Redis（十二）：redis请求转发的实现

请求转发一般的原因为： 1. 该请求自身无法处理，需要转发给对应的服务器处理； 2. 为实现负载均衡，使用路由服务，选择目标实例进行转发；

在集群模式下，请求可以打到任何一台 redis 服务器上。然而并不是所有的服务器都会处理真正的请求，而是只有符合redis slot规则的实例才会处理真正的请求；

这就存在一个情况，当请求打到了一台不应该打到的redis实例上，它应该是要进行转发的。

那么，这个转发该如何做呢？

1. 集群模式下的命令转发如何实现？

// server.c, 在统一处理请求时，会判断出集群模式，进行处理
int processCommand(client *c) {
    ...
    /* If cluster is enabled perform the cluster redirection here.
     * However we don't perform the redirection if:
     * 1) The sender of this command is our master.
     * 2) The command has no key arguments. */
    // 集群模下，根据 hashslot 找到对应的redis节点处理
    if (server.cluster_enabled &&
        !(c->flags & CLIENT_MASTER) &&
        !(c->flags & CLIENT_LUA &&
          server.lua_caller->flags & CLIENT_MASTER) &&
        !(c->cmd->getkeys_proc == NULL && c->cmd->firstkey == 0))
    {
        int hashslot;

        if (server.cluster->state != CLUSTER_OK) {
            flagTransaction(c);
            clusterRedirectClient(c,NULL,0,CLUSTER_REDIR_DOWN_STATE);
            return C_OK;
        } else {
            int error_code;
            // 查找相应的redis节点
            clusterNode *n = getNodeByQuery(c,c->cmd,c->argv,c->argc,&hashslot,&error_code);
            // 除非是应该自己处理的数据，否则响应数据节点不在此处，让客户端另外查找数据节点
            // 因此 redis 节点不做数据转发，只是提示客户再寻找
            // 客户端拿送返回的信息，再向对应的节点发起请求处理
            if (n == NULL || n != server.cluster->myself) {
                flagTransaction(c);
                clusterRedirectClient(c,n,hashslot,error_code);
                return C_OK;
            }
        }
    }
    ...
}

// cluster.c, 查找key对应的redis节点
/* Return the pointer to the cluster node that is able to serve the command.
 * For the function to succeed the command should only target either:
 *
 * 1) A single key (even multiple times like LPOPRPUSH mylist mylist).
 * 2) Multiple keys in the same hash slot, while the slot is stable (no
 *    resharding in progress).
 *
 * On success the function returns the node that is able to serve the request.
 * If the node is not 'myself' a redirection must be perfomed. The kind of
 * redirection is specified setting the integer passed by reference
 * 'error_code', which will be set to CLUSTER_REDIR_ASK or
 * CLUSTER_REDIR_MOVED.
 *
 * When the node is 'myself' 'error_code' is set to CLUSTER_REDIR_NONE.
 *
 * If the command fails NULL is returned, and the reason of the failure is
 * provided via 'error_code', which will be set to:
 *
 * CLUSTER_REDIR_CROSS_SLOT if the request contains multiple keys that
 * don't belong to the same hash slot.
 *
 * CLUSTER_REDIR_UNSTABLE if the request contains multiple keys
 * belonging to the same slot, but the slot is not stable (in migration or
 * importing state, likely because a resharding is in progress).
 *
 * CLUSTER_REDIR_DOWN_UNBOUND if the request addresses a slot which is
 * not bound to any node. In this case the cluster global state should be
 * already "down" but it is fragile to rely on the update of the global state,
 * so we also handle it here. */
clusterNode *getNodeByQuery(client *c, struct redisCommand *cmd, robj **argv, int argc, int *hashslot, int *error_code) {
    clusterNode *n = NULL;
    robj *firstkey = NULL;
    int multiple_keys = 0;
    multiState *ms, _ms;
    multiCmd mc;
    int i, slot = 0, migrating_slot = 0, importing_slot = 0, missing_keys = 0;

    /* Set error code optimistically for the base case. */
    if (error_code) *error_code = CLUSTER_REDIR_NONE;

    /* We handle all the cases as if they were EXEC commands, so we have
     * a common code path for everything */
    if (cmd->proc == execCommand) {
        /* If CLIENT_MULTI flag is not set EXEC is just going to return an
         * error. */
        if (!(c->flags & CLIENT_MULTI)) return myself;
        ms = &c->mstate;
    } else {
        /* In order to have a single codepath create a fake Multi State
         * structure if the client is not in MULTI/EXEC state, this way
         * we have a single codepath below. */
        ms = &_ms;
        _ms.commands = &mc;
        _ms.count = 1;
        mc.argv = argv;
        mc.argc = argc;
        mc.cmd = cmd;
    }

    /* Check that all the keys are in the same hash slot, and obtain this
     * slot and the node associated. */
    for (i = 0; i < ms->count; i++) {
        struct redisCommand *mcmd;
        robj **margv;
        int margc, *keyindex, numkeys, j;

        mcmd = ms->commands[i].cmd;
        margc = ms->commands[i].argc;
        margv = ms->commands[i].argv;
        // 获取所有的 keyIndex, 用于后续依次取 key
        keyindex = getKeysFromCommand(mcmd,margv,margc,&numkeys);
        for (j = 0; j < numkeys; j++) {
            robj *thiskey = margv[keyindex[j]];
            // 计算hashSlot, crc16 算法
            int thisslot = keyHashSlot((char*)thiskey->ptr,
                                       sdslen(thiskey->ptr));

            if (firstkey == NULL) {
                /* This is the first key we see. Check what is the slot
                 * and node. */
                firstkey = thiskey;
                slot = thisslot;
                n = server.cluster->slots[slot];

                /* Error: If a slot is not served, we are in "cluster down"
                 * state. However the state is yet to be updated, so this was
                 * not trapped earlier in processCommand(). Report the same
                 * error to the client. */
                if (n == NULL) {
                    getKeysFreeResult(keyindex);
                    if (error_code)
                        *error_code = CLUSTER_REDIR_DOWN_UNBOUND;
                    return NULL;
                }

                /* If we are migrating or importing this slot, we need to check
                 * if we have all the keys in the request (the only way we
                 * can safely serve the request, otherwise we return a TRYAGAIN
                 * error). To do so we set the importing/migrating state and
                 * increment a counter for every missing key. */
                if (n == myself &&
                    server.cluster->migrating_slots_to[slot] != NULL)
                {
                    migrating_slot = 1;
                } else if (server.cluster->importing_slots_from[slot] != NULL) {
                    importing_slot = 1;
                }
            } else {
                /* If it is not the first key, make sure it is exactly
                 * the same key as the first we saw. */
                if (!equalStringObjects(firstkey,thiskey)) {
                    if (slot != thisslot) {
                        /* Error: multiple keys from different slots. */
                        getKeysFreeResult(keyindex);
                        if (error_code)
                            *error_code = CLUSTER_REDIR_CROSS_SLOT;
                        return NULL;
                    } else {
                        /* Flag this request as one with multiple different
                         * keys. */
                        multiple_keys = 1;
                    }
                }
            }

            /* Migarting / Improrting slot? Count keys we don't have. */
            // 查找0号库是否存在该值，没找到则增加未命中率
            if ((migrating_slot || importing_slot) &&
                lookupKeyRead(&server.db[0],thiskey) == NULL)
            {
                missing_keys++;
            }
        }
        getKeysFreeResult(keyindex);
    }

    /* No key at all in command? then we can serve the request
     * without redirections or errors. */
    if (n == NULL) return myself;

    /* Return the hashslot by reference. */
    if (hashslot) *hashslot = slot;

    /* MIGRATE always works in the context of the local node if the slot
     * is open (migrating or importing state). We need to be able to freely
     * move keys among instances in this case. */
    if ((migrating_slot || importing_slot) && cmd->proc == migrateCommand)
        return myself;

    /* If we don't have all the keys and we are migrating the slot, send
     * an ASK redirection. */
    if (migrating_slot && missing_keys) {
        if (error_code) *error_code = CLUSTER_REDIR_ASK;
        return server.cluster->migrating_slots_to[slot];
    }

    /* If we are receiving the slot, and the client correctly flagged the
     * request as "ASKING", we can serve the request. However if the request
     * involves multiple keys and we don't have them all, the only option is
     * to send a TRYAGAIN error. */
    if (importing_slot &&
        (c->flags & CLIENT_ASKING || cmd->flags & CMD_ASKING))
    {
        if (multiple_keys && missing_keys) {
            if (error_code) *error_code = CLUSTER_REDIR_UNSTABLE;
            return NULL;
        } else {
            return myself;
        }
    }

    /* Handle the read-only client case reading from a slave: if this
     * node is a slave and the request is about an hash slot our master
     * is serving, we can reply without redirection. */
    if (c->flags & CLIENT_READONLY &&
        cmd->flags & CMD_READONLY &&
        nodeIsSlave(myself) &&
        myself->slaveof == n)
    {
        return myself;
    }

    /* Base case: just return the right node. However if this node is not
     * myself, set error_code to MOVED since we need to issue a rediretion. */
    if (n != myself && error_code) *error_code = CLUSTER_REDIR_MOVED;
    return n;
}
// cluster.c, 计算hashSlot, 使用 crc16算法
// 特殊语法: {key_with_hash}key_without_hash
/* We have 16384 hash slots. The hash slot of a given key is obtained
 * as the least significant 14 bits of the crc16 of the key.
 *
 * However if the key contains the {...} pattern, only the part between
 * { and } is hashed. This may be useful in the future to force certain
 * keys to be in the same node (assuming no resharding is in progress). */
unsigned int keyHashSlot(char *key, int keylen) {
    int s, e; /* start-end indexes of { and } */

    for (s = 0; s < keylen; s++)
        if (key[s] == '{') break;

    /* No '{' ? Hash the whole key. This is the base case. */
    if (s == keylen) return crc16(key,keylen) & 0x3FFF;

    /* '{' found? Check if we have the corresponding '}'. */
    for (e = s+1; e < keylen; e++)
        if (key[e] == '}') break;

    /* No '}' or nothing betweeen {} ? Hash the whole key. */
    if (e == keylen || e == s+1) return crc16(key,keylen) & 0x3FFF;

    /* If we are here there is both a { and a } on its right. Hash
     * what is in the middle between { and }. */
    return crc16(key+s+1,e-s-1) & 0x3FFF;
}


// 根据状态值，响应客户端，数据节点不在本节点
/* Send the client the right redirection code, according to error_code
 * that should be set to one of CLUSTER_REDIR_* macros.
 *
 * If CLUSTER_REDIR_ASK or CLUSTER_REDIR_MOVED error codes
 * are used, then the node 'n' should not be NULL, but should be the
 * node we want to mention in the redirection. Moreover hashslot should
 * be set to the hash slot that caused the redirection. */
void clusterRedirectClient(client *c, clusterNode *n, int hashslot, int error_code) {
    if (error_code == CLUSTER_REDIR_CROSS_SLOT) {
        addReplySds(c,sdsnew("-CROSSSLOT Keys in request don't hash to the same slot\r\n"));
    } else if (error_code == CLUSTER_REDIR_UNSTABLE) {
        /* The request spawns mutliple keys in the same slot,
         * but the slot is not "stable" currently as there is
         * a migration or import in progress. */
        addReplySds(c,sdsnew("-TRYAGAIN Multiple keys request during rehashing of slot\r\n"));
    } else if (error_code == CLUSTER_REDIR_DOWN_STATE) {
        addReplySds(c,sdsnew("-CLUSTERDOWN The cluster is down\r\n"));
    } else if (error_code == CLUSTER_REDIR_DOWN_UNBOUND) {
        addReplySds(c,sdsnew("-CLUSTERDOWN Hash slot not served\r\n"));
    } else if (error_code == CLUSTER_REDIR_MOVED ||
               error_code == CLUSTER_REDIR_ASK)
    {
        // 当对应的数据节点不是自身，而且已经找到了应当处理的节点时，响应客户端对应信息
        // ASK错误说明数据正在迁移，不知道何时迁移完成，因此重定向是临时的，不应刷新slot缓存
        // MOVED错误重定向则是(相对)永久的，应刷新slot缓存
        addReplySds(c,sdscatprintf(sdsempty(),
            "-%s %d %s:%d\r\n",
            (error_code == CLUSTER_REDIR_ASK) ? "ASK" : "MOVED",
            hashslot,n->ip,n->port));
    } else {
        serverPanic("getNodeByQuery() unknown error.");
    }
}

所以，redis集群模式下的请求转发，并非redis服务端直接转发请求，而是通过向客户端响应 转移指令，由客户端重新发起目标请求，从而实现命令转发的。

其实，redis做响应转移的处理，应只会发生在redis节点发生变更的时候，比如增加节点或减少节点时，redis为实现数据再均衡，才会出现。正常情况下，具体哪个数据应该请求向哪个redis节点，则完全由客户端负责。这也是集群的优势所在，各个数据节点只处理对应的范围数据。因此，需要客户端将服务端的slot存放规则或者位置缓存起来(通过 cluster slots 可以获取槽存放信息)，从而实现向正确的节点请求操作。

2. 主从模式的命令转发如何实现？

主从模式下，只有主节点可以写请求，而从节点则负责同步主节点的数据即可。然而，在我们做读写分离的时候，从节点是可以承受读流量的。但是，如果写流程打到了从节点上，这是否又涉及到一个请求转发呢？我们来看一下：

// 主从的命令处理判断，也是在 processCommand 中统一处理的
int processCommand(client *c) {
    ...
    /* Don't accept write commands if this is a read only slave. But
     * accept write commands if this is our master. */
    // 针对从节点，只能接受读请求，如果是写请求，直接响应
    if (server.masterhost && server.repl_slave_ro &&
        // master 请求除外，因为master过来的请求，是用于同步数据的
        !(c->flags & CLIENT_MASTER) &&
        c->cmd->flags & CMD_WRITE)
    {
        // -READONLY You can't write against a read only slave.
        addReply(c, shared.roslaveerr);
        return C_OK;
    }
    ...
    return C_OK;
}

所以，redis主从模式下，服务端并不做转发处理。而要实现读写分离的功能，必然要客户端自行处理了。比如要自行定位master节点，然后将写请求发送过去，读请求则可以做负载均衡处理。这也是很多数据库中间件的职责所在。

3. 如何使用redis集群？

redis集群，本质上提供了数据的分片存储能力（当然要实现这个功能有相当多的工作要做），但是访问数据需要客户端自行处理。所以，我们以jedis作为客户端，看看客户端都是如何利用集群的吧！测试用例如下：

    @Test
    public void testCluster() throws Exception {
        // 添加集群的服务节点Set集合
        Set<HostAndPort> hostAndPortsSet = new HashSet<HostAndPort>();
        // 添加节点
        hostAndPortsSet.add(new HostAndPort("192.168.1.103", 7000));
        hostAndPortsSet.add(new HostAndPort("192.168.1.103", 7001));
        hostAndPortsSet.add(new HostAndPort("192.168.1.103", 8000));
        hostAndPortsSet.add(new HostAndPort("192.168.1.103", 8001));
        hostAndPortsSet.add(new HostAndPort("192.168.1.103", 9000));
        hostAndPortsSet.add(new HostAndPort("192.168.1.103", 9001));

        // Jedis连接池配置
        JedisPoolConfig jedisPoolConfig = new JedisPoolConfig();
        // 最大空闲连接数, 默认8个
        jedisPoolConfig.setMaxIdle(5);
        // 最大连接数, 默认8个
        jedisPoolConfig.setMaxTotal(10);
        //最小空闲连接数, 默认0
        jedisPoolConfig.setMinIdle(0);
        // 获取连接时的最大等待毫秒数(如果设置为阻塞时BlockWhenExhausted),如果超时就抛异常, 小于零:阻塞不确定的时间,  默认-1
        jedisPoolConfig.setMaxWaitMillis(2000);
        //对拿到的connection进行validateObject校验
        jedisPoolConfig.setTestOnBorrow(true);
        // JedisCluster 会继承 JedisSlotBasedConnectionHandler, 即会处理 slot 定位问题
        JedisCluster jedis = new JedisCluster(hostAndPortsSet, jedisPoolConfig);
        String key = "key1";
        String value = "Value1";
        jedis.set(key, value);
        System.out.println("set a value to Redis over. " + key + "->" + value);
        value = jedis.get("key1");
        System.out.println("get a value from Redis over. " + key + "->" + value);
        jedis.close();
    }

如上，就是jedis访问redis集群的方式了，sdk封装之后的应用，总是简单易用。主要就是通过 JedisCluster 进行访问即可。而与单机的redis访问的很大不同点，是在于数据key的定位上，我们可以详细看看。

如下是 JedisCluster 的类继承图：

与之对比的是 Jedis 的类继承图：

它们两个都实现的接口有: BasicCommands, Closeable, JedisCommands.

可见，cluster下的redis操作上，与普通的redis还是有许多不同的。不过，我们只想探讨的是，key如何定位的问题，所以一个set/get就够了。

    // JedisCluster 初始化时会初始化 slot 信息到本地缓存中
    // redis.clients.jedis.JedisClusterConnectionHandler#JedisClusterConnectionHandler
  public JedisClusterConnectionHandler(Set<HostAndPort> nodes,
                                       final GenericObjectPoolConfig poolConfig, int connectionTimeout, int soTimeout, String password) {
    this.cache = new JedisClusterInfoCache(poolConfig, connectionTimeout, soTimeout, password);
    // 在初始化 JedisCluster 时，会先触发一次 slot 信息的拉取，以备后续使用
    initializeSlotsCache(nodes, poolConfig, password);
  }
  private void initializeSlotsCache(Set<HostAndPort> startNodes, GenericObjectPoolConfig poolConfig, String password) {
    for (HostAndPort hostAndPort : startNodes) {
      Jedis jedis = new Jedis(hostAndPort.getHost(), hostAndPort.getPort());
      if (password != null) {
        jedis.auth(password);
      }
      try {
        // 只要某个节点成功响应，就够了
        // 遍历的目的，是为了高可用保证，为了避免某些节点故障而拿不到信息
        cache.discoverClusterNodesAndSlots(jedis);
        break;
      } catch (JedisConnectionException e) {
        // try next nodes
      } finally {
        if (jedis != null) {
          jedis.close();
        }
      }
    }
  }

    // set 的操作，则是使用 JedisClusterCommand 包装了一层 Jedis
    // redis.clients.jedis.JedisCluster#set(java.lang.String, java.lang.String)
  @Override
  public String set(final String key, final String value) {
      // connectionHandler 是 JedisSlotBasedConnectionHandler 的实例
      // 默认重试次数: 5
    return new JedisClusterCommand<String>(connectionHandler, maxAttempts) {
      @Override
      public String execute(Jedis connection) {
        return connection.set(key, value);
      }
    }.run(key);
  }
  // redis.clients.jedis.JedisClusterCommand#run(java.lang.String)
  public T run(String key) {
    if (key == null) {
      throw new JedisClusterException("No way to dispatch this command to Redis Cluster.");
    }

    return runWithRetries(SafeEncoder.encode(key), this.maxAttempts, false, false);
  }
  // 带重试的访问 redis 节点， 重试的场景有：数据节点不在访问节点; 访问的节点正在进行数据迁移; 访问节点不可用;
  // redis.clients.jedis.JedisClusterCommand#runWithRetries
  private T runWithRetries(byte[] key, int attempts, boolean tryRandomNode, boolean asking) {
    if (attempts <= 0) {
      throw new JedisClusterMaxRedirectionsException("Too many Cluster redirections?");
    }

    Jedis connection = null;
    try {

      if (asking) {
        // TODO: Pipeline asking with the original command to make it
        // faster....
        connection = askConnection.get();
        connection.asking();

        // if asking success, reset asking flag
        asking = false;
      } else {
        if (tryRandomNode) {
          connection = connectionHandler.getConnection();
        } else {
            // 直接调用 connectionHandler.getConnectionFromSlot 获取对应的redis连接
            // 此处计算的 slot 就是redis服务端实现的那套 crc16 % 0x3FFF, 即各端保持一致，就可以做出相同的判定了
          connection = connectionHandler.getConnectionFromSlot(JedisClusterCRC16.getSlot(key));
        }
      }

      return execute(connection);

    } catch (JedisNoReachableClusterNodeException jnrcne) {
      throw jnrcne;
    } catch (JedisConnectionException jce) {
      // release current connection before recursion
      releaseConnection(connection);
      connection = null;

      if (attempts <= 1) {
        //We need this because if node is not reachable anymore - we need to finally initiate slots renewing,
        //or we can stuck with cluster state without one node in opposite case.
        //But now if maxAttempts = 1 or 2 we will do it too often. For each time-outed request.
        //TODO make tracking of successful/unsuccessful operations for node - do renewing only
        //if there were no successful responses from this node last few seconds
        this.connectionHandler.renewSlotCache();

        //no more redirections left, throw original exception, not JedisClusterMaxRedirectionsException, because it's not MOVED situation
        throw jce;
      }
        // 连接异常，再次请求随机节点
      return runWithRetries(key, attempts - 1, tryRandomNode, asking);
    } catch (JedisRedirectionException jre) {
      // if MOVED redirection occurred,
      if (jre instanceof JedisMovedDataException) {
        // it rebuilds cluster's slot cache
        // recommended by Redis cluster specification
        this.connectionHandler.renewSlotCache(connection);
      }

      // release current connection before recursion or renewing
      releaseConnection(connection);
      connection = null;

      if (jre instanceof JedisAskDataException) {
        asking = true;
        askConnection.set(this.connectionHandler.getConnectionFromNode(jre.getTargetNode()));
      } else if (jre instanceof JedisMovedDataException) {
      } else {
        throw new JedisClusterException(jre);
      }
        // 收到 MOVED/ASK 响应，刷新slot信息后，重新再访问
      return runWithRetries(key, attempts - 1, false, asking);
    } finally {
      releaseConnection(connection);
    }
  }
  // 计算hashSlot值
  // redis.clients.util.JedisClusterCRC16#getSlot(byte[])
  public static int getSlot(byte[] key) {
    int s = -1;
    int e = -1;
    boolean sFound = false;
    for (int i = 0; i < key.length; i++) {
      if (key[i] == '{' && !sFound) {
        s = i;
        sFound = true;
      }
      if (key[i] == '}' && sFound) {
        e = i;
        break;
      }
    }
    if (s > -1 && e > -1 && e != s + 1) {
      return getCRC16(key, s + 1, e) & (16384 - 1);
    }
    return getCRC16(key) & (16384 - 1);
  }
  // 根据hashSlot, 得到对应的 redis 连接实例
  @Override
  public Jedis getConnectionFromSlot(int slot) {
      // 先从缓存中获取slot对应的连接信息，初始时自然是空的
    JedisPool connectionPool = cache.getSlotPool(slot);
    if (connectionPool != null) {
      // It can't guaranteed to get valid connection because of node
      // assignment
      return connectionPool.getResource();
    } else {
        // 刷新slot缓存信息，大概就是请求 cluster slot, 获取slot的分布信息，然后存入JedisClusterInfoCache中
      renewSlotCache(); //It's abnormal situation for cluster mode, that we have just nothing for slot, try to rediscover state
      connectionPool = cache.getSlotPool(slot);
      // 如果还是获取不到，则随机选择一个连接
      // 此时请求该随机节点，服务端有可能会响应正确的节点位置信息
      if (connectionPool != null) {
        return connectionPool.getResource();
      } else {
        //no choice, fallback to new connection to random node
        return getConnection();
      }
    }
  }
    // redis.clients.jedis.JedisClusterConnectionHandler#renewSlotCache()
  public void renewSlotCache() {
    cache.renewClusterSlots(null);
  }
  // redis.clients.jedis.JedisClusterInfoCache#renewClusterSlots
  public void renewClusterSlots(Jedis jedis) {
    //If rediscovering is already in process - no need to start one more same rediscovering, just return
    if (!rediscovering) {
      try {
        w.lock();
        rediscovering = true;

        if (jedis != null) {
          try {
            discoverClusterSlots(jedis);
            return;
          } catch (JedisException e) {
            //try nodes from all pools
          }
        }
        // 依次遍历集群节点，直到有一个正确的响应为止
        for (JedisPool jp : getShuffledNodesPool()) {
          try {
            jedis = jp.getResource();
            discoverClusterSlots(jedis);
            return;
          } catch (JedisConnectionException e) {
            // try next nodes
          } finally {
            if (jedis != null) {
              jedis.close();
            }
          }
        }
      } finally {
        rediscovering = false;
        w.unlock();
      }
    }
  }
  
  private void discoverClusterSlots(Jedis jedis) {
    // 发送 cluster slots, 命令，获取 slot 分布信息
    List<Object> slots = jedis.clusterSlots();
    this.slots.clear();

    for (Object slotInfoObj : slots) {
      List<Object> slotInfo = (List<Object>) slotInfoObj;

    /* Format: 1) 1) start slot
     *            2) end slot
     *            3) 1) master IP
     *               2) master port
     *               3) node ID
     *            4) 1) replica IP
     *               2) replica port
     *               3) node ID
     *           ... continued until done
     */
      if (slotInfo.size() <= MASTER_NODE_INDEX) {
        continue;
      }

      List<Integer> slotNums = getAssignedSlotArray(slotInfo);

      // hostInfos
      // 第三个元素是 master 信息
      List<Object> hostInfos = (List<Object>) slotInfo.get(MASTER_NODE_INDEX);
      if (hostInfos.isEmpty()) {
        continue;
      }

      // at this time, we just use master, discard slave information
      HostAndPort targetNode = generateHostAndPort(hostInfos);
      // 只存储master信息
      assignSlotsToNode(slotNums, targetNode);
    }
  }

  private List<Integer> getAssignedSlotArray(List<Object> slotInfo) {
    List<Integer> slotNums = new ArrayList<Integer>();
    // 依次将所管辖slot范围，添加到列表中
    // 如 0 ~ 5999
    for (int slot = ((Long) slotInfo.get(0)).intValue(); slot <= ((Long) slotInfo.get(1))
        .intValue(); slot++) {
      slotNums.add(slot);
    }
    return slotNums;
  }
  // 将所有给定的 slot, 放到 targetNode 的管辖范围，方便后续获取
  // redis.clients.jedis.JedisClusterInfoCache#assignSlotsToNode
  public void assignSlotsToNode(List<Integer> targetSlots, HostAndPort targetNode) {
    // 此处的锁为读写锁 ReentrantReadWriteLock 中的 writeLock
    w.lock();
    try {
        // 创建redis连接
      JedisPool targetPool = setupNodeIfNotExist(targetNode);
      // 依次将范围内的slot指向 targetNode
      // 正常情况下，slots的大小应该都是16384
      for (Integer slot : targetSlots) {
        // slots = new HashMap<Integer, JedisPool>();
        slots.put(slot, targetPool);
      }
    } finally {
      w.unlock();
    }
  }
  // redis.clients.jedis.JedisClusterInfoCache#setupNodeIfNotExist(redis.clients.jedis.HostAndPort)
  public JedisPool setupNodeIfNotExist(HostAndPort node) {
    w.lock();
    try {
      String nodeKey = getNodeKey(node);
      JedisPool existingPool = nodes.get(nodeKey);
      if (existingPool != null) return existingPool;

      JedisPool nodePool = new JedisPool(poolConfig, node.getHost(), node.getPort(),
          connectionTimeout, soTimeout, password, 0, null, false, null, null, null);
      nodes.put(nodeKey, nodePool);
      return nodePool;
    } finally {
      w.unlock();
    }
  }
  // 刷新slot缓存信息后，再重新请求获取redis连接就简单了
  // redis.clients.jedis.JedisClusterInfoCache#getSlotPool
  public JedisPool getSlotPool(int slot) {
    r.lock();
    try {
      return slots.get(slot);
    } finally {
      r.unlock();
    }
  }

从上面的描述，我们清楚了整个客户如何处理集群请求的。整体就两个步骤： 1. 通过 cluster slot 获取redis集群的slot分布信息，然后缓存到本地； 2. 根据slot分布信息，向对应的redis节点发起请求即可。

另外，还有些意外情况，即客户端拿到的 slot 信息如果是错误的怎么办？如何保持客户端缓存与服务端的一致性？

事实上，客户端既不保证slot信息的准确性，也不保证与服务端数据的一致性，而是在发生错误的时候，再进行刷新即可。通过 JedisClusterCommand#runWithRetries, 进行错误重试，slot数据刷新。

4. 通常的请求转发如何实现？

可以看到，redis实际上一直避开了转发这个问题。

那么，实际中，我们的转发工作都是如何实现的呢？

最简单的，接收到客户端的请求之后，将数据重新封装好，然后构建一个目标地址的新请求，发送过去，然后等待结果响应。当目标服务器响应后，再将结果响应给客户端即可。如：应用网关、代理服务器；

其次，是响应客户端一个状态码(如302)，让客户端自主进行跳转。这和redis实现倒是如出一辙；

相对复杂的，直接使用流进行对接，接收到客户端的请求后，直接将数据传到目标服务器，同样，目标服务器响应后，直接将数据写入客户端通道即可。这种情况避免大量数据的重新封装，极大减少了转发带来的性能损失，从而提高响应速度。这种场景，一般用于传输大文件。