Cassandra 的启动和初始化

Cassandra 的入口

CassandraDaemon 作为Cassandra的入口，做了以下几件事:

• load configuration
• 注册 MBean service
• load schemas
• scrub data directories
• init keyspaces
• start StorageService
• start gossip service
• start native/thrift server

main()

    public static void main(String[] args)
    {
        instance.activate();
    }

activate()

    public void activate()
    {
        MBeanServer mbs = ManagementFactory.getPlatformMBeanServer();
        mbs.registerMBean(new StandardMBean(new NativeAccess(), NativeAccessMBean.class), new ObjectName(MBEAN_NAME));

        setup();
        start();
    }

setup()

    protected void setup()
    {
        // check the system keyspace to keep user from shooting self in foot by changing partitioner, cluster name, etc.
        // we do a one-off scrub of the system keyspace first; we can't load the list of the rest of the keyspaces,
        // until system keyspace is opened.
        for (CFMetaData cfm : Schema.instance.getKeyspaceMetaData(Keyspace.SYSTEM_KS).values())
            ColumnFamilyStore.scrubDataDirectories(cfm);
    
        SystemKeyspace.checkHealth();

        // load keyspace descriptions.
        DatabaseDescriptor.loadSchemas();

        // clean up compaction leftovers
        Map, Map> unfinishedCompactions = SystemKeyspace.getUnfinishedCompactions();

        // clean up debris in the rest of the keyspaces
        for (String keyspaceName : Schema.instance.getKeyspaces())
        {
            //...
        }

        Keyspace.setInitialized();

        // initialize keyspaces
        for (String keyspaceName : Schema.instance.getKeyspaces())
        {
            //...
        }

        // replay the log if necessary
        CommitLog.instance.recover();

        // start server internals
        StorageService.instance.registerDaemon(this);
        StorageService.instance.initServer();
        
        // Metrics
        String metricsReporterConfigFile = System.getProperty("cassandra.metricsReporterConfigFile");

        // schedule periodic background compaction task submission. this is simply a backstop against compactions stalling
        // due to scheduling errors or race conditions
        ScheduledExecutors.optionalTasks.scheduleWithFixedDelay(ColumnFamilyStore.getBackgroundCompactionTaskSubmitter(), 5, 1, TimeUnit.MINUTES);

        // Thrift
        thriftServer = new ThriftServer(rpcAddr, rpcPort, listenBacklog);
        // Native transport
        nativeServer = new org.apache.cassandra.transport.Server(nativeAddr, nativePort);

        completeSetup();
    }

start()

public void start()
{
    nativeServer.start();
    thriftServer.start();
}

初始化 DatabaseDescriptor

通过 static block 初始化

    static
    {
        applyConfig(loadConfig());
    }

首先是读入config, 用yaml的library读取config (org.yaml.snakeyaml).

    @VisibleForTesting
    public static Config loadConfig() throws ConfigurationException
    {
        String loaderClass = System.getProperty("cassandra.config.loader");
        ConfigurationLoader loader = loaderClass == null
                                   ? new YamlConfigurationLoader()
                                   : FBUtilities.construct(loaderClass, "configuration loading");
        return loader.loadConfig();
    }

再apply config, 主要是validate config, 用config 初始化某些service比如snitch, comparator, memoryAllocator.