managing-digitalocean-kubernetes

name	managing-digitalocean-kubernetes
description	Use when working with Digitalocean Kubernetes — digitalOcean Kubernetes (DOKS) management via the doctl CLI. Covers clusters, node pools, upgrades, kubeconfig, and cluster health. Use when managing DOKS clusters or checking Kubernetes infrastructure on DigitalOcean.
connection_type	digitalocean-kubernetes
preload	false

Managing DigitalOcean Kubernetes (DOKS)

Manage DigitalOcean Kubernetes using the doctl kubernetes CLI.

MANDATORY: Discovery-First Pattern

Always discover available resources before performing analysis.

Phase 1: Discovery

#!/bin/bash

echo "=== Kubernetes Clusters ==="
doctl kubernetes cluster list --format ID,Name,Region,VersionSlug,Status,NodePools.Count,Endpoint,CreatedAt --no-header 2>/dev/null | head -20

echo ""
echo "=== Available Versions ==="
doctl kubernetes options versions --format Slug,KubernetesVersion --no-header 2>/dev/null | head -10

echo ""
echo "=== Available Sizes ==="
doctl kubernetes options sizes --format Name,Slug --no-header 2>/dev/null | head -10

echo ""
echo "=== Available Regions ==="
doctl kubernetes options regions --format Name,Slug --no-header 2>/dev/null | head -10

Phase 2: Analysis

#!/bin/bash

CLUSTER_ID="${1:?Cluster ID required}"

echo "=== Cluster Details ==="
doctl kubernetes cluster get "$CLUSTER_ID" --format ID,Name,Region,VersionSlug,Status,Endpoint,AutoUpgrade,SurgeUpgrade,HA,CreatedAt --no-header 2>/dev/null

echo ""
echo "=== Node Pools ==="
doctl kubernetes cluster node-pool list "$CLUSTER_ID" --format ID,Name,Size,Count,AutoScale,MinNodes,MaxNodes --no-header 2>/dev/null | head -10

echo ""
echo "=== Nodes ==="
doctl kubernetes cluster node-pool list "$CLUSTER_ID" -o json 2>/dev/null | jq -r '.[].nodes[] | "\(.id)\t\(.name)\t\(.status.state)\t\(.droplet_id)\t\(.created_at)"' | head -20

echo ""
echo "=== Available Upgrades ==="
doctl kubernetes cluster get-upgrades "$CLUSTER_ID" --format Slug,KubernetesVersion --no-header 2>/dev/null | head -5

echo ""
echo "=== Cluster Kubeconfig Check ==="
doctl kubernetes cluster kubeconfig show "$CLUSTER_ID" 2>/dev/null | grep -c "clusters:" && echo "Kubeconfig: available" || echo "Kubeconfig: not accessible"

echo ""
echo "=== Associated LBs ==="
doctl compute load-balancer list --format ID,Name,Region,Status,IP --no-header 2>/dev/null |  -5

 
 
doctl compute volume list --format ID,Name,Region,Size,DropletIDs --no-header 2>/dev/null |  -10

 
 
doctl registry get --format Name,Endpoint,Region,StorageUsageBytes,CreatedAt --no-header 2>/dev/null |  -3

Output Format

CLUSTER_ID                            NAME       REGION  VERSION  STATUS   NODES
abc123-def456-ghi789                  prod-k8s   nyc1    1.29.1   running  6
def456-ghi789-jkl012                  dev-k8s    sfo3    1.29.1   running  3

Safety Rules

• Use read-only commands: list, get, get-upgrades, show
• Never run delete, update, remove-node without explicit user confirmation
• Use --format and --no-header for clean output
• Limit output with | head -N to stay under 50 lines

Anti-Hallucination Rules

1. NEVER assume resource names — always discover via CLI/API in Phase 1 before referencing in Phase 2.
2. NEVER fabricate metric names or dimensions — verify against the service documentation or --help output.
3. NEVER mix CLI commands between service versions — confirm which version/API you are targeting.
4. ALWAYS use the discovery → verify → analyze chain — every resource referenced must have been discovered first.
5. ALWAYS handle empty results gracefully — an empty response is valid data, not an error to retry.

Counter-Rationalizations

Shortcut	Counter	Why
"I'll skip discovery and check known resources"	Always run Phase 1 discovery first	Resource names change, new resources appear — assumed names cause errors
"The user only asked for a quick check"	Follow the full discovery → analysis flow	Quick checks miss critical issues; structured analysis catches silent failures
"Default configuration is probably fine"	Audit configuration explicitly	Defaults often leave logging, security, and optimization features disabled
"Metrics aren't needed for this"	Always check relevant metrics when available	API/CLI responses show current state; metrics reveal trends and intermittent issues
"I don't have access to that"	Try the command and report the actual error	Assumed permission failures prevent useful investigation; actual errors are informative