eviden-logo

Evidian > Products > SafeKit: All-in-One SANless High Availability & Application Clustering Software > Network isolation case and power outage case in a cluster

Network isolation case and power outage case in a cluster

Evidian SafeKit

What are the different scenarios in case of network isolation in a cluster?

A single network

When there is a network isolation, the default behavior is:

  • as heartbeats are lost for each node, each node goes to ALONE and runs the application with its virtual IP address (double execution of the application modifying its local data),
  • when the isolation is repaired, one ALONE node is forced to stop and to resynchronize its data from the other node,
  • at the end the cluster is PRIM-SECOND (or SECOND-PRIM according the duplicate virtual IP address detection made by Windows).

Two networks with a dedicated replication network

When there is a network isolation, the behavior with a dedicated replication network is:

  • a dedicated replication network is implemented on a private network,
  • heartbeats on the production network are lost (isolated network),
  • heartbeats on the replication network are working (not isolated network),
  • the cluster stays in PRIM/SECOND state.

A single network and a splitbrain checker

When there is a network isolation, the behavior with a split-brain checker is:

  • a split-brain checker has been configured with the IP address of a witness (typically a router),
  • the split-brain checker operates when a server goes from PRIM to ALONE or from SECOND to ALONE,
  • in case of network isolation, before going to ALONE, both nodes test the IP address,
  • the node which can access the IP address goes to ALONE, the other one goes to WAIT,
  • when the isolation is repaired, the WAIT node resynchronizes its data and becomes SECOND.

Note: If the witness is down or disconnected, both nodes go to WAIT and the application is no more running. That's why you must choose a robust witness like a router.

What are the different scenarios in case of power outage in a cluster?

Primary node power outage

When a power outage stops only the primary node:

  • there is an automatic failover on the secondary node, which becomes ALONE and restarts the application,
  • when node 1 is rebooted, it becomes SEDOND after resynchronization of replicated data,
  • the roles of primary and secondary can be swapped by an adminsitrator if needed.

Secondary node power outage

When a power outage stops only the secondary node:

  • there is no failover, the primary becomes ALONE and the application continues its execution on node 1,
  • when node 2 is rebooted, it becomes SEDOND after resynchronization of replicated data.

General power outage - case 1

When a power outage stops both nodes, the default behavior is:

  • both nodes goes to STOP,
  • when node 1 is rebooted, it does not go into ALONE state and does not restart the application because it doesn't know if it has the up-to-date data. So it goes to the WAIT state waiting for the restart of the other node,
  • when node 2 is rebooted, both nodes return to their previous PRIM/SECOND states.

General power outage - case 2

When a power outage stops both nodes, the behavior with syncdelta is :

  • syncdelta is set for example to 10 minutes in the configuration meaning that start of a node is accepted even if its data is 10 minutes behind the last sync,
  • when node 1 is rebooted, it goes to ALONE and restarts the application assuming that the restart is done within 10 minutes after the power failure,
  • when node 2 is rebooted, it becomes SECOND after resynchronization of replicated data,
  • Note: if node 2 is rebooted the first, then it becomes ALONE and node 1 will become SECOND at its start.

🔍 SafeKit High Availability Navigation Hub

Explore SafeKit: Features, technical videos, documentation, and free trial
Resource Type Description Direct Link
Key Features Why Choose SafeKit for Simple and Cost-Effective High Availability? See Why Choose SafeKit for High Availability
Deployment Model All-in-One SANless HA: Shared-Nothing Software Clustering See SafeKit All-in-One SANless HA
Partners SafeKit: The Benchmark in High Availability for Partners See Why SafeKit Is the HA Benchmark for Partners
HA Strategies SafeKit: Infrastructure (VM) vs. Application-Level High Availability See SafeKit HA & Redundancy: VM vs. Application Level
Technical Specifications Technical Limitations for SafeKit Clustering See SafeKit High Availability Limitations
Proof of Concept SafeKit: High Availability Configuration & Failover Demos See SafeKit Failover Tutorials
Architecture How the SafeKit Mirror Cluster works (Real-Time Replication & Failover) See SafeKit Mirror Cluster: Real-Time Replication & Failover
Architecture How the SafeKit Farm Cluster works (Network Load Balancing & Failover) See SafeKit Farm Cluster: Network Load Balancing & Failover
Competitive Advantages Comparison: SafeKit vs. Traditional High Availability (HA) Clusters See SafeKit vs. Traditional HA Cluster Comparison
Technical Resources SafeKit High Availability: Documentation, Downloads & Trial See SafeKit HA Free Trial & Technical Documentation
Pre-configured Solutions SafeKit Application Module Library: Ready-to-Use HA Solutions See SafeKit High Availability Application Modules