Why are both servers active when clustering Pacemaker Corosync?

Question

linux

Greetings.
I'm trying to set up HA between two servers that are located at different data centers of different companies.
Since it is not possible to use IP addresses from the same subnet with one common gateway, I raised IPSec between servers, an identical network between servers 10.11.12.0/27 passes through the tunnel.
Raise the Pacemaker Corosync cluster on the servers, but both servers consider themselves active, although synchronization and authorization via PCS are running. But pcs status corosync prints only localhost
-------------------------------NODE-1------ ------------------------------------------------
[[email protected] /] # pcs status nodes
Pacemaker Nodes:
Online: dyu.lb.server
Standby:
Maintenance:
Offline: huj.lb.server
Pacemaker Remote Nodes:
Online:
Standby:
Maintenance:
Offline:
[[email protected] /]# pcs status
Cluster name: lbcluster
Stack: corosync
Current DC: dyu.lb.server (version 1.1.20-5.el7_7.1- 3c4c782f70) - partition WITHOUT quorum
Last updated: Thu Oct 31 09:33:04 2019
Last change: Thu Oct 31 09:16:59 2019 by root via cibadmin on dyu.lb.server
2 nodes configured
2 resources configured
Online: [ dyu .lb.server ]
OFFLINE: [ huj.lb.server ]
Full list of resources:
virtual_ip (ocf::heartbeat:IPaddr2): Started dyu.lb.server
webserver (ocf::heartbeat:nginx): Started dyu.lb. server
daemon status:
corosync: active/enabled
pacemaker: active/enabled
pcsd: active/enabled
[[email protected] /]# pcs status corosync
Membership information
--------------------- -
Nodeid Votes Name
1 1 dyu.lb.server (local)
-------------------------------NODE-2-- ----------------------------------------
[[email protected] -2 ~]# pcs status nodes
Pacemaker Nodes:
Online: huj.lb.server
Standby:
Maintenance:
Offline: dyu.lb.server
Pacemaker Remote Nodes:
Online:
Standby:
Maintenance:
Offline:
[[email protected] ~]# pcs status
Cluster name: lbcluster
Stack: corosync
Current DC: huj.lb.server (version 1.1.20-5.el7_7.1-3c4c782f70) - partition WITHOUT quorum
Last updated: Thu Oct 31 09:33:13 2019
Last change: Thu Oct 31 09:15 :42 2019 by root via cibadmin on huj.lb.server
2 nodes configured
0 resources configured
Online: [ huj.lb.server ]
OFFLINE: [ dyu.lb.server ]
No resources
Daemon Status:
corosync: active/enabled
pacemaker: active /enabled
pcsd: active/enabled
[[email protected] ~]# pcs status corosync
Membership information
----------------------
Nodeid Votes Name
2 1 huj. lb.server (local)

Answer 1

The problem was that the servers were not on the same broadcas domain.
Solved by creating a GRE tunnel between servers.