HBase源代码调试(6)-HRegion的启动

前提:debug configurations里面

MainClass : HRegion

Arguments: .META.

-----------------------------------------------

第一步构造HLog 对象,构造函数中

首先在/tmp/hbase-user下创建 hlog.META.1357446070560文件夹, 数字串是hregion启动的时间(currentTimeMills).

然后看有无 ".oldlogs'文件夹, 无则创建

然后是读取各种conf里面的设置 如 hbase.regionserver.maxlogs, default is 32

then under the hlog.META.XXXXX/ , create a file named 'hlog.135747xxxx', still use the System.currentTimeMills()

then call ->Writer createWriter(); ->SequenceFile$Writer.init()

        it will call writeFileHeader, and put VERSION, keyClass etc into header

launch a LogSyncer Thread

new a WALCoprocessorHost

In method  initializeRegionInternals

      ThreadPoolExecutor storeOpenerThreadPool =
        getStoreOpenAndCloseThreadPool(
          "StoreOpenerThread-" + this.regionInfo.getRegionNameAsString());
      CompletionService<Store> completionService =
        new ExecutorCompletionService<Store>(storeOpenerThreadPool);

      // initialize each store in parallel
      for (final HColumnDescriptor family : htableDescriptor.getFamilies()) {
        status.setStatus("Instantiating store for column family " + family);
        completionService.submit(new Callable<Store>() {
          public Store call() throws IOException {
            return instantiateHStore(tableDir, family);
          }
        });

可以看到对每个family,启动一个线程, 执行

instantiateHStore

返回一个new Store对象,而在store的构造函数中,调用了method

 this.storefiles = sortAndClone(loadStoreFiles());

in loadStoreFiles,会启动线程

     completionService.submit(new Callable<StoreFile>() {
        public StoreFile call() throws IOException {
          StoreFile storeFile = new StoreFile(fs, p, conf, cacheConf,
              family.getBloomFilterType(), dataBlockEncoder);
          passSchemaMetricsTo(storeFile);
          storeFile.createReader();
          return storeFile;
        }
      });

在createReader的过程中读取分析HFile