Request for Guidance on 3CX High Availability and Failover Design

[Danushka]

Dear Team,
We require guidance regarding the failover mechanism for our 3CX deployment.
The customer has specifically requested a high-availability solution with minimal downtime in the event of a failure.
After reviewing the 3CX documentation, we understand that 3CX uses an active-passive approach with built-in configuration replication, which may have a synchronization interval of up to 12–24 hours. Our customer is concerned about this limitation, as the environment is business-critical and requires a robust high-availability solution.
We are currently evaluating the following approaches:
Option 1 – Handle 3CX Failover via Hyper-V

• Real-time/continuous VM replication (approximately a 3-second replication interval) from the primary server to the secondary server.
• Failover would be handled by Hyper-V VM replication rather than by 3CX itself.

Option 2 – Use the Native 3CX Failover Mechanism

• Utilize the failover functionality provided by 3CX.

We would appreciate clarification on the following concerns:
1. Regarding Option 1 (Hyper-V Failover)
If the primary server becomes unavailable and the replicated VM is started on the secondary site using the same IP address as the primary 3CX server, will 3CX continue operating without issues?
Is this deployment scenario supported or recommended by 3CX? We are asking because we have not previously implemented or tested this failover scenario.
2. Regarding Option 2 (3CX Native Failover)
If 3CX handles the failover process, what is the typical failover time under normal operating conditions?
Our current environment is as follows:
Primary Site

• Hyper-V Host IP: 172.25.35.130
• 3CX VM IP: 172.25.35.100

Secondary Site

• Hyper-V Host IP: 172.25.35.132
• 3CX VM IP: 172.25.35.104

Additional Information

• We are using a custom domain.
• Approximately 1,500 IP phones are registered to the 3CX system.
• Both servers are running Windows Server 2022 with Hyper-V.
• 3CX is deployed as virtual machines.
• Veeam Backup & Replication is also available for backup and recovery purposes.

We look forward to your guidance and recommendations regarding the most suitable high-availability approach for this deployment.
Thank you for your assistance.

 

[Alphabetic]

1. Yes - continue as normal
2. You configure the parameters for when a failover is considered. Though, what you need to be aware of is that the DNS will need updating for your FQDN so there may be some disruption depending on if you use split DNS for internal devices and if the WAN IP is going to change.

 

[KyriacosS_3CX]

Hello,

For the 3CX Native failover, it should be mentioned that you can have backups scheduled down to hourly. Failover should take only a few minutes (would depend on backup size/location + DNS switchover), 2 minutes is achievable, but the specifics of each case vary.