogStartOffsetsInDir)
}
/**
* Write the recovery checkpoint file for all logs in provided directory and clean older snapshots for provided logs.
*
* @param dir the directory in which logs are checkpointed
* @param logsToCleanSnapshot logs whose snapshots need to be cleaned
*/
// Only for testing
private[log] def checkpointRecoveryOffsetsAndCleanSnapshot(dir: File, logsToCleanSnapshot: Seq[Log]): Unit = {
try {
checkpointLogRecoveryOffsetsInDir(dir)
logsToCleanSnapshot.foreach(_.deleteSnapshotsAfterRecoveryPointCheckpoint())
} catch {
case e: IOException =>
logDirFailureChannel.maybeAddOfflineLogDir(dir.getAbsolutePath, s"Disk error while writing to recovery point " +
s"file in directory $dir", e)
}
}
private def checkpointLogRecoveryOffsetsInDir(dir: File): Unit = {
for {
partitionToLog <- logsByDir.get(dir.getAbsolutePath)
checkpoint <- recoveryPointCheckpoints.get(dir)
} {
checkpoint.write(partitionToLog.map { case (tp, log) => tp -> log.recoveryPoint })
}
}
/**
* Checkpoint log start offset for all logs in provided directory.
*/
private def checkpointLogStartOffsetsInDir(dir: File): Unit = {
for {
partitionToLog <- logsByDir.get(dir.getAbsolutePath)
checkpoint <- logStartOffsetCheckpoints.get(dir)
} {
try {
val logStartOffsets = partitionToLog.collect {
case (k, log) if log.logStartOffset > log.logSegments.head.baseOffset => k -> log.logStartOffset
}
checkpoint.write(logStartOffsets)
} catch {
case e: IOException =>
logDirFailureChannel.maybeAddOfflineLogDir(dir.getAbsolutePath, s"Disk error while writing to logStartOffset file in directory $dir", e)
}
}
}
// The logDir should be an absolute path
def maybeUpdatePreferredLogDir(topicPartition: TopicPartition, logDir: String): Unit = {
// Do not cache the preferred log directory if either the current log or the future log for this partition exists in the specified logDir
if (!getLog(topicPartition).exists(_.dir.getParent == logDir) &&
!getLog(topicPartition, isFuture = true).exists(_.dir.getParent == logDir))
preferredLogDirs.put(topicPartition, logDir)
}
def abortAndPauseCleaning(topicPartition: TopicPartition): Unit = {
if (cleaner != null)
cleaner.abortAndPauseCleaning(topicPartition)
}
/**
* Get the log if it exists, otherwise return None
*
* @param topicPartition the partition of the log
* @param isFuture True iff the future log of the specified partition should be returned
*/
def getLog(topicPartition: TopicPartition, isFuture: Boolean = false): Option[Log] = {
if (isFuture)
Option(futureLogs.get(topicPartition))
else
Option(currentLogs.get(topicPartition))
}
/**
* Method to indicate that logs are getting initialized for the partition passed in as argument.
* This method should always be followed by [[kafka.log.LogManager#finishedInitializingLog]] to indicate that log
* initialization is done.
*/
def initializingLog(topicPartition: TopicPartition): Unit = {
partitionsInitializing(topicPartition) = false
}
/**
* Mark the partition configuration for all partitions that are getting initialized for topic
* as dirty. That will result in reloading of configuration once initialization is done.
*/
def topicConfigUpdated(topic: String): Unit = {
partitionsInitializing.keys.filter(_.topic() == topic).foreach {
topicPartition => partitionsInitializing.replace(topicPartition, false, true)
}
}
/**
* Mark all in progress partitions having dirty configuration if broker configuration is updated.
*/
def brokerConfigUpdated(): Unit = {
partitionsInitializing.keys.foreach {
topicPartition => partitionsInitializing.replace(topicPartition, false, true)
}
}
/**
* Method to indicate that the log initialization for the partition passed in as argument is
* finished. This method should follow a call to [[kafka.log.LogManager#initializingLog]]
*/
def finishedInitializingLog(topicPartition: TopicPartition,
maybeLog: Option[Log],
fetchLogConfig: () => LogConfig): Unit = {
if (partitionsInitializing(topicPartition)) {
maybeLog.foreach(_.updateConfig(fetchLogConfig()))
}
partitionsInitializing -= topicPartition
}
/**
* If the log already exists, just return a copy of the existing log
* Otherwise if isNew=true or if there is no offline log directory, create a log for the given topic and the given partition
* Otherwise throw KafkaStorageException
*
* @param topicPartition The partition whose log needs to be returned or created
* @param config The configuration of the log that should be applied for log creation
* @param isNew Whether the replica should have existed on the broker or not
* @param isFuture True iff the future log of the specified partition should be returned or created
* @throws KafkaStorageException if isNew=false, log is not found in the cache and there is offline log directory on the broker
*/
def getOrCreateLog(topicPartition: TopicPartition, config: LogConfig, isNew: Boolean = false, isFuture: Boolean = false): Log = {
logCreationOrDeletionLock synchronized {
getLog(topicPartition, isFuture).getOrElse {
// create the log if it has not already been created in another thread
if (!isNew && offlineLogDirs.nonEmpty)
throw new KafkaStorageException(s"Can not create log for $topicPartition because log directories ${offlineLogDirs.mkString(",")} are offline")
val logDirs: List[File] = {
val preferredLogDir = preferredLogDirs.get(topicPartition)
if (isFuture) {
if (preferredLogDir == null)
throw new IllegalStateException(s"Can not create the future log for $topicPartition without having a preferred log directory")
else if (getLog(topicPartition).get.dir.getParent == preferredLogDir)
throw new IllegalStateException(s"Can not create the future log for $topicPartition in the current log directory of this partition")
}
if (preferredLogDir != null)
List(new File(preferredLogDir))
else
nextLogDirs()
}
val logDirName = {
if (isFuture)
Log.logFutureDirName(topicPartition)
else
Log.logDirName(topicPartition)
}
val logDir = logDirs
.toStream // to prevent actually mapping the whole list, lazy map
.map(createLogDirectory(_, logDirName))
.find(_.isSuccess)
.getOrElse(Failure(new KafkaStorageException("No log directories available. Tried " + logDirs.map(_.getAbsolutePath).mkString(", "))))
.get // If Failure, will throw
val log = Log(
dir = logDir,
config = config,
logStartOffset = 0L,
recoveryPoint = 0L,
maxProducerIdExpirationMs = maxPidExpirationMs,
producerIdExpirationCheckIntervalMs = LogManager.ProducerIdExpirationCheckIntervalMs,
scheduler = scheduler,
time = time,
brokerTopicStats = brokerTopicStats,
logDirFailureChannel = logDirFailureChannel)
if (isFuture)
futureLogs.put(topicPartition, log)
else
currentLogs.put(topicPartition, log)
info(s"Created log for partition $topicPartition in $logDir with properties " + s"{${config.originals.asScala.mkString(", ")}}.")
// Remove the preferred log dir since it has already been satisfied
preferredLogDirs.remove(topicPartition)
log
}
}
}
private[log] def createLogDirectory(logDir: File, logDirName: String): Try[File] = {
val logDirPath = logDir.getAbsolutePath
if (isLogDirOnline(logDirPath)) {
val dir = new File(logDirPath, logDirName)
try {
Files.createDirectories(dir.toPath)
Success(dir)
} catch {
case e: IOException =>
val msg = s"Error while creating log for $logDirName in dir $logDirPath"
logDirFailureChannel.maybeAddOfflineLogDir(logDirPath, msg, e)
warn(msg, e)
Failure(new KafkaStorageException(msg, e))
}
} else {
Failure(new KafkaStorageException(s"Can not create log $logDirName because log directory $logDirPath is offline"))
}
}
/**
* Delete logs marked for deletion. Delete all logs for which `currentDefaultConfig.fileDeleteDelayMs`
* has elapsed after the delete was scheduled. Logs for which this interval has not yet elapsed will be
* considered for deletion in the next iteration of `deleteLogs`. The next iteration will be executed
* after the remaining time for the first log that is not deleted. If there are no more `logsToBeDeleted`,
* `deleteLogs` will be executed after `currentDefaultConfig.fileDeleteDelayMs`.
*/
private def deleteLogs(): Unit = {
var nextDelayMs = 0L
try {
def nextDeleteDelayMs: Long = {
if (!logsToBeDeleted.isEmpty) {
val (_, scheduleTimeMs) = logsToBeDeleted.peek()
scheduleTimeMs + currentDefaultConfig.fileDeleteDelayMs - time.milliseconds()
} else
currentDefaultConfig.fileDeleteDelayMs
}
while ({nextDelayMs = nextDeleteDelayMs; nextDelayMs <= 0}) {
val (removedLog, _) = logsToBeDeleted.take()
if (removedLog != null) {
try {
removedLog.delete()
info(s"Deleted log for partition ${removedLog.topicPartition} in ${removedLog.dir.getAbsolutePath}.")
} catch {
case e: KafkaStorageException =>
error(s"Exception while deleting $removedLog in dir ${removedLog.dir.getParent}.", e)
}
}
}
} catch {
case e: Throwable =>
error(s"Exception in kafka-delete-logs thread.", e)
} finally {
try {
scheduler.schedule("kafka-delete-logs",
deleteLogs _,
delay = nextDelayMs,
unit = TimeUnit.MILLISECONDS)
} catch {
case e: Throwable =>
if (scheduler.isStarted) {
// No errors should occur unless scheduler has been shutdown
error(s"Failed to schedule next delete in kafka-delete-logs thread", e)
}
}
}
}
/**
* Mark the partition directory in the source log directory for deletion and
* rename the future log of this partition in the destination log directory to be the current log
*
* @param topicPartition TopicPartition that needs to be swapped
*/
def replaceCurrentWithFutureLog(topicPartition: TopicPartition): Unit = {
logCreationOrDeletionLock synchronized {
val sourceLog = currentLogs.get(topicPartition)
val destLog = futureLogs.get(topicPartition)
info(s"Attempting to replace current log $sourceLog with $destLog for $topicPartition")
if (sourceLog == null)
throw new KafkaStorageException(s"The current replica for $topicPartition is offline")
if (destLog == null)
throw new KafkaStorageException(s"The future replica for $topicPartition is offline")
destLog.renameDir(Log.logDirName(topicPartition))
destLog.updateHighWatermark(sourceLog.highWatermark)
// Now that future replica has been successfully renamed to be the current replica
// Update the cached map and log cleaner as appropriate.
futureLogs.remove(topicPartition)
currentLogs.put(topicPartition, destLog)
if (cleaner != null) {
cleaner.alterCheckpointDir(topicPartition, sourceLog.dir.getParentFile, destLog.dir.getParentFile)
cleaner.resumeCleaning(Seq(topicPartition))
info(s"Compaction for partition $topicPartition is resumed")
}
try {
sourceLog.renameDir(Log.logDeleteDirName(topicPartition))
// Now that replica in source log directory has been successfully renamed for deletion.
// Close the log, update checkpoint files, and enqueue this log to be deleted.
sourceLog.close()
checkpointRecoveryOffsetsAndCleanSnapshot(sourceLog.dir.getParentFile, ArrayBuffer.empty)
checkpointLogStartOffsetsInDir(sourceLog.dir.getParentFile)
addLogToBeDeleted(sourceLog)
} catch {
case e: KafkaStorageException =>
// If sourceLog's log directory is offline, we need close its handlers here.
// handleLogDirFailure() will not close handlers of sourceLog because it has been removed from currentLogs map
sourceLog.closeHandlers()
sourceLog.removeLogMetrics()
throw e
}
info(s"The current replica is successfully replaced with the future replica for $topicPartition")
}
}
/**
* Rename the directory of the given topic-partition "logdir" as "logdir.uuid.delete" and
* add it in the queue for deletion.
*
* @param topicPartition TopicPartition that needs to be deleted
* @param isFuture True iff the future log of the specified partition should be deleted
* @return the removed log
*/
def asyncDelete(topicPartition: TopicPartition, isFuture: Boolean = false): Log = {
val removedLog: Log = logCreationOrDeletionLock synchronized {
if (isFuture)
futureLogs.remove(topicPartition)
else
currentLogs.remove(topicPartition)
}
if (removedLog != null) {
//We need to wait until there is no more cleaning task on the log to be deleted before actually deleting it.
if (cleaner != null && !isFuture) {
cleaner.abortCleaning(topicPartition)
cleaner.updateCheckpoints(removedLog.dir.getParentFile)
}
removedLog.renameDir(Log.logDeleteDirName(topicPartition))
checkpointRecoveryOffsetsAndCleanSnapshot(removedLog.dir.getParentFile, ArrayBuffer.empty)
checkpointLogStartOffsetsInDir(removedLog.dir.getParentFile)
addLogToBeDeleted(removedLog)
info(s"Log for partition ${removedLog.topicPartition} is renamed to ${removedLog.dir.getAbsolutePath} and is scheduled for deletion")
} else if (offlineLogDirs.nonEmpty) {
throw new KafkaStorageException(s"Failed to delete log for ${if (isFuture) "future" else ""} $topicPartition because it may be in one of the offline directories ${offlineLogDirs.mkString(",")}")
}
removedLog
}
/**
* Provides the full ordered list of suggested directories for the next partition.
* Currently this is done by calculating the number of partitions in each directory and then sorting the
* data directories by fewest partitions.
*/
private def nextLogDirs(): List[File] = {
if(_liveLogDirs.size == 1) {
List(_liveLogDirs.peek())
} else {
// count the number of logs in each parent directory (including 0 for empty directories
val logCounts = allLogs.groupBy(_.dir.getParent).mapValues(_.size)
val zeros = _liveLogDirs.asScala.map(dir => (dir.getPath, 0)).toMap
val dirCounts = (zeros ++ logCounts).toBuffer
// choose the directory with the least logs in it
dirCounts.sortBy(_._2).map {
case (path: String, _: Int) => new File(path)
}.toList
}
}
/**
* Delete any eligible logs. Return the number of segments deleted.
* Only consider logs that are not compacted.
*/
def cleanupLogs(): Unit = {
debug("Beginning log cleanup...")
var total = 0
val startMs = time.milliseconds
// clean current logs.
val deletableLogs = {
if (cleaner != null) {
// prevent cleaner from working on same partitions when changing cleanup policy
cleaner.pauseCleaningForNonCompactedPartitions()
} else {
currentLogs.filter {
case (_, log) => !log.config.compact
}
}
}
try {
deletableLogs.foreach {
case (topicPartition, log) =>
debug(s"Garbage collecting '${log.name}'")
total += log.deleteOldSegments()
val futureLog = futureLogs.get(topicPartition)
if (futureLog != null) {
// clean future logs
debug(s"Garbage collecting future log '${futureLog.name}'")
total += futureLog.deleteOldSegments()
}
}
} finally {
if (cleaner != null) {
cleaner.resumeCleaning(deletableLogs.map(_._1))
}
}
debug(s"Log cleanup completed. $total files deleted in " +
(time.milliseconds - startMs) / 1000 + " seconds")
}
/**
* Get all the partition logs
*/
def allLogs: Iterable[Log] = currentLogs.values ++ futureLogs.values
def logsByTopic(topic: String): Seq[Log] = {
(currentLogs.toList ++ futureLogs.toList).collect {
case (topicPartition, log) if topicPartition.topic == topic => log
}
}
/**
* Map of log dir to logs by topic and partitions in that dir
*/
private def logsByDir: Map[String, Map[TopicPartition, Log]] = {
(this.currentLogs.toList ++ this.futureLogs.toList).toMap
.groupBy { case (_, log) => log.dir.getParent }
}
// logDir should be an absolute path
def isLogDirOnline(logDir: String): Boolean = {
// The logDir should be an absolute path
if (!logDirs.exists(_.getAbsolutePath == logDir))
throw new LogDirNotFoundException(s"Log dir $logDir is not found in the config.")
_liveLogDirs.contains(new File(logDir))
}
/**
* Flush any log which has exceeded its flush interval and has unwritten messages.
*/
private def flushDirtyLogs(): Unit = {
debug("Checking for dirty logs to flush...")
for ((topicPartition, log) <- currentLogs.toList ++ futureLogs.toList) {
try {
val timeSinceLastFlush = time.milliseconds - log.lastFlushTime
debug(s"Checking if flush is needed on ${topicPartition.topic} flush interval ${log.config.flushMs}" +
s" last flushed ${log.lastFlushTime} time since last flush: $timeSinceLastFlush")
if(timeSinceLastFlush >= log.config.flushMs)
log.flush
} catch {
case e: Throwable =>
error(s"Error flushing topic ${topicPartition.topic}", e)
}
}
}
}
object LogManager {
val RecoveryPointCheckpointFile = "recovery-point-offset-checkpoint"
val LogStartOffsetCheckpointFile = "log-start-offset-checkpoint"
val ProducerIdExpirationCheckIntervalMs = 10 * 60 * 1000
def apply(config: KafkaConfig,
initialOfflineDirs: Seq[String],
zkClient: KafkaZkClient,
brokerState: BrokerState,
kafkaScheduler: KafkaScheduler,
time: Time,
brokerTopicStats: BrokerTopicStats,
logDirFailureChannel: LogDirFailureChannel): LogManager = {
val defaultProps = KafkaServer.copyKafkaConfigToLog(config)
LogConfig.validateValues(defaultProps)
val defaultLogConfig = LogConfig(defaultProps)
// read the log configurations from zookeeper
val (topicConfigs, failed) = zkClient.getLogConfigs(
zkClient.getAllTopicsInCluster,
defaultProps
)
if (!failed.isEmpty) throw failed.head._2
val cleanerConfig = LogCleaner.cleanerConfig(config)
new LogManager(logDirs = config.logDirs.map(new File(_).getAbsoluteFile),
initialOfflineDirs = initialOfflineDirs.map(new File(_).getAbsoluteFile),
topicConfigs = topicConfigs,
initialDefaultConfig = defaultLogConfig,
cleanerConfig = cleanerConfig,
recoveryThreadsPerDataDir = config.numRecoveryThreadsPerDataDir,
flushCheckMs = config.logFlushSchedulerIntervalMs,
flushRecoveryOffsetCheckpointMs = config.logFlushOffsetCheckpointIntervalMs,
flushStartOffsetCheckpointMs = config.logFlushStartOffsetCheckpointIntervalMs,
retentionCheckMs = config.logCleanupIntervalMs,
maxPidExpirationMs = config.transactionalIdExpirationMs,
scheduler = kafkaScheduler,
brokerState = brokerState,
brokerTopicStats = brokerTopicStats,
logDirFailureChannel = logDirFailureChannel,
time = time)
}
}