您好,登錄后才能下訂單哦!
這篇“InterProcessMutex實現zookeeper分布式鎖原理是什么”文章的知識點大部分人都不太理解,所以小編給大家總結了以下內容,內容詳細,步驟清晰,具有一定的借鑒價值,希望大家閱讀完這篇文章能有所收獲,下面我們一起來看看這篇“InterProcessMutex實現zookeeper分布式鎖原理是什么”文章吧。
zookeeper實現分布式鎖的原理就是多個節點同時在一個指定的節點下面創建臨時會話順序節點,誰創建的節點序號最小,誰就獲得了鎖,并且其他節點就會監聽序號比自己小的節點,一旦序號比自己小的節點被刪除了,其他節點就會得到相應的事件,然后查看自己是否為序號最小的節點,如果是,則獲取鎖。
InterProcessMutex實現的鎖機制是公平且互斥的,公平的方式是按照每個請求的順序進行排隊的。
InterProcessMutex實現的InterProcessLock接口,InterProcessLock主要規范了如下幾個方法:
// 獲取互斥鎖 public void acquire() throws Exception; // 在給定的時間內獲取互斥鎖 public boolean acquire(long time, TimeUnit unit) throws Exception; // 釋放鎖處理 public void release() throws Exception; // 如果此JVM中的線程獲取了互斥鎖,則返回true boolean isAcquiredInThisProcess();
接下來我們看看InterProcessMutex中的實現,它究竟有哪些屬性,以及實現細節
public class InterProcessMutex implements InterProcessLock, Revocable<InterProcessMutex> { // LockInternals是真正實現操作zookeeper的類,它內部包含連接zookeeper客戶端的CuratorFramework // LockInternals的具體實現后面我會講到 private final LockInternals internals; // basePath是鎖的根結點,所有的臨時有序的節點都是basePath的子節點, private final String basePath; // private final ConcurrentMap<Thread, LockData> threadData = Maps.newConcurrentMap(); // LockData封裝了請求對應的線程(owningThread)、鎖的重入的次數(lockCount)、線程對應的臨時節點(lockPath) private static class LockData { final Thread owningThread; final String lockPath; // 原子性的 final AtomicInteger lockCount = new AtomicInteger(1); private LockData(Thread owningThread, String lockPath) { this.owningThread = owningThread; this.lockPath = lockPath; } } private static final String LOCK_NAME = "lock-"; // 獲取互斥鎖,阻塞【InterProcessLock的實現】 @Override public void acquire() throws Exception { // 獲取鎖,一直等待 if ( !internalLock(-1, null) ) { throw new IOException("Lost connection while trying to acquire lock: " + basePath); } } // 獲取互斥鎖,指定時間time【InterProcessLock的實現】 @Override public boolean acquire(long time, TimeUnit unit) throws Exception { return internalLock(time, unit); } // 當前線程是否占用鎖中【InterProcessLock的實現】 @Override public boolean isAcquiredInThisProcess() { return (threadData.size() > 0); } //如果調用線程與獲取互斥鎖的線程相同,則執行一次互斥鎖釋放。如果線程已多次調用acquire,當此方法返回時,互斥鎖仍將保留 【InterProcessLock的實現】 @Override public void release() throws Exception { Thread currentThread = Thread.currentThread(); //當前線程 LockData lockData = threadData.get(currentThread); //線程對應的鎖信息 if ( lockData == null ) { throw new IllegalMonitorStateException("You do not own the lock: " + basePath); } // 因為獲取到的鎖是可重入的,對lockCount進行減1,lockCount=0時才是真正釋放鎖 int newLockCount = lockData.lockCount.decrementAndGet(); if ( newLockCount > 0 ) { return; } if ( newLockCount < 0 ) { throw new IllegalMonitorStateException("Lock count has gone negative for lock: " + basePath); } try { // 到這里時lockCount=0,具體釋放鎖的操作交給LockInternals中的releaseLock方法實現 internals.releaseLock(lockData.lockPath); } finally { threadData.remove(currentThread); } } // 獲取basePath根結點下的所有臨時節點的有序集合 public Collection<String> getParticipantNodes() throws Exception { return LockInternals.getParticipantNodes(internals.getClient(), basePath, internals.getLockName(), internals.getDriver()); } boolean isOwnedByCurrentThread() { LockData lockData = threadData.get(Thread.currentThread()); return (lockData != null) && (lockData.lockCount.get() > 0); } protected String getLockPath() { LockData lockData = threadData.get(Thread.currentThread()); return lockData != null ? lockData.lockPath : null; } // acquire()中調用的internalLock()方法 private boolean internalLock(long time, TimeUnit unit) throws Exception { Thread currentThread = Thread.currentThread(); LockData lockData = threadData.get(currentThread); if ( lockData != null ) { // 如果當前線程已經獲取到了鎖,那么將重入次數lockCount+1,返回true lockData.lockCount.incrementAndGet(); return true; } // attemptLock方法是獲取鎖的真正實現,lockPath是當前線程成功在basePath下創建的節點,若lockPath不為空代表成功獲取到鎖 String lockPath = internals.attemptLock(time, unit, getLockNodeBytes()); if ( lockPath != null ) { // lockPath封裝到當前線程對應的鎖信息中 LockData newLockData = new LockData(currentThread, lockPath); threadData.put(currentThread, newLockData); return true; } return false; } }
接下來我們看看InterProcessMutex中使用的LockInternals類的實現細節
public class LockInternals { private final CuratorFramework client; // 連接zookeeper的客戶端 private final String path; // 等于basePath,InterProcessMutex中傳進來的 private final String basePath; // 根結點 private final LockInternalsDriver driver; // 操作zookeeper節點的driver private final String lockName; // "lock-" private final AtomicReference<RevocationSpec> revocable = new AtomicReference<RevocationSpec>(null); private final CuratorWatcher revocableWatcher = new CuratorWatcher() { @Override public void process(WatchedEvent event) throws Exception { if ( event.getType() == Watcher.Event.EventType.NodeDataChanged ) { checkRevocableWatcher(event.getPath()); } } }; // 監聽節點的監聽器,若被監聽的節點有動靜,則喚醒 notifyFromWatcher()=>notifyAll(); private final Watcher watcher = new Watcher() { @Override public void process(WatchedEvent event) { notifyFromWatcher(); } }; private volatile int maxLeases; // 獲取basePath的子節點,排序后的 public static List<String> getSortedChildren(CuratorFramework client, String basePath, final String lockName, final LockInternalsSorter sorter) throws Exception { List<String> children = client.getChildren().forPath(basePath); List<String> sortedList = Lists.newArrayList(children); Collections.sort ( sortedList, new Comparator<String>() { @Override public int compare(String lhs, String rhs) { return sorter.fixForSorting(lhs, lockName).compareTo(sorter.fixForSorting(rhs, lockName)); } } ); return sortedList; } // 嘗試獲取鎖【internalLock=>attemptLock】 String attemptLock(long time, TimeUnit unit, byte[] lockNodeBytes) throws Exception { // 開始時間 final long startMillis = System.currentTimeMillis(); // 記錄等待時間 final Long millisToWait = (unit != null) ? unit.toMillis(time) : null; final byte[] localLockNodeBytes = (revocable.get() != null) ? new byte[0] : lockNodeBytes; // 重試次數 int retryCount = 0; // 當前節點 String ourPath = null; // 是否獲取到鎖的標志 boolean hasTheLock = false; // 是否放棄獲取到標志 boolean isDone = false; // 不停嘗試獲取 while ( !isDone ) { isDone = true; try { // 創建當前線程對應的節點 ourPath = driver.createsTheLock(client, path, localLockNodeBytes); // internalLockLoop中獲取 hasTheLock = internalLockLoop(startMillis, millisToWait, ourPath); } catch ( KeeperException.NoNodeException e ) { // 是否可再次嘗試 if ( client.getZookeeperClient().getRetryPolicy().allowRetry(retryCount++, System.currentTimeMillis() - startMillis, RetryLoop.getDefaultRetrySleeper()) ) { isDone = false; } else { throw e; } } } // 獲取到鎖后,返回當前線程對應創建的節點路徑 if ( hasTheLock ) { return ourPath; } return null; } // 循環獲取【attemptLock=>internalLockLoop】 private boolean internalLockLoop(long startMillis, Long millisToWait, String ourPath) throws Exception { boolean haveTheLock = false; // 是否擁有分布式鎖 boolean doDelete = false; // 是否需要刪除當前節點 try { if ( revocable.get() != null ) { client.getData().usingWatcher(revocableWatcher).forPath(ourPath); } // 循環嘗試獲取鎖 while ( (client.getState() == CuratorFrameworkState.STARTED) && !haveTheLock ) { // 得到basePath下排序后的臨時子節點 List<String> children = getSortedChildren(); // 獲取之前創建的當前線程對應的子節點 String sequenceNodeName = ourPath.substring(basePath.length() + 1); // +1 to include the slash // 判斷是否獲取到鎖,沒有就返回監聽路徑 PredicateResults predicateResults = driver.getsTheLock(client, children, sequenceNodeName, maxLeases); // 成功獲取到 if ( predicateResults.getsTheLock() ) { haveTheLock = true; } else { // 沒有獲取到鎖,監聽前一個臨時順序節點 String previousSequencePath = basePath + "/" + predicateResults.getPathToWatch(); synchronized(this) { try { // 上一個臨時順序節點如果被刪除,會喚醒當前線程繼續競爭鎖 client.getData().usingWatcher(watcher).forPath(previousSequencePath); if ( millisToWait != null ) { millisToWait -= (System.currentTimeMillis() - startMillis); startMillis = System.currentTimeMillis(); // 獲取鎖超時 if ( millisToWait <= 0 ) { doDelete = true; // timed out - delete our node break; } wait(millisToWait); } else { wait(); } } catch ( KeeperException.NoNodeException e ) { // it has been deleted (i.e. lock released). Try to acquire again } } } } } catch ( Exception e ) { ThreadUtils.checkInterrupted(e); doDelete = true; throw e; } finally { if ( doDelete ) { // 因為獲取鎖超時,所以刪除之前創建的臨時子節點 deleteOurPath(ourPath); } } return haveTheLock; } private void deleteOurPath(String ourPath) throws Exception { try { // 刪除 client.delete().guaranteed().forPath(ourPath); } catch ( KeeperException.NoNodeException e ) { // ignore - already deleted (possibly expired session, etc.) } } }
StandardLockInternalsDriver implements LockInternalsDriver
// 前面internalLockLoop方法中driver.getsTheLock執行的方法 @Override public PredicateResults getsTheLock(CuratorFramework client, List<String> children, String sequenceNodeName, int maxLeases) throws Exception { // 獲取子節點在臨時順序節點列表中的位置 int ourIndex = children.indexOf(sequenceNodeName); // 檢驗子節點在臨時順序節點列表中是否有效 validateOurIndex(sequenceNodeName, ourIndex); // 若當前子節點的位置<maxLeases,代表可獲取鎖【maxLeases默認=1,若ourIndex=0,代筆自己位置最小】 boolean getsTheLock = ourIndex < maxLeases; // getsTheLock=true,則不需要監聽前maxLeases的節點【maxLeases默認=1,代表監聽前面最靠近自己的節點】 String pathToWatch = getsTheLock ? null : children.get(ourIndex - maxLeases); return new PredicateResults(pathToWatch, getsTheLock); }
以上就是關于“InterProcessMutex實現zookeeper分布式鎖原理是什么”這篇文章的內容,相信大家都有了一定的了解,希望小編分享的內容對大家有幫助,若想了解更多相關的知識內容,請關注億速云行業資訊頻道。
免責聲明:本站發布的內容(圖片、視頻和文字)以原創、轉載和分享為主,文章觀點不代表本網站立場,如果涉及侵權請聯系站長郵箱:is@yisu.com進行舉報,并提供相關證據,一經查實,將立刻刪除涉嫌侵權內容。