Re-check Primary-Failover / Disabling Fail-Over-Management

[clauspruefer]

"It manages fail-over: if a Primary PostgreSql crashes, it automatically promotes a Replica PostgreSql as a Primary."

1. How is this feature tested? Is it rock solid?
2. Is it possible to disable this feature?

Due to the fact that kubernetes network is virtualized, it should work 100%, in bare-metal scenarios switch / hardware outages can lead to split-brain scenario. This should be re-verified on kubernetes.

[alex-arica]

Yes it is possible to disable it. More details in the doc here: https://www.kubegres.io/doc/properties-explained.html#spec_failover

It was rock tested however, as you pointed out Kubernetes network can be unstable on some cloud provider leading to Kubernetes reporting that an instance of Postgres is not healthy. When this happens, Kubegres might trigger a failover. But the failover itself was rock tested.