delete-nodegroup.sh

Previous Next

Purpose

Delete an EKS managed nodegroup with explicit confirmation and optional wait.

Location

cloud/aws/eks/delete-nodegroup.sh

Preconditions

  • Required tools: bash, aws
  • Required permissions: eks:DeleteNodegroup, eks:DescribeNodegroup
  • Required environment variables: none

Arguments

Flag Required Default Description
--cluster-name NAME Yes N/A Cluster name
--nodegroup-name NAME Yes N/A Nodegroup name
--if-missing No false Return success when missing
--wait / --no-wait No wait enabled Wait for deletion completion
--timeout SEC No 1800 Wait timeout
--poll-interval SEC No 20 Poll interval
--region REGION, --profile PROFILE No AWS defaults AWS context override
--yes Cond. false Required for non-dry-run deletion
--dry-run No false Print planned commands

Scenarios

  • Happy path: nodegroup deletion succeeds and resource is removed.
  • Common operational path: remove outdated worker pools during migration/upgrade.
  • Failure path: missing permissions or nodegroup busy/pending updates.
  • Recovery/rollback path: recreate nodegroup or retry once conflicting updates clear.

Usage

cloud/aws/eks/delete-nodegroup.sh --cluster-name prod-eks --nodegroup-name legacy-workers --yes
cloud/aws/eks/delete-nodegroup.sh --cluster-name dev-eks --nodegroup-name test-workers --if-missing --yes
cloud/aws/eks/delete-nodegroup.sh --cluster-name dev-eks --nodegroup-name test-workers --dry-run

Behavior

  • Main execution flow:
  • validates target and deletion confirmation
  • handles optional missing-resource skip
  • submits delete request
  • optionally waits until nodegroup disappears
  • Idempotency notes: convergent with --if-missing.
  • Side effects: worker capacity removal and workload eviction/rescheduling impact.

Output

  • Standard output format: timestamped logs on stderr.
  • Exit codes:
  • 0 success
  • 2 invalid arguments/safety guard failure
  • non-zero on AWS API failures/timeouts

Failure Modes

  • Common errors and likely causes:
  • missing --yes confirmation
  • nodegroup not found without --if-missing
  • delete denied or blocked by existing operations
  • Recovery and rollback steps:
  • re-run with proper flags
  • wait for pending updates to complete
  • inspect EKS events and retry

Security Notes

  • Secret handling: none.
  • Least-privilege requirements: deletion permission limited to approved principals.
  • Audit/logging expectations: nodegroup deletions should be tracked with operational approvals.

Testing

  • Unit tests:
  • deletion safety checks and wait logic
  • Integration tests:
  • delete nodegroup in disposable test cluster
  • Manual verification:
  • describe-nodegroup returns not found
scale-nodegroup.sh kubeconfig-update.sh