-retry-join-wan sometimes not retry

[hehailong5]

Hi,
version: v1.6.10

I have two clusters join with -retry-join-wan. each cluster has only one running server.
during the reboot test, I found sometimes the two can not be joined with each other.
I have used the -retry-join-wan option in the command line, and suppose the two could find each other in the end.

the command line for the server #1:

consul agent -server -bootstrap -data-dir consul-works/data-dir -config-dir consul-works/config-dir  -node=server-a50_50__3  -disable-host-node-id -raft-protocol=2 -datacenter=dca50_50__3 -bind=a50:50::3 -client=a50:50::3 -retry-join-wan=[a50:50::2] -retry-join-wan=[a50:50::3]

the command line for the server #2

consul agent -server -bootstrap -data-dir consul-works/data-dir -config-dir consul-works/config-dir  -node=server-a50_50__2  -disable-host-node-id -raft-protocol=2 -datacenter=dca50_50__2 -bind=a50:50::2 -client=a50:50::2 -retry-join-wan=[a50:50::2] -retry-join-wan=[a50:50::3]

the testing output from consul.log

[root@01:/log]$ grep "joined:" consul.log
    2021/06/15 14:20:37 [INFO] agent: (WAN) joined: 2
    2021/06/15 20:36:19 [INFO] agent: (WAN) joined: 1
    2021/06/15 20:39:26 [INFO] agent: (WAN) joined: 1
    2021/06/16 07:44:10 [INFO] agent: (WAN) joined: 2
    2021/06/16 10:53:34 [INFO] agent: (WAN) joined: 2
    2021/06/17 14:14:42 [INFO] agent: (WAN) joined: 2
    2021/06/17 15:39:49 [INFO] agent: (WAN) joined: 1
    2021/06/17 16:39:59 [INFO] agent: (WAN) joined: 2
    2021/06/17 17:33:33 [INFO] agent: (WAN) joined: 1
    2021/06/17 18:33:27 [INFO] agent: (WAN) joined: 1