How to Configure Split-Brain Avoidance For Neverfail Engine

How to Configure Split-Brain Avoidance For Neverfail Engine

Summary

This Knowledgebase article provides the procedure for configuring Split-Brain Avoidance when running Neverfail Continuity Engine.

More Information

Split-Brain Avoidance is a mechanism meant to ensure that only one server remains Active if Automatic Failover is enabled and the Channel network connection is lost, but both servers remain connected to the Production (public) network.


Split-brain Avoidance works by "pinging" from the Passive server to the Active server across the production (public) network. If the Active server responds, the Passive server will not failover, even if the channel connection is lost. This feature requires that the Active and Passive servers have an additional IP address besides the Production (public) IP. This additional IP address is usually called Management IP.


Procedure

To enable this feature, you will need to first configure a unique "Management IP address" on each of the servers. The Management IP address should be in the same subnet with the Production (public) IP address of each server.


In a LAN (HA) deployment the Production (public) IP is the same for Primary and Secondary servers. However, in a WAN (DR) deployment, the Production (public) IP is different, because the Secondary server is at a different location and so in a different subnet.


The Management IP can be configured on the same network card (NIC) with the Production (public) IP or on a different/separated NIC.


Management IP address allows the Passive server to communicate because the Production (public) IP address is always filtered/blocked by Neverfail Packet Filter on the Passive server. The Management IP allows the Passive server to send "pings", and is also required to allow the passive server to send email alerts.


To configure Management IP addressed on your Primary and Secondary servers please follow the instructions in Knowledgebase article #208.


To configure Split-Brain Avoidance in the Neverfail Engine Advanced Management Client, follow the steps below:
  1. Open the Neverfail Engine Advanced Management Client.
  2. Go to Server ->Monitoring .
  3. Open Configure Pings --> Ping Routing .
  4. For each server enter the Management and Production (Public) IPs such that the ping is configured from the Management IP to the Public IP of the other server.
  5. Click OK .
  6. Open Configure Failover
  7. Enable 'Prevent failover if channel heartbeat is lost but Active server is still visible to other servers'.
  8. Click OK .

 Example

In the example below, we have a HA (LAN) installation where the Production (Principal) IP is 192.168.10.221 and is the same for Primary and Secondary servers.

The Management IP of the Primary server is 192.168.10.224.

The Management IP of the Secondary server is 192.168.10.225


The configuration of the Management and Public IPs in the Advanced Management Client looks like this:

 

Applies To

All Versions Of Neverfail Continuity Engine

 

Related Information

None

 

KBID-1615


    • Related Articles

    • Continuity Engine Switchover/Failover Processes

      This article discusses Switchovers and Failovers, their similarities and differences. It also discusses a condition called False Failover, which can result in a Split Brain Syndrome. Learning objectives At the end of the session you should be able ...

    • How to Configure Application Services In Neverfail Continuity Engine

      Summary This Knowledgebase article provides information about how to configure application services in Neverfail Continuity Engine.  More Information  Services page The Engine Management Service Services page shows services specified by plug-ins or ...

    • How to Configure Rules in Neverfail Continuity Engine

      Summary This Knowledgebase article provides information about configuring rules in Neverfail Continuity Engine. More Information  Rules Overview Rules are implemented by plug-ins (there are no user-defined rules). Rules can be either timed (they must ...

    • Continuity Engine Troubleshooting - Channel Drops

      This article discusses unexpected channel drops. Under normal operations, Neverfail Continuity Engine maintains continuous communications between servers using the Neverfail Channel. When communications between servers fail, the condition is referred ...

    • How To Resolve Two Active Servers (Split-Brain)

      Summary This Knowledgebase article provides information about the symptoms, causes, and resolutions of two active servers. More Information The occurrence of two active servers is not by design and when detected, should be resolved immediately. When ...