Failover detection on AWS Lightsail failing

[Mike O]

We have two enterprise servers running on two AWS Lightsail instances (one in GB, the other in DE).

The issue is this: when pulling the plug on the GB instance, the DE instance does not detect the downed server. It seems the connection request is blackholed by the AWS firewall, i.e. the connection never times out, closes, or gets rejected.

1. The failover configuration is set to GB being active and DE being passive.
2. The standard 3CX/AWS firewall has been configured on both systems.
3. When telnet'ing to the downed server the connection just hangs.

I think that 3CX needs to set a hard connection timeout on the passive server checks to determine if the active server is offline. Has anyone else encountered this?

 

[cobaltit]

I mean if you are paying for ENT and need failover maybe pay a few more bucks and move up to real cloud and not AWS light? Seems simple enough if you think the firewall is to blame.