Solved Failover instance with port 5001 suffix

[Justin Moors]

Hi all

I have a failover setup across two cloud providers, the active is on Google Cloud and the failover is on AWS, it fails over without problem, but I noticed that none of the handsets reconnected. On investigation, the primary is configured and responds to https://domain.3cx.uk but the failover always requires https://domain.3cx.uk:5001. I tried seeing if I could connect to the active with a :5001 suffix, however I get server not responding.

I figure that as the SBCs are configured to point to https://domain.3cx.uk when in a failover state they fail to connect because the configured URL doesnt contain the suffix. I have enabled Failover in the SBC settings and specified the IP address of the passive instance - but still no joy.

Assuming that the correct course of action is to ensure the port suffix is not required on the failover, how do I change this?

Thanks
Justin

 

[YiannisH_3CX]

Hi @Justin Moors

The FQDN needs to be identical between the primary and failover server. So you will need to reinstall the failover server using the same FQDN and same port. If there is no suffix then the port is 443.

 

[Justin Moors]

Thanks @YiannisH_3CX

So am I right in thinking that during the setup for the failover I must have inadvertently added the 5001 suffix to the URL?

 

[YiannisH_3CX]

Yes, but to be fair its the default port suggested during the setup. If you use the same key for the setup (which you should) the FQDN would be auto populated so I am thinking you used a different key during the setup.

So you need a enterprise key for failover and use the same key for both installs so you can have the same FQDN.

 

[Justin Moors]

I definitely used the same key and they are both associated with the same FQDN, as I say the failover between the two works perfectly, its just the SBCs dont reconnect to the passive node when in failover mode, this is when I discovered that I needed the :5001 suffix to reach the management UI and made the connection.

 

[YiannisH_3CX]

Ok then that is what you need to fix. Just make sure to use port 443 when reinstalling.

 

[Justin Moors]

That's great, thanks @YiannisH_3CX

 

[YiannisH_3CX]

Let us know how it goes

 

[Justin Moors]

Will do

 

[Justin Moors]

Rebuilt failover on AWS and spotted the default for management UI was set to 5001 when using the AMI - changed this to 443 and all is now working as anticipated.