What problem are you trying to solve?

Focus on the what and why of the need you have, not the how you'd like it solved.

Current Behavior

When spec.members is set to 0 on a MongoDBCommunity resource, the operator rejects the configuration and leaves the previous StatefulSet replicas running.

Observed behavior:

• the CR accepts the desired spec update

• reconciliation fails

• old MongoDB pods continue running

• PVCs remain, but compute is not reduced

Example status message:

error validating new Spec: number of arbiters specified (0) is greater or equal than the number of members in the replicaset (0). At least one member must not be an arbiter

What would you like to see happen?

Describe the desired outcome or enhancement.

Expected Behavior

One of these approaches would solve the problem:

1. Allow spec.members: 0 as a supported paused state

2. Introduce a dedicated field such as spec.paused: true or spec.hibernated: true

3. Add an operator-supported suspend mode that:

   • scales managed pods to zero

   • preserves PVCs

   • preserves enough metadata to resume safely

   • reports a healthy paused/suspended status instead of Failed

Why is this important to you or your team?

Explain how the request adds value or solves a business need.

Motivation

In some environments, we need a cost-saving or maintenance mode where application workloads are stopped, but stateful data is preserved. For most services, this means:

• stop running pods

• keep PVCs and other durable resources

• allow a later resume without data loss

This is especially useful for:

• non-production environments

• temporary shutdown windows

• cost optimization

• platform-wide pause/resume workflows

What steps, if any, are you taking today to manage this problem?