LVS 内核实现分析（3）

4. LVS内核源码分析------NAT模式下TCP报文

  前面已经讲过了LVS支持三种模式，其中NAT模式的应用比较广泛，该种模式下的配置仅仅涉及到LVS服务器，Real Server上不需要进行额外的配置。

4.1  LVS的HOOK点函数Linux2.6.32

  在内核中IPVS的源码在net/netfilter/ipvs目录下，LVS是以netfilter框架为基础，先看一下LVS在哪些HOOK点挂载了自己的处理函数。IPVS的处理函数在挂载在下面三个HOOK点，NF_INET_LOCAL_IN, NF_INET_FORWARD，NF_INET_POST_ROUTING，LVS同时支持IPv4/IPv6两种协议。

static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
    /* After packet filtering, forward packet through VS/DR, VS/TUN,
     * or VS/NAT(change destination), so that filtering rules can be
     * applied to IPVS. */
    {
        .hook        = ip_vs_in,
        .owner        = THIS_MODULE,
        .pf        = PF_INET,
        .hooknum = NF_INET_LOCAL_IN,
        .priority = 100,
    },
    /* After packet filtering, change source only for VS/NAT */
    {
        .hook        = ip_vs_out,
        .owner        = THIS_MODULE,
        .pf        = PF_INET,
        .hooknum = NF_INET_FORWARD,
        .priority = 100,
    },
    /* After packet filtering (but before ip_vs_out_icmp), catch icmp
     * destined for 0.0.0.0/0, which is for incoming IPVS connections */
    {
        .hook        = ip_vs_forward_icmp,
        .owner        = THIS_MODULE,
        .pf        = PF_INET,
        .hooknum = NF_INET_FORWARD,
        .priority = 99,
    },
    /* Before the netfilter connection tracking, exit from POST_ROUTING */
    {
        .hook        = ip_vs_post_routing,
        .owner        = THIS_MODULE,
        .pf        = PF_INET,
        .hooknum = NF_INET_POST_ROUTING,
        .priority = NF_IP_PRI_NAT_SRC-1,
    },
#ifdef CONFIG_IP_VS_IPV6
    /* After packet filtering, forward packet through VS/DR, VS/TUN,
     * or VS/NAT(change destination), so that filtering rules can be
     * applied to IPVS. */
    {
        .hook        = ip_vs_in,
        .owner        = THIS_MODULE,
        .pf        = PF_INET6,
        .hooknum = NF_INET_LOCAL_IN,
        .priority = 100,
    },
    /* After packet filtering, change source only for VS/NAT */
    {
        .hook        = ip_vs_out,
        .owner        = THIS_MODULE,
        .pf        = PF_INET6,
        .hooknum = NF_INET_FORWARD,
        .priority = 100,
    },
    /* After packet filtering (but before ip_vs_out_icmp), catch icmp
     * destined for 0.0.0.0/0, which is for incoming IPVS connections */
    {
        .hook        = ip_vs_forward_icmp_v6,
        .owner        = THIS_MODULE,
        .pf        = PF_INET6,
        .hooknum = NF_INET_FORWARD,
        .priority = 99,
    },
    /* Before the netfilter connection tracking, exit from POST_ROUTING */
    {
        .hook        = ip_vs_post_routing,
        .owner        = THIS_MODULE,
        .pf        = PF_INET6,
        .hooknum = NF_INET_POST_ROUTING,
        .priority = NF_IP6_PRI_NAT_SRC-1,
    },
#endif
};

4.2  ip_vs_in函数

当远端的客户端发送数据到达服务器时，也就是LVS Server时，该数据包的目的IP地址就是到达本地的，所以会先进入NF_LOCAL_IN的HOOK点，进行本地处理。

static unsigned int
ip_vs_in(unsigned int hooknum, struct sk_buff *skb,
     const struct net_device *in, const struct net_device *out,
     int (*okfn)(struct sk_buff *))
{
    struct ip_vs_iphdr iph;
    struct ip_vs_protocol *pp;
    struct ip_vs_conn *cp;
    int ret, restart, af, pkts;
   //判断是IPV4还是IPV6协议
    af = (skb->protocol == htons(ETH_P_IP)) ? AF_INET : AF_INET6;
   /*获取到IP头*/
    ip_vs_fill_iphdr(af, skb_network_header(skb), &iph);

    /*
     *    Big tappo: only PACKET_HOST, including loopback for local client
     *    Don't handle local packets on IPv6 for now
     判断是否是达到本机的*/
    if (unlikely(skb->pkt_type != PACKET_HOST)) {
        IP_VS_DBG_BUF(12, "packet type=%d proto=%d daddr=%s ignored\n",
                  skb->pkt_type,
                  iph.protocol,
                  IP_VS_DBG_ADDR(af, &iph.daddr));
        return NF_ACCEPT;
    }
/*对IPV6的判断忽略*/
#ifdef CONFIG_IP_VS_IPV6
    if (af == AF_INET6) {
        if (unlikely(iph.protocol == IPPROTO_ICMPV6)) {
            int related, verdict = ip_vs_in_icmp_v6(skb, &related, hooknum);

            if (related)
                return verdict;
            ip_vs_fill_iphdr(af, skb_network_header(skb), &iph);
        }
    } else
#endif
/*判断是否是ICMP报文，*/
        if (unlikely(iph.protocol == IPPROTO_ICMP)) {
            int related, verdict = ip_vs_in_icmp(skb, &related, hooknum);

            if (related)
                return verdict;
            ip_vs_fill_iphdr(af, skb_network_header(skb), &iph);
        }

    /* Protocol supported? 判断LVS支持的协议，目前支持TCP、UDP、sctp、ah、esp    */
    pp = ip_vs_proto_get(iph.protocol);
    if (unlikely(!pp))
        return NF_ACCEPT;

    /*
     * Check if the packet belongs to an existing connection entry，检查该数据包是否已经属于已经存在的连接实例，对于第一个包，这里返回NULL，这里对应的回调函数为ip_vs_conn_in_get_proto，该函数主要是根据源IP，目的IP，源端口和目的端口,协议号来查找。 */
    cp = pp->conn_in_get(af, skb, pp, &iph, iph.len, 0);
/*对于新到来的数据包包，查找不到cp为空，进入下面的if处理*/
    if (unlikely(!cp)) {
        int v;

        /* For local client packets, it could be a response 对于TCP这里调用的函数为ip_vs_conn_out_get_proto  新的内核版本已经没有了response的处理，这里不作分析*/
        cp = pp->conn_out_get(af, skb, pp, &iph, iph.len, 0);
        if (cp)
            return handle_response(af, skb, pp, cp, iph.len); /*这里查找是否是出去的报文，if (res_dir == IP_VS_CIDX_F_IN2OUT) {
            return handle_response(af, skb, pp, cp, iph.len);
        } 做SNAT*/
    /*没有查找到connection ,创建新的connection ，这里的函数为tcp_conn_schedule */
        if (!pp->conn_schedule(af, skb, pp, &v, &cp))
            return v;
    }

    if (unlikely(!cp)) {
        /* sorry, all this trouble for a no-hit :) 到这里说明创建也没有成功，直接返回ACCEPT*/
        IP_VS_DBG_PKT(12, pp, skb, 0,
             "packet continues traversal as normal");
        return NF_ACCEPT;
    }

    IP_VS_DBG_PKT(11, pp, skb, 0, "Incoming packet");

    /* Check the server status 检查server的状态，如果cp->dest不为空，但是服务器不可用，则直接丢弃该数据包*/
    if (cp->dest && !(cp->dest->flags & IP_VS_DEST_F_AVAILABLE)) {
        /* the destination server is not available */

        if (sysctl_ip_vs_expire_nodest_conn) {
            /* try to expire the connection immediately */
            ip_vs_conn_expire_now(cp);
        }
        /* don't restart its timer, and silently
           drop the packet. 引用计数减一 */
        __ip_vs_conn_put(cp);
        return NF_DROP;
    }
   /*更新统计和状态信息*/
    ip_vs_in_stats(cp, skb);
  /*设置现在的状态为IP_VS_DIR_INPUT,调用的函数为，最终调用的函数为tcp_state_transition */
    restart = ip_vs_set_state(cp, IP_VS_DIR_INPUT, skb, pp);
    if (cp->packet_xmit)
        ret = cp->packet_xmit(skb, cp, pp); /*调用发包函数，对于NAT模式来说这里的发包函数为ip_vs_nat_xmit */
        /* do not touch skb anymore */
    else {
        IP_VS_DBG_RL("warning: packet_xmit is null");
        ret = NF_ACCEPT;
    }

    /* Increase its packet counter and check if it is needed
     * to be synchronized
     *
     * Sync connection if it is about to close to
     * encorage the standby servers to update the connections timeout
      增加发送的包的数量，并进行同步*/
    pkts = atomic_add_return(1, &cp->in_pkts);
    if (af == AF_INET && (ip_vs_sync_state & IP_VS_STATE_MASTER) &&
        cp->protocol == IPPROTO_SCTP) {
        if ((cp->state == IP_VS_SCTP_S_ESTABLISHED &&
            (atomic_read(&cp->in_pkts) %
             sysctl_ip_vs_sync_threshold[1]
             == sysctl_ip_vs_sync_threshold[0])) ||
                (cp->old_state != cp->state &&
                 ((cp->state == IP_VS_SCTP_S_CLOSED) ||
                 (cp->state == IP_VS_SCTP_S_SHUT_ACK_CLI) ||
                 (cp->state == IP_VS_SCTP_S_SHUT_ACK_SER)))) {
            ip_vs_sync_conn(cp);//这里进行同步，同步主要是同步备份LVS server，这里是针对SCTP协议
            goto out;
        }
    }

    if (af == AF_INET &&
     (ip_vs_sync_state & IP_VS_STATE_MASTER) &&
     (((cp->protocol != IPPROTO_TCP ||
           cp->state == IP_VS_TCP_S_ESTABLISHED) &&
     (pkts % sysctl_ip_vs_sync_threshold[1]
     == sysctl_ip_vs_sync_threshold[0])) ||
     ((cp->protocol == IPPROTO_TCP) && (cp->old_state != cp->state) &&
     ((cp->state == IP_VS_TCP_S_FIN_WAIT) ||
     (cp->state == IP_VS_TCP_S_CLOSE_WAIT) ||
     (cp->state == IP_VS_TCP_S_TIME_WAIT)))))
        ip_vs_sync_conn(cp); //这里进行同步，同步主要是同步备份LVS server，这里是针对TCP协议

out:
    cp->old_state = cp->state;

    ip_vs_conn_put(cp);
    return ret;
}

4.2.1  tcp_conn_schedule

在该函数中调用调度策略函数，按照调度策略找到真正的real Server。

static int
tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp,
         int *verdict, struct ip_vs_conn **cpp)
{
    struct ip_vs_service *svc;
    struct tcphdr _tcph, *th;
    struct ip_vs_iphdr iph;

    ip_vs_fill_iphdr(af, skb_network_header(skb), &iph);

    th = skb_header_pointer(skb, iph.len, sizeof(_tcph), &_tcph);
    if (th == NULL) {
        *verdict = NF_DROP;
        return 0;
    }
 /*根据协议号和目的地址、目的端口号查找到ip_vs_service 实例*/
    if (th->syn &&
     (svc = ip_vs_service_get(af, skb->mark, iph.protocol, &iph.daddr,
                     th->dest))) {
        if (ip_vs_todrop()) {
            /*
             * It seems that we are very loaded.
        &nbs