概览
本文跟一下leaf的雪花模式的算法
关注点:
- workerid生成
- 时间回拨问题解决
leaf是美团开源的分布式id 项目
源码分析
-
首先从server的Controller出发,看一下雪花算法生成的方法
-
@RequestMapping(value = "/api/snowflake/get/{key}") public String getSnowflakeId(@PathVariable("key") String key) { return get(key, snowflakeService.getId(key)); }
-
进入到
snowflakeService
-
发现核心生成id的类是
SnowflakeIDGenImpl
-
private final long workerIdBits = 10L; private final long sequenceBits = 12L;
-
正常的workerid长度1024,以及序号4096
-
时间起始时间是
//Thu Nov 04 2010 09:42:54 GMT+0800 (中国标准时间) this(zkAddress, port, 1288834974657L);
-
public SnowflakeIDGenImpl(String zkAddress, int port, long twepoch) { this.twepoch = twepoch;Preconditions.checkArgument(timeGen() > twepoch, "Snowflake not support twepoch gt currentTime");final String ip = Utils.getIp();SnowflakeZookeeperHolder holder = new SnowflakeZookeeperHolder(ip, String.valueOf(port), zkAddress);//获得zookeeper的连接其LOGGER.info("twepoch:{} ,ip:{} ,zkAddress:{} port:{}", twepoch, ip, zkAddress, port);boolean initFlag = holder.init();//在这里是生成idif (initFlag) { workerId = holder.getWorkerID();//获得当前生成器机器的idLOGGER.info("START SUCCESS USE ZK WORKERID-{}", workerId);} else { Preconditions.checkArgument(initFlag, "Snowflake Id Gen is not init ok");}Preconditions.checkArgument(workerId >= 0 && workerId <= maxWorkerId, "workerID must gte 0 and lte 1023");}
-
来看看
SnowflakeZookeeperHolder.init()
方法 -
public boolean init() { try { CuratorFramework curator = createWithOptions(connectionString, new RetryUntilElapsed(1000, 4), 10000, 6000);curator.start();Stat stat = curator.checkExists().forPath(PATH_FOREVER);if (stat == null) { //不存在根节点,机器第一次启动,创建/snowflake/ip:port-000000000,并上传数据zk_AddressNode = createNode(curator);//worker id 默认是0updateLocalWorkerID(workerID);//定时上报本机时间给forever节点ScheduledUploadData(curator, zk_AddressNode);return true;} else { Map<String, Integer> nodeMap = Maps.newHashMap();//ip:port->00001Map<String, String> realNode = Maps.newHashMap();//ip:port->(ipport-000001)//存在根节点,先检查是否有属于自己的根节点List<String> keys = curator.getChildren().forPath(PATH_FOREVER);for (String key : keys) { String[] nodeKey = key.split("-");realNode.put(nodeKey[0], key);nodeMap.put(nodeKey[0], Integer.parseInt(nodeKey[1]));}Integer workerid = nodeMap.get(listenAddress);if (workerid != null) { //有自己的节点,zk_AddressNode=ip:portzk_AddressNode = PATH_FOREVER + "/" + realNode.get(listenAddress);workerID = workerid;//启动worder时使用会使用if (!checkInitTimeStamp(curator, zk_AddressNode)) { throw new CheckLastTimeException("init timestamp check error,forever node timestamp gt this node time");}//准备创建临时节点doService(curator);updateLocalWorkerID(workerID);LOGGER.info("[Old NODE]find forever node have this endpoint ip-{} port-{} workid-{} childnode and start SUCCESS", ip, port, workerID);} else { //表示新启动的节点,创建持久节点 ,不用check时间String newNode = createNode(curator);zk_AddressNode = newNode;String[] nodeKey = newNode.split("-");workerID = Integer.parseInt(nodeKey[1]);doService(curator);updateLocalWorkerID(workerID);LOGGER.info("[New NODE]can not find node on forever node that endpoint ip-{} port-{} workid-{},create own node on forever node and start SUCCESS ", ip, port, workerID);}}} catch (Exception e) { LOGGER.error("Start node ERROR {}", e);try { Properties properties = new Properties();properties.load(new FileInputStream(new File(PROP_PATH.replace("{port}", port + ""))));workerID = Integer.valueOf(properties.getProperty("workerID"));LOGGER.warn("START FAILED ,use local node file properties workerID-{}", workerID);} catch (Exception e1) { LOGGER.error("Read file error ", e1);return false;}}return true;}
-
逻辑梳理 连接zookeeper if(业务节点为空)创建根节点记录根节点信息 else(业务节点不为空)获取业务节点下的所有子节点,放到map中//ip:port->00001检查有没有自己if 有://这里是防止机器宕机重启,从而可以来获取wokeridworkid 则就是zookeeper的序号else 没有:创建持久有序节点获取workerid
-
workerid是最关键的,其他都是次要
-
再来看雪花算法如何获取id
-
这里从leafController的
getId()
来获取id -
一路跟到SnowflakeIDGenImpl
-
@Overridepublic synchronized Result get(String key) { long timestamp = timeGen();if (timestamp < lastTimestamp) { //时钟回拨发生 这里单位是毫秒long offset = lastTimestamp - timestamp;if (offset <= 5) { try { wait(offset << 1); // 等待最多10mstimestamp = timeGen();if (timestamp < lastTimestamp) { //如果还出现问题,则是无法解决,召唤程序员return new Result(-1, Status.EXCEPTION);}} catch (InterruptedException e) { LOGGER.error("wait interrupted");return new Result(-2, Status.EXCEPTION);}} else { //超过5ms, 直接召唤程序员return new Result(-3, Status.EXCEPTION);}}if (lastTimestamp == timestamp) { //时间相同, 则seq累加呗sequence = (sequence + 1) & sequenceMask;if (sequence == 0) { //seq 为0的时候表示是下一毫秒时间开始对seq做随机sequence = RANDOM.nextInt(100);timestamp = tilNextMillis(lastTimestamp);}} else { //如果是新的ms开始sequence = RANDOM.nextInt(100);//这里的随机启动我有点不太明白}lastTimestamp = timestamp;long id = ((timestamp - twepoch) << timestampLeftShift) | (workerId << workerIdShift) | sequence;//移位处理,组成64位的longreturn new Result(id, Status.SUCCESS);}
-
可以看到, leaf解决时钟回拨的思路就是 延迟等待 , 不行就召唤程序员
-
除了雪花模式,leaf也支持号段模式,因为之前详细分析了tinyid, 所以这里就不再解释了
-
本文的目的也是在关注leaf在雪花模式中如何来生成workerid以及解决时钟回拨问题
总结
- 用zookeeper来生成workerid, 通过 持久有序节点, 保证生成器宕机重启,还能使用之前的workerid
- 延迟等待来解决时间回拨问题
源码地址
https://github.com/Meituan-Dianping/Leaf