> ## Documentation Index
> Fetch the complete documentation index at: https://docs.budecosystem.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Cluster Operations

> Run day-2 operations for health, deployments, and lifecycle management

## Overview

After onboarding, operators use Clusters for continuous health management, deployment oversight, and safe lifecycle changes.

## Daily Operations Loop

```mermaid theme={null}
flowchart TD
    A[Review Cluster Health] --> B[Inspect Deployments]
    B --> C[Check Node Events]
    C --> D[Tune Capacity or Settings]
    D --> E[Document and Audit Changes]
    E --> A
```

## 1) Health Monitoring

* Review **General** tab for CPU/GPU/memory/storage trends.
* Check for abrupt drops in available workers or nodes.
* Use time windows to detect regressions.

## 2) Deployment Oversight

* Open **Deployments** tab to review active workloads.
* Confirm deployment status, worker counts, and routing behavior.
* Escalate unhealthy deployments before cluster-wide degradation.

<img src="https://mintcdn.com/budecosystem-b7b14df4/VWVW0RGNFnJu1JHC/images/image-35.png?fit=max&auto=format&n=VWVW0RGNFnJu1JHC&q=85&s=c9f0ea5e4a2895eb6302477825ce2dea" alt="Image" width="1919" height="874" data-path="images/image-35.png" />

## 3) Node Diagnostics

* Use **Nodes** tab for request-vs-allocatable analysis.
* Open node event panels for warnings (scheduling failures, connectivity, taints).
* Prioritize repeated or high-severity events for immediate remediation.

## 4) Safe Editing and Deletion

* Use edit actions to update metadata like name or ingress.
* Before deletion, confirm no active endpoints depend on the cluster.
* Record change rationale for governance and post-incident reviews.

## Incident Response Pattern

```mermaid theme={null}
flowchart LR
    A[Alert Triggered] --> B[Open Cluster General]
    B --> C[Correlate with Deployments]
    C --> D[Inspect Node Events]
    D --> E[Mitigate: Reschedule, Scale, or Roll Back]
    E --> F[Verify Recovery]
```

## Best Practices

<Check>
  Separate production and non-production clusters with clear naming.
</Check>

<Check>
  Review node events before and after major deployment rollouts.
</Check>

<Check>
  Avoid destructive operations during unresolved incidents.
</Check>

<Check>
  Align cluster actions with RBAC and audit policy requirements.
</Check>