Issue
During the deployment of a new Liferay Pod, a standard resource optimization action reallocated pods from one node to another, causing a deployment timeout. This behavior occurs as Kubernetes manages auto-scaling, as detailed here: Kubernetes Cluster Autoscaler.
Environment
- PaaS
Resolution
To prevent service interruptions during such resource optimizations, enabling High Availability (HA) is strongly recommended. A HA setup involves deploying redundant components and services, ensuring continuous availability and minimizing the impact of potential disruptions or failures.
If the environment does not implement High Availability (HA), your environment may experience timeouts when pods are reallocated. The only definitive solution to prevent such service interruptions is to enable HA.
For more information about high-availability configurations, visit the resource: Upgrading to a High Availability Subscription
