七張圖完全講清楚ZooKeeper分佈式鎖的實現原理【石杉的架構筆記】

歡迎關注微信公衆號：石杉的架構筆記（id：shishan100）

個人新課**《C2C 電商系統微服務架構120天實戰訓練營》在公衆號儒猿技術窩**上線了，感興趣的同窗，能夠點擊下方連接瞭解詳情：

《C2C 電商系統微服務架構120天實戰訓練營》

1、寫在前面

以前寫過一篇文章（《拜託，面試請不要再問我Redis分佈式鎖的實現原理》），給你們說了一下Redisson這個開源框架是如何實現Redis分佈式鎖原理的，這篇文章再給你們聊一下ZooKeeper實現分佈式鎖的原理。

同理，我是直接基於比較經常使用的Curator這個開源框架，聊一下這個框架對ZooKeeper（如下簡稱zk）分佈式鎖的實現。

通常除了大公司是自行封裝分佈式鎖框架以外，建議你們用這些開源框架封裝好的分佈式鎖實現，這是一個比較快捷省事兒的方式。

2、ZooKeeper分佈式鎖機制

接下來咱們一塊兒來看看，多客戶端獲取及釋放zk分佈式鎖的整個流程及背後的原理。

首先你們看看下面的圖，若是如今有兩個客戶端一塊兒要爭搶zk上的一把分佈式鎖，會是個什麼場景？

若是你們對zk還不太瞭解的話，建議先自行百度一下，簡單瞭解點基本概念，好比zk有哪些節點類型等等。

參見上圖。zk裏有一把鎖，這個鎖就是zk上的一個節點。而後呢，兩個客戶端都要來獲取這個鎖，具體是怎麼來獲取呢？

我們就假設客戶端A搶先一步，對zk發起了加分佈式鎖的請求，這個加鎖請求是用到了zk中的一個特殊的概念，叫作**「臨時順序節點」。**

簡單來講，就是直接在"my_lock"這個鎖節點下，建立一個順序節點，這個順序節點有zk內部自行維護的一個節點序號。

好比說，第一個客戶端來搞一個順序節點，zk內部會給起個名字叫作：xxx-000001。而後第二個客戶端來搞一個順序節點，zk可能會起個名字叫作：xxx-000002。你們注意一下，最後一個數字都是依次遞增的，從1開始逐次遞增。zk會維護這個順序。

因此這個時候，假如說客戶端A先發起請求，就會搞出來一個順序節點，你們看下面的圖，Curator框架大概會弄成以下的樣子：

你們看，客戶端A發起一個加鎖請求，先會在你要加鎖的node下搞一個臨時順序節點，這一大坨長長的名字都是Curator框架本身生成出來的。

而後，那個最後一個數字是"1"。你們注意一下，由於客戶端A是第一個發起請求的，因此給他搞出來的順序節點的序號是"1"。

接着客戶端A建立完一個順序節點。還沒完，他會查一下"my_lock"這個鎖節點下的全部子節點，而且這些子節點是按照序號排序的，這個時候他大概會拿到這麼一個集合：

接着客戶端A會走一個關鍵性的判斷，就是說：唉！兄弟，這個集合裏，我建立的那個順序節點，是否是排在第一個啊？

若是是的話，那我就能夠加鎖了啊！由於明明我就是第一個來建立順序節點的人，因此我就是第一個嘗試加分佈式鎖的人啊！

bingo！加鎖成功！你們看下面的圖，再來直觀的感覺一下整個過程。

接着假如說，客戶端A都加完鎖了，客戶端B過來想要加鎖了，這個時候他會幹同樣的事兒：先是在"my_lock"這個鎖節點下建立一個臨時順序節點，此時名字會變成相似於：

你們看看下面的圖：

客戶端B由於是第二個來建立順序節點的，因此zk內部會維護序號爲"2"。

接着客戶端B會走加鎖判斷邏輯，查詢"my_lock"鎖節點下的全部子節點，按序號順序排列，此時他看到的相似於：

同時檢查本身建立的順序節點，是否是集合中的第一個？

明顯不是啊，此時第一個是客戶端A建立的那個順序節點，序號爲"01"的那個。因此加鎖失敗！

加鎖失敗了之後，客戶端B就會經過ZK的API對他的順序節點的**上一個順序節點加一個監聽器。**zk自然就能夠實現對某個節點的監聽。

若是你們還不知道zk的基本用法，能夠百度查閱，很是的簡單。客戶端B的順序節點是：

他的上一個順序節點，不就是下面這個嗎？

即客戶端A建立的那個順序節點！

因此，客戶端B會對：

這個節點加一個監聽器，監聽這個節點是否被刪除等變化！你們看下面的圖。

接着，客戶端A加鎖以後，可能處理了一些代碼邏輯，而後就會釋放鎖。那麼，釋放鎖是個什麼過程呢？

其實很簡單，就是把本身在zk裏建立的那個順序節點，也就是：

這個節點給刪除。

刪除了那個節點以後，zk會負責通知監聽這個節點的監聽器，也就是客戶端B以前加的那個監聽器，說：兄弟，你監聽的那個節點被刪除了，有人釋放了鎖。

此時客戶端B的監聽器感知到了上一個順序節點被刪除，也就是排在他以前的某個客戶端釋放了鎖。

此時，就會通知客戶端B從新嘗試去獲取鎖，也就是獲取"my_lock"節點下的子節點集合，此時爲：

集合裏此時只有客戶端B建立的惟一的一個順序節點了！

而後呢，客戶端B判斷本身竟然是集合中的第一個順序節點，bingo！能夠加鎖了！直接完成加鎖，運行後續的業務代碼便可，運行完了以後再次釋放鎖。

3、總結

其實若是有客戶端C、客戶端D等N個客戶端爭搶一個zk分佈式鎖，原理都是相似的。

• 你們都是上來直接建立一個鎖節點下的一個接一個的臨時順序節點
• 若是本身不是第一個節點，就對本身上一個節點加監聽器
• 只要上一個節點釋放鎖，本身就排到前面去了，至關因而一個排隊機制。

並且用臨時順序節點的另一個用意就是，若是某個客戶端建立臨時順序節點以後，不當心本身宕機了也不要緊，zk感知到那個客戶端宕機，會自動刪除對應的臨時順序節點，至關於自動釋放鎖，或者是自動取消本身的排隊。

最後，我們來看下用Curator框架進行加鎖和釋放鎖的一個過程：

其實用開源框架就是這點好，方便。這個Curator框架的zk分佈式鎖的加鎖和釋放鎖的實現原理，其實就是上面咱們說的那樣子。

可是若是你要手動實現一套那個代碼的話。仍是有點麻煩的，要考慮到各類細節，異常處理等等。因此你們若是考慮用zk分佈式鎖，能夠參考下本文的思路。

END

若有收穫，請幫忙轉發，您的鼓勵是做者最大的動力，謝謝！

一大波微服務、分佈式、高併發、高可用的原創系列文章正在路上

歡迎掃描下方二維碼，持續關注：

石杉的架構筆記（id:shishan100）

十餘年BAT架構經驗傾囊相授

**> **推薦閱讀：** > > 一、[拜託！面試請不要再問我Spring Cloud底層原理](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Flink.juejin.im%2525253Ftarget%2525253Dhttps%252525253A%252525252F%252525252Flink.juejin.im%252525252F%252525253Ftarget%252525253Dhttps%25252525253A%25252525252F%25252525252Flink.juejin.im%25252525252F%25252525253Ftarget%25252525253Dhttps%2525252525253A%2525252525252F%2525252525252Fjuejin.im%2525252525252Fpost%2525252525252F5be13b83f265da6116393fc7) > > 二、[【雙11狂歡的背後】微服務註冊中心如何承載大型系統的千萬級訪問？](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Flink.juejin.im%2525253Ftarget%2525253Dhttps%252525253A%252525252F%252525252Flink.juejin.im%252525252F%252525253Ftarget%252525253Dhttps%25252525253A%25252525252F%25252525252Flink.juejin.im%25252525252F%25252525253Ftarget%25252525253Dhttps%2525252525253A%2525252525252F%2525252525252Fjuejin.im%2525252525252Fpost%2525252525252F5be3f8dcf265da613a5382ca) > > 三、[【性能優化之道】每秒上萬併發下的Spring Cloud參數優化實戰](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Flink.juejin.im%2525253Ftarget%2525253Dhttps%252525253A%252525252F%252525252Flink.juejin.im%252525252F%252525253Ftarget%252525253Dhttps%25252525253A%25252525252F%25252525252Flink.juejin.im%25252525252F%25252525253Ftarget%25252525253Dhttps%2525252525253A%2525252525252F%2525252525252Fjuejin.im%2525252525252Fpost%2525252525252F5be83e166fb9a049a7115580) > > 四、[微服務架構如何保障雙11狂歡下的99.99%高可用](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Flink.juejin.im%2525253Ftarget%2525253Dhttps%252525253A%252525252F%252525252Flink.juejin.im%252525252F%252525253Ftarget%252525253Dhttps%25252525253A%25252525252F%25252525252Flink.juejin.im%25252525252F%25252525253Ftarget%25252525253Dhttps%2525252525253A%2525252525252F%2525252525252Fjuejin.im%2525252525252Fpost%2525252525252F5be99a68e51d4511a8090440) > > 五、[兄弟，用大白話告訴你小白都能聽懂的Hadoop架構原理](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Flink.juejin.im%2525253Ftarget%2525253Dhttps%252525253A%252525252F%252525252Flink.juejin.im%252525252F%252525253Ftarget%252525253Dhttps%25252525253A%25252525252F%25252525252Flink.juejin.im%25252525252F%25252525253Ftarget%25252525253Dhttps%2525252525253A%2525252525252F%2525252525252Fjuejin.im%2525252525252Fpost%2525252525252F5beaf02ce51d457e90196069) > > 六、[大規模集羣下Hadoop NameNode如何承載每秒上千次的高併發訪問](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Flink.juejin.im%2525253Ftarget%2525253Dhttps%252525253A%252525252F%252525252Flink.juejin.im%252525252F%252525253Ftarget%252525253Dhttps%25252525253A%25252525252F%25252525252Flink.juejin.im%25252525252F%25252525253Ftarget%25252525253Dhttps%2525252525253A%2525252525252F%2525252525252Fjuejin.im%2525252525252Fpost%2525252525252F5bec278c5188253e64332c76) > > 七、【[性能優化的祕密】Hadoop如何將TB級大文件的上傳性能優化上百倍](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Flink.juejin.im%2525253Ftarget%2525253Dhttps%252525253A%252525252F%252525252Flink.juejin.im%252525252F%252525253Ftarget%252525253Dhttps%25252525253A%25252525252F%25252525252Flink.juejin.im%25252525252F%25252525253Ftarget%25252525253Dhttps%2525252525253A%2525252525252F%2525252525252Fjuejin.im%2525252525252Fpost%2525252525252F5bed82a9e51d450f9461cfc7) > > 八、[拜託，面試請不要再問我TCC分佈式事務的實現原理](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Flink.juejin.im%2525253Ftarget%2525253Dhttps%252525253A%252525252F%252525252Flink.juejin.im%252525252F%252525253Ftarget%252525253Dhttps%25252525253A%25252525252F%25252525252Fjuejin.im%25252525252Fpost%25252525252F5bf201f7f265da610f63528a)[坑爹呀！](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Flink.juejin.im%2525253Ftarget%2525253Dhttps%252525253A%252525252F%252525252Fjuejin.im%252525252Fpost%252525252F5bf2c6b6e51d456693549af4) > > 九、[【坑爹呀！】最終一致性分佈式事務如何保障實際生產中99.99%高可用？](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Flink.juejin.im%2525253Ftarget%2525253Dhttps%252525253A%252525252F%252525252Fjuejin.im%252525252Fpost%252525252F5bf2c6b6e51d456693549af4) > > 十、[拜託，面試請不要再問我Redis分佈式鎖的實現原理！](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Flink.juejin.im%2525253Ftarget%2525253Dhttps%252525253A%252525252F%252525252Fjuejin.im%252525252Fpost%252525252F5bf3f15851882526a643e207) > > **十一、****[【眼前一亮！】看Hadoop底層算法如何優雅的將大規模集羣性能提高10倍以上？](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Flink.juejin.im%2525253Ftarget%2525253Dhttps%252525253A%252525252F%252525252Fjuejin.im%252525252Fpost%252525252F5bf5396f51882509a768067e)** > > **十二、****[億級流量系統架構之如何支撐百億級數據的存儲與計算](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Flink.juejin.im%25252F%25253Ftarget%25253Dhttps%2525253A%2525252F%2525252Fjuejin.im%2525252Fpost%2525252F5bfab59fe51d4551584c7bcf)** > > 1三、[億級流量系統架構之如何設計高容錯分佈式計算系統](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Flink.juejin.im%253Ftarget%253Dhttps%25253A%25252F%25252Fjuejin.im%25252Fpost%25252F5bfbeeb9f265da61407e9679) > > 1四、[億級流量系統架構之如何設計承載百億流量的高性能架構](https://link.juejin.im/?target=https%3A%2F%2Flink.juejin.im%3Ftarget%3Dhttps%253A%252F%252Fjuejin.im%252Fpost%252F5bfd2df1e51d4574b133dd3a) > > 1五、[億級流量系統架構之如何設計每秒十萬查詢的高併發架構](https://link.juejin.im/?target=https%3A%2F%2Fjuejin.im%2Fpost%2F5bfe771251882509a7681b3a) > > 1六、[億級流量系統架構之如何設計全鏈路99.99%高可用架構](https://juejin.cn/post/6844903727699230734)**