TOPIC:

This is my situation:
Implemented a HA of two HOST, consisting of a Primary Server and Secondary Server, the primary server has the load of all VMs, the proposed scenarios were:
1- Turn off the Primary Server and the secondary will take full control of the VMs so as not to cut the service.
2- Disconnect the Primary Server from the main network, so that the secondary server will take control of the VMs again.

The result obtained was:
Scenario 1: When the primary server is turned off as it is he who has the master role of the HA, he indiscriminately restarts the secondary server after 60 seconds trying to find the connectivity of himself.

Scenario 2: When disconnecting the primary server from the network, it starts the countdown of 60 seconds to reconnect with the primary server and when it can not find it, it restarts the pool and the secondary server.

Why does the slave or secondary server restart when it is intentionally switched off or removed from the network to the master server?

The version of HA-Lizard used is 2.1.3 - NOSAN for Xenserver 7.2

It sounds like the heuristic IP setting is not working. In a loss of the master host, the slave would need to reach the IP address you set as a heuristic IP during the setup. If the slave cannot reach this IP, it will self-fence and enter a suspended HA mode until the network connectivity is restored.

This is just a guess based on what you are seeing. The logs will tell you exactly what happened.

thanks for answering

to where that heuristic ip address has to be connected, according to the diagram it must refer to a physical team.
How can I solve it since it does not respond to me from any physical equipment and is not connected to anything?

thanks for answering

you're rigth the heuristic ip address is not response.

to where that heuristic ip address has to be connected, according to the diagram it must refer to a physical team.
How can I solve it since it does not respond to me from any physical equipment and is not connected to anything?

The heuristic IP should be configured to any reliable interface on your network. The below is taken from our documentation

FENCE_HEURISTICS_IPS
Create a list of IP addresses or domain names to check. The list should be
delimited by “:”
As of version 1.7.7 the default value is set to 8.8.8.8 which is a well-known
address. However, live production systems should NOT use the default address.
This address should ideally be set to a local IP on the management network
which is very close to the management interface of the hosts. Typically this
would be the IP address of the switch the hosts are connected to. If more than
one IP address is specified then all of the addresses must be successfully
reached with ping in order to count as a single vote.