1. I can start nodes(total six nodes);
  2. I run them on cloud servers;
  3. then, I execute command : -a "Iwhalecloud@2022redis" \ Ip1:port1\ Ip1:port 2\ Ip2:port1 \ Ip2:port2 \ Ip3:port1 \ Ip3:port2 it can create cluster successfully，but when I pressure measurement reids cluster by my program， throw Too many Cluster redirections excpetion. IP is outer net ip ,not Intranet IP .

lastly, I found node.conf is wrong。

in every node.conf myself node ip become Intranet IP. this cause sometimes my program cannot connect it, so throw Too many Cluster redirections excpetion

for example , execute command[cluster nodes] in IP3 ECS （port1）。 like this ：

ef911780081a6feb0f01b4005413bcc42385ac44 IP1:port1@16379 slave b2738cf1475321bd4478d7267f5a8b5bca53987e 0 1655977440435 5 connected de1c2b96baadb04faf87c7c98927f21e9a56a4e9 IP2:port2@16379 slave fde3323af28e48dfb428e49d481ae0962e9761a2 0 1655977439433 1 connected a009a6b2698e5c106ca5dc14a598c7484df7a9a8 IP2:port1@16378 master - 0 1655977438132 3 connected 5461-10922 9e8bae64fd2da60333686fdf421eca2e6e1e615c IP2:port2@16379 slave a009a6b2698e5c106ca5dc14a598c7484df7a9a8 0 1655977439000 3 connected fde3323af28e48dfb428e49d481ae0962e9761a2 172.23.159.24:port1@16378 myself,master - 0 1655977438000 1 connected 0-5460 b2738cf1475321bd4478d7267f5a8b5bca53987e IP3:port2@16378 master - 0 1655977438133 5 connected 10923-16383

or example , execute command[cluster nodes] in IP3 ECS （port2）。 like this ：

ef911780081a6feb0f01b4005413bcc42385ac44 IP1:port1@16379 slave b2738cf1475321bd4478d7267f5a8b5bca53987e 0 1655977440435 5 connected de1c2b96baadb04faf87c7c98927f21e9a56a4e9 IP2:port2@16379 slave fde3323af28e48dfb428e49d481ae0962e9761a2 0 1655977439433 1 connected a009a6b2698e5c106ca5dc14a598c7484df7a9a8 IP2:port1@16378 master - 0 1655977438132 3 connected 5461-10922 9e8bae64fd2da60333686fdf421eca2e6e1e615c IP2:port2@16379 slave a009a6b2698e5c106ca5dc14a598c7484df7a9a8 0 1655977439000 3 connected fde3323af28e48dfb428e49d481ae0962e9761a2 IP3:port1@16378 master - 0 1655977438000 1 connected 0-5460 b2738cf1475321bd4478d7267f5a8b5bca53987e 172.23.159.24:port2@16378 myself,master - 0 1655977438133 5 connected 10923-16383

or example , execute command[cluster nodes] in IP2 ECS （port2）。 like this ：

ef911780081a6feb0f01b4005413bcc42385ac44 IP1:port1@16379 slave b2738cf1475321bd4478d7267f5a8b5bca53987e 0 1655977440435 5 connected de1c2b96baadb04faf87c7c98927f21e9a56a4e9 IP2:port2@16379 slave fde3323af28e48dfb428e49d481ae0962e9761a2 0 1655977439433 1 connected a009a6b2698e5c106ca5dc14a598c7484df7a9a8 IP2:port1@16378 master - 0 1655977438132 3 connected 5461-10922 9e8bae64fd2da60333686fdf421eca2e6e1e615c 172.23.159.22:port2@16379 myself,slave a009a6b2698e5c106ca5dc14a598c7484df7a9a8 0 1655977439000 3 connected

fde3323af28e48dfb428e49d481ae0962e9761a2 IP3:port1@16378 master - 0 1655977438000 1 connected 0-5460 b2738cf1475321bd4478d7267f5a8b5bca53987e IP3:port2@16378 master - 0 1655977438133 5 connected 10923-16383

now ，I stop six redis nodes, update intranet IP -> out net Ip in node.conf , restart them . my program can work correctly.

so I want to know how I can solve the question elegantly. I don‘t know why

Comment From: uvletter

when you build the cluster, try cluster meet with a WAN IP instead of LAN IP?

then, I execute command : -a "Iwhalecloud@2022redis"

I'm not sure what the command is

Comment From: yetian12

构建集群时，尝试cluster meet使用 WAN IP 而不是 LAN IP？

然后，我执行命令：-a "Iwhalecloud@2022redis"

我不确定命令是什么

redis-cli --cluster create \ --cluster-replicas 1 \ -a " password" ........ 。 I use this command to create cluster . And I use wan ip wholly

Comment From: yetian12

构建集群时，尝试cluster meet使用 WAN IP 而不是 LAN IP？

然后，我执行命令：-a "Iwhalecloud@2022redis"

我不确定命令是什么

redis-cli --cluster create --cluster-replicas 1 \ -a " password" ........ 。 I use this command to create cluster . And I use wan ip wholly