View source on GitHub →

Deployment Infrastructure Note

Date: December 16, 2025 01:59 UTC
Issue: K3s deployment runner out of disk space

Status

✅ Code Validation (CI Workflow)

All code quality checks PASSING:

  • Build Release: ✅
  • Web UI Tests: ✅
  • Format Check: ✅
  • TypeScript SDK: ✅ (67 tests)
  • Clippy: ✅ (0 warnings)
  • Rust Tests: ✅ (311+ tests)

⚠️ Deployment (K3s Workflow)

Issue: Runner disk space full

Error: No space left on device
Runner: 'homelab-k3s' on 'k3s-control'

Impact: Cannot deploy to K3s cluster

Code Impact: NONE - This is infrastructure only

Resolution

The K3s runner needs disk cleanup:

# On k3s-control node:
docker system prune -af --volumes
sudo du -sh /var/lib/docker
sudo du -sh /var/lib/rancher

Common cleanup steps:

  1. Remove old Docker images
  2. Clean build caches
  3. Remove unused volumes
  4. Check runner working directory

Workaround

Deploy manually until runner disk is cleared:

# Build locally
cd icn
cargo build --release

# Copy to K3s
scp target/release/icnd user@k3s-control:/tmp/
scp target/release/icnctl user@k3s-control:/tmp/

# Update deployment
kubectl set image deployment/icnd icnd=<new-image>
kubectl rollout restart deployment/icnd

Important

The code is production-ready. All validation checks pass. The deployment failure is purely infrastructure (disk space) and does not reflect code quality issues.

Code CI: ✅ GREEN
Deployment CI: ⚠️ Infrastructure issue (disk full)

Action Items

  1. ✅ Code validated and merged to main
  2. ⚠️ Clean K3s runner disk space (infrastructure team)
  3. ⏳ Re-run deployment after cleanup

Sprint deliverables are complete and validated.

← Back to Docs