The LVS Server Load balancer environment is set up, and SYN_RECV state processing is successful when the LVS + KEEPALIVED environment is set up for the first time. After a while, a problem occurs when the environment is set up again. No matter how you configure and modify parameters, the client is always unable to connect to the realserver. view the result through ipvsadm-lc: www.2cto.com CP SYN_RECV h100: 12949 192.168.4.200: 5678 h104: 5678, I finally found the cause of the problem. Here I despise Baidu and found no resources available at all. The reason is as follows: I chose the IP address specified for listening to the Service Listening IP address of realserver, that is, the IP address of eth0, but did not listen to lo, that is, the local address, even if you receive a request forwarded by the directoryserver, lo: 0 is locally broadcast to the local machine. As the service does not listen to lo, no response is returned. Cup, it took me so much time .................. the ifconfig result of realserver is as follows: eth0 Link encap: Ethernet HWaddr 00: 30: 48: C8: 7D: CA inet addr: 192.168.10.55 Bcast: 192.168.15.255 Mask: 255.255.240.0 inet6 addr: fe80 :: 230: 48ff: fec8: 7dca/64 Scope: Link up broadcast running multicast mtu: 1500 Metric: 1 RX packets: 5369998 errors: 0 dropped: 0 overruns: 0 frame: 0 TX packets: 357128 errors: 0 dropped: 0 overruns: 0 carrier: 0 collisions: 0 txqueuelen: 1000 RX bytes: 493767975 (470.8 MiB) TX bytes: 181277721 (172.8 MiB) memory: fbee0000-fbf00000lo Link encap: Local Loopback inet addr: 127.0.0.1 Mask: 255.0.0.0 inet6 addr ::: 1/128 Scope: Host up loopback running mtu: 16436 Metric: 1 RX packets: 11414 errors: 0 dropped: 0 overruns: 0 frame: 0 TX packets: 11414 errors: 0 dropped: 0 overruns: 0 carrier: 0 collisions: 0 txqueuelen: 0 RX bytes: 3994708 (3.8 MiB) TX bytes: 3994708 (3.8 MiB) lo: 0 Link encap: Local Loopback inet addr: 192.168.4.200 Mask: 255.255.255 up loopback running mtu: 16436 Metric: 1