Agentic Planner - February 7, 2026: KSail Development Roadmap

[botantler[bot]]

Executive Summary

KSail development maintains momentum with 14 open issues across feature development and infrastructure. The project shows excellent health with stable CI/CD and a clear roadmap focused on cluster lifecycle enhancements and LoadBalancer support.

Current State:

• ✅ Repository Health: Excellent - clean builds, passing tests, dependencies current
• 🚧 Active Development: Focus on LoadBalancer support ([feature]: Add --load-balancer flag to ksail cluster init|create #1714) and cluster update command ([feature]: add ksail cluster update for in-place updating a cluster #1734)
• ⚠️ Infrastructure: Minor issues ([agentics] Update Docs failed #2061 Update Docs, Daily Test Coverage Improver - Go Module Proxy Infrastructure Failure #2079 Go proxy) - non-blocking
• 📋 Recent Progress: CI test fixes (CI Failure DoctorSystem Test failure: resource quota evaluation timeout in Talos + Calico + Gatekeeper configuration #2086 closed), dependency updates ongoing

Strategic Direction: Complete production-ready cluster lifecycle features (LoadBalancer, Update, Backup/Restore) before expanding to new distributions (vCluster, Omni).

---

Current State Analysis

Activity Since February 6

Recent Fixes:

• ✅ CI Failure DoctorSystem Test failure: resource quota evaluation timeout in Talos + Calico + Gatekeeper configuration #2086: CI test failure (resource quota timeout) - closed, transient API server issue
• ✅ Dependency Updates: docker/login-action bump (3.6.0 → 3.7.0)
• ✅ CI/CD Stability: System tests stable after transient failures resolved

Outstanding Issues:

• ⚠️ Daily Test Coverage Improver - Go Module Proxy Infrastructure Failure #2079: Go module proxy infrastructure issue - external Google infrastructure problem, blocking test coverage improver workflow only
• ⚠️ [agentics] Update Docs failed #2061: Update Docs workflow failure - minor automation issue, non-blocking

Open Issues Inventory (14 Total)

🔴 Critical Priority (6 issues - LoadBalancer Support)

Parent Issue:

1. [feature]: Add --load-balancer flag to ksail cluster init|create #1714: LoadBalancer support (implied parent for [feature]: Add MetalLB installer for Talos × Docker LoadBalancer support #2069, [feature]: Add hcloud-ccm installer for Talos × Hetzner LoadBalancer support #2070, [feature]: Add Cloud Provider KIND installer for Vanilla × Docker LoadBalancer support #2071, [feature]: Document LoadBalancer support matrix and configuration examples #2074)

Implementation Sub-Tasks:
2. #2069: MetalLB installer for Talos × Docker ⭐ START HERE
3. #2070: hcloud-ccm installer for Talos × Hetzner
4. #2071: Cloud Provider KIND installer for Vanilla × Docker ⭐ START HERE
5. #2074: Documentation for LoadBalancer support (final deliverable)
6. #2073: CI workflow-specific concurrency groups (enables #1903 fix)

🔴 Critical Priority (2 issues - Cluster Update)

Parent Issue:
7. #1734: Cluster update command

Implementation Sub-Task:
8. #2072: Core cluster update implementation

🟡 High Priority (1 issue - CI Stability)

9. CI Failure DoctorRecurring merge_group concurrency race causing false CI failures #1903: CI matrix aggregation recurring failures - blocked by [feature]: Implement workflow-specific concurrency groups to fix matrix aggregation #2073

🟢 Medium Priority (3 issues - Future Features)

10. [feature]: Cluster backup and restore commands (Velero-inspired) #1836: Cluster backup/restore commands (Velero-inspired)
11. [feature]: Implement vCluster as a distribution, and use vind for the docker provider #2017: vCluster distribution support
12. [feature]: Add Omni provider for Talos #1970: Omni provider for Talos

ℹ️ Infrastructure (2 issues - Tracking/Monitoring)

13. [agentic-workflows] Failed runs #2060: Failed runs tracker (parent issue for workflow failures)
14. [agentics] Update Docs failed #2061: Update Docs workflow failure (minor, non-blocking)

⚠️ Blocked/External (1 issue)

15. Daily Test Coverage Improver - Go Module Proxy Infrastructure Failure #2079: Test coverage improver blocked by Google Go proxy infrastructure

---

Priority Roadmap

Phase 1: Infrastructure Stability (Week 1-2, Feb 7-20)

Goal: Resolve CI/CD stability issues before major feature work.

Week 1 (Feb 7-13)

• Fix CI concurrency ([feature]: Implement workflow-specific concurrency groups to fix matrix aggregation #2073) ⭐ HIGH PRIORITY
  • Change concurrency group to include workflow name: "ci-ksail-${{ github.workflow }}-${{ github.ref }}"
  • Apply pattern to all workflows (.github/workflows/*.yaml)
  • Test with merge queue runs
  • Monitor 5-10 runs to verify no cross-workflow cancellations
  • Estimated effort: 1-2 days
  • Success criteria: Zero false failures from concurrency races
  • Assignable: Yes

Week 2 (Feb 14-20)

• Debug Update Docs workflow ([agentics] Update Docs failed #2061) (Low priority, opportunistic)
  • Can use agentic-workflows agent or manual debugging
  • Not blocking other work
  • Estimated effort: 2-3 hours

Success Criteria:

• ✅ No false CI failures from concurrency races
• ✅ Merge queue runs complete successfully
• ✅ Workflow stability >95%
• ✅ Issue CI Failure DoctorRecurring merge_group concurrency race causing false CI failures #1903 resolved as side effect

---

Phase 2: LoadBalancer Support (Weeks 2-7, Feb 13 - Mar 26)

Goal: Enable LoadBalancer services across all distribution × provider combinations.

Week 2-3 (Feb 13-26) - Parallel Development

Track A: MetalLB for Talos × Docker (#2069) ⭐

• Create pkg/svc/installer/loadbalancer/metallb/ package
• Implement MetalLB installer following existing installer patterns (see CNI installers)
• Configure L2 mode by default for Talos × Docker
• Add unit tests for installation logic (mock client patterns)
• Add integration test: Talos cluster + LoadBalancer service
• Document MetalLB configuration options
• Estimated effort: 1.5 weeks
• Assignable: Yes

Track B: Cloud Provider KIND for Vanilla × Docker (#2071) ⭐

• Create pkg/svc/installer/loadbalancer/cloudproviderkind/ package
• Implement Cloud Provider KIND installer
• Add unit tests for installation logic
• Add integration test: Vanilla cluster + LoadBalancer service
• Document Cloud Provider KIND setup
• Estimated effort: 1.5 weeks
• Assignable: Yes (can run parallel with MetalLB)

Week 4-6 (Feb 27 - Mar 19) - Hetzner Support

• hcloud-ccm Installer ([feature]: Add hcloud-ccm installer for Talos × Hetzner LoadBalancer support #2070)
  • Create pkg/svc/installer/loadbalancer/hcloudccm/ package
  • Implement hcloud-cloud-controller-manager installer
  • Configure API token injection (HCLOUD_TOKEN env var)
  • Add unit tests for installation logic
  • Add integration test: Talos × Hetzner + LoadBalancer service
  • Document prerequisites and authentication setup
  • Estimated effort: 2-3 weeks
  • Assignable: Yes (requires Hetzner account for testing)

Week 7 (Mar 20-26) - Documentation Finalization

• LoadBalancer Documentation ([feature]: Document LoadBalancer support matrix and configuration examples #2074)
  • Support matrix table (Distribution × Provider × LoadBalancer)
  • Configuration examples for each combination
  • Troubleshooting guide (IP not assigned, service pending, etc.)
  • CLI flags reference (auto-generated)
  • README updates mentioning LoadBalancer support
  • Estimated effort: 4-5 days
  • Assignable: Yes

Success Criteria:

• ✅ LoadBalancer services work on Vanilla × Docker (Cloud Provider KIND)
• ✅ LoadBalancer services work on K3s × Docker (ServiceLB - already working)
• ✅ LoadBalancer services work on Talos × Docker (MetalLB)
• ✅ LoadBalancer services work on Talos × Hetzner (hcloud-ccm)
• ✅ Comprehensive documentation published
• ✅ Integration tests passing for all combinations
• ✅ Support matrix documented clearly

---

Phase 3: Cluster Update Command (Weeks 7-10, Mar 20 - Apr 16)

Goal: Enable in-place cluster configuration updates without recreation.

Week 7-8 (Mar 20 - Apr 2) - Core Implementation

• Core Update Implementation ([feature]: Implement ksail cluster update core command and config diff detection #2072)
  • Add ksail cluster update command to CLI
  • Implement config diff detection (compare running vs ksail.yaml)
  • Component addition/removal orchestration
  • --dry-run flag to preview changes
  • Confirmation prompt (reuse deletion confirmation pattern)
  • Estimated effort: 2 weeks
  • Assignable: Yes

Week 9-10 (Apr 3-16) - Error Handling & Testing

• Update Command: Immutability & Rollback
  • Handle immutable fields with clear error messages
  • Rollback on partial failure
  • Unit tests for diff detection and orchestration
  • Integration tests for update scenarios
  • Documentation with examples
  • Estimated effort: 2 weeks
  • Assignable: Yes

Success Criteria:

• ✅ ksail cluster update applies mutable config changes in-place
• ✅ --dry-run shows planned changes without applying
• ✅ Confirmation prompt prevents accidental updates
• ✅ Immutable changes trigger clear error messages
• ✅ Rollback works on failures
• ✅ Comprehensive test coverage
• ✅ Documentation includes examples for common scenarios

---

Phase 4: Backup/Restore & New Distributions (Q2 2026 - Apr-Jun)

Medium Priority Features (start after Phase 3):

Cluster Backup/Restore (#1836)

• Timeline: 6-8 weeks (mid-April to early June)
• Scope:
  • ksail cluster backup (resources + PV data)
  • ksail cluster restore with ordering
  • Metadata tracking (.tar.gz archives)
  • File system backup (v1)
• Limitations: No encryption, no cloud snapshots in v1
• Dependencies: Phase 3 complete (enables testing backup/restore workflows)

vCluster Distribution (#2017)

• Timeline: 4-6 weeks (can run parallel with backup/restore)
• Scope:
  • Add vCluster distribution
  • vind integration for Docker provider
  • Provisioner implementation
  • Multi-tenancy use cases
• Strategic Value: Unique differentiator vs competitors (lightweight virtual clusters)
• Dependencies: None (can run parallel)

Omni Provider for Talos (#1970)

• Timeline: 6-8 weeks (can run parallel)
• Scope:
  • Omni client SDK integration
  • Talos provisioner + Omni
  • API authentication
  • Production cluster management via Omni UI
• Prerequisites: Omni account required for testing
• Dependencies: None (can run parallel)

---

Dependencies & Critical Path

graph TD
    A[CI Concurrency `Fix #2073`] -->|Stabilizes CI| B[LoadBalancer Phase]
    B -->|MetalLB #2069| C[LoadBalancer Complete]
    B -->|Cloud Provider KIND #2071| C
    B -->|hcloud-ccm #2070| C
    C -->|Documentation #2074| D[Phase 2 Done]
    D -->|Start Phase 3| E[Update Command #2072]
    E -->|Enables workflows| F[Backup/Restore #1836]
    E -->|Independent| G[vCluster #2017]
    E -->|Independent| H[Omni Provider #1970]
    
    A -->|Fixes| I[#1903 CI Matrix Issue]
    
    style A fill:#ffd93d,stroke:#ffa94d
    style B fill:#ff6b6b,stroke:#c92a2a
    style C fill:#ff6b6b,stroke:#c92a2a
    style E fill:#ff6b6b,stroke:#c92a2a
    style F fill:#6bcf7f,stroke:#37b24d
    style G fill:#6bcf7f,stroke:#37b24d
    style H fill:#6bcf7f,stroke:#37b24d
    style I fill:#ffd93d,stroke:#ffa94d

Loading

Critical Path: CI Fix → LoadBalancer (MetalLB + Cloud Provider KIND in parallel) → hcloud-ccm → Documentation → Cluster Update → {Backup/Restore OR vCluster OR Omni}

---

Recommended Timeline (February - June 2026)

February 2026

• Week 1-2: CI concurrency fix ([feature]: Implement workflow-specific concurrency groups to fix matrix aggregation #2073), start LoadBalancer installers ([feature]: Add MetalLB installer for Talos × Docker LoadBalancer support #2069, [feature]: Add Cloud Provider KIND installer for Vanilla × Docker LoadBalancer support #2071 in parallel)
• Week 3-4: Complete MetalLB and Cloud Provider KIND installers

March 2026

• Week 1-3: Complete hcloud-ccm installer ([feature]: Add hcloud-ccm installer for Talos × Hetzner LoadBalancer support #2070)
• Week 3: LoadBalancer documentation ([feature]: Document LoadBalancer support matrix and configuration examples #2074)
• Week 4: Release LoadBalancer support, start cluster update command ([feature]: Implement ksail cluster update core command and config diff detection #2072)

April 2026

• Week 1-2: Complete cluster update command core implementation
• Week 3-4: Testing, error handling, rollback for update command

May-June 2026

• Backup/restore implementation ([feature]: Cluster backup and restore commands (Velero-inspired) #1836)
• vCluster support ([feature]: Implement vCluster as a distribution, and use vind for the docker provider #2017) - parallel option
• Omni provider ([feature]: Add Omni provider for Talos #1970) - parallel option

---

Risk Assessment

Risk	Likelihood	Impact	Mitigation
LoadBalancer scope creep	Medium	Medium	Strict MVP: basic L4 only, defer IPAM/advanced features
hcloud-ccm testing costs	Low	Low	Use Hetzner free tier, coordinate with community
Update command breaks clusters	Medium	High	Extensive dry-run mode, rollback, conservative immutability handling
CI matrix issue persists	Low	Medium	Quick fix (#2073), monitor closely, escalate to GitHub if needed
Go proxy infrastructure (#2079)	High	Low	External issue, wait for Google recovery, not blocking main work
Backup/restore data loss	Low	High	Extensive testing, clear v1 limitations documentation
vCluster integration complexity	High	Medium	Research/POC first, fallback to "future work" if too complex
Transient API server timeouts	Low	Low	Monitor #2086 pattern, add retry logic if recurs

---

Success Metrics

February Targets (Week 4)

• ✅ CI concurrency fix validated (no false failures)
• ✅ MetalLB installer 80% complete
• ✅ Cloud Provider KIND installer 80% complete

March Targets (End of Month)

• ✅ All LoadBalancer installers complete
• ✅ LoadBalancer documentation published
• ✅ Cluster update command 50% complete

Q1 2026 Targets (End of March)

• ✅ LoadBalancer support released (all installers working)
• ✅ Integration tests passing for all distribution × provider combos
• ✅ Cluster update command in active development

Q2 2026 Targets (End of June)

• ✅ Cluster update command released
• ✅ At least one of: Backup/Restore, vCluster, or Omni completed

Quality Metrics (Ongoing)

• Test Coverage: Maintain >70%
• Build Time: Keep <5 minutes
• CI Success Rate: Target >95%
• Documentation: Every feature has usage guide + examples

---

Alignment with Previous Plan (Feb 6)

What Changed:

1. Minor status update: Issue CI Failure DoctorSystem Test failure: resource quota evaluation timeout in Talos + Calico + Gatekeeper configuration #2086 closed (transient API server timeout, not actionable)
2. Dependency update: docker/login-action bumped
3. Timeline refinement: Week-by-week breakdown remains accurate
4. Priority confirmation: LoadBalancer and Update command remain critical priority

What Stayed the Same:

• LoadBalancer as critical priority ([feature]: Add --load-balancer flag to ksail cluster init|create #1714 parent, installers as sub-issues)
• Cluster Update as critical priority ([feature]: add ksail cluster update for in-place updating a cluster #1734 parent, [feature]: Implement ksail cluster update core command and config diff detection #2072 implementation)
• CI matrix fix as high priority (CI Failure DoctorRecurring merge_group concurrency race causing false CI failures #1903 parent, [feature]: Implement workflow-specific concurrency groups to fix matrix aggregation #2073 fix)
• vCluster ([feature]: Implement vCluster as a distribution, and use vind for the docker provider #2017) and Omni ([feature]: Add Omni provider for Talos #1970) as medium priority
• Strategic focus: Complete lifecycle features before expanding

Key Insight: Plan remains valid and on track. No major issues or blockers emerged in the past 24 hours.

---

Suggested New Issues to Create

None at this time. All necessary issues have been created from previous planning sessions:

• ✅ [feature]: Add MetalLB installer for Talos × Docker LoadBalancer support #2069: MetalLB installer
• ✅ [feature]: Add hcloud-ccm installer for Talos × Hetzner LoadBalancer support #2070: hcloud-ccm installer
• ✅ [feature]: Add Cloud Provider KIND installer for Vanilla × Docker LoadBalancer support #2071: Cloud Provider KIND installer
• ✅ [feature]: Implement ksail cluster update core command and config diff detection #2072: Cluster update core implementation
• ✅ [feature]: Implement workflow-specific concurrency groups to fix matrix aggregation #2073: CI workflow-specific concurrency
• ✅ [feature]: Document LoadBalancer support matrix and configuration examples #2074: LoadBalancer documentation

These issues provide sufficient granularity for tracking Phase 1-3 work. Additional issues can be created as needed when starting Phase 4 (Backup/Restore, vCluster, Omni).

Future Issue Suggestions (create when Phase 4 starts):

When ready to implement Backup/Restore (#1836), consider breaking into sub-issues:

# Backup Command Implementation
gh issue create \
  --title "[feature]: Implement ksail cluster backup core functionality" \
  --body "**Parent Issue**: #1836

## Scope
Implement core backup functionality for Kubernetes resources and persistent volume data.

## Tasks
- [ ] Create backup archive structure (.tar.gz)
- [ ] Implement resource export (YAML organized by namespace)
- [ ] Implement PV data backup (file system copy)
- [ ] Add backup metadata file (version, timestamp, cluster info)
- [ ] Unit tests for backup logic
- [ ] Integration tests for backup scenarios

## Acceptance Criteria
- \`ksail cluster backup\` creates valid backup archives
- Resources backed up in correct order
- PV data included when \`--include-volumes=true\`
- Metadata file includes all required fields

**Timeline**: 3-4 weeks" \
  --label "go" \
  --label "next"

# Restore Command Implementation
gh issue create \
  --title "[feature]: Implement ksail cluster restore core functionality" \
  --body "**Parent Issue**: #1836

## Scope
Implement core restore functionality to apply backup archives to target clusters.

## Tasks
- [ ] Parse backup archive structure
- [ ] Implement resource restore with ordering (CRDs → Namespaces → PVs → PVCs → Workloads)
- [ ] Implement PV data restore
- [ ] Add \`--existing-resource-policy\` flag (none, update)
- [ ] Add restore labels for traceability
- [ ] Unit tests for restore logic
- [ ] Integration tests for restore scenarios

## Acceptance Criteria
- \`ksail cluster restore\` applies backup archives successfully
- Resources restored in correct order
- PV data restored to new PVCs
- Existing resources handled per policy flag

**Timeline**: 3-4 weeks" \
  --label "go" \
  --label "next"

When ready to implement vCluster (#2017), consider breaking into sub-issues:

# vCluster Distribution Provisioner
gh issue create \
  --title "[feature]: Implement vCluster distribution provisioner" \
  --body "**Parent Issue**: #2017

## Scope
Add vCluster as a distribution option in KSail with vind for Docker provider.

## Tasks
- [ ] Research vCluster SDK/API
- [ ] Create vCluster provisioner in \`pkg/svc/provisioner/cluster/vcluster/\`
- [ ] Integrate vind for Docker provider
- [ ] Add distribution enum value in \`pkg/apis/cluster/v1alpha1/enums.go\`
- [ ] Update CLI to support vCluster distribution
- [ ] Unit tests for provisioner logic
- [ ] Integration tests for vCluster creation

## Acceptance Criteria
- \`ksail cluster create --distribution vCluster --provider Docker\` works
- Virtual clusters run inside host cluster
- Multi-tenancy use cases supported

**Timeline**: 4-6 weeks" \
  --label "go" \
  --label "next"

---

Notes for Maintainers

Immediate Actions (Week 1)

1. Implement CI concurrency fix ([feature]: Implement workflow-specific concurrency groups to fix matrix aggregation #2073) ⭐ START HERE

   • Update .github/workflows/ci.yaml concurrency configuration
   • Apply pattern to other workflows
   • Test with merge queue runs
   • Close #1903 when validated

2. Start LoadBalancer installers in parallel:

   • Assign [feature]: Add MetalLB installer for Talos × Docker LoadBalancer support #2069 (MetalLB) to developer 1
   • Assign [feature]: Add Cloud Provider KIND installer for Vanilla × Docker LoadBalancer support #2071 (Cloud Provider KIND) to developer 2
   • Both can proceed independently

3. Monitor infrastructure issues:

   • Track Daily Test Coverage Improver - Go Module Proxy Infrastructure Failure #2079 (Go proxy) for Google infrastructure recovery
   • Optionally debug [agentics] Update Docs failed #2061 (Update Docs) as time permits

Parallelization Opportunities

• MetalLB ([feature]: Add MetalLB installer for Talos × Docker LoadBalancer support #2069) + Cloud Provider KIND ([feature]: Add Cloud Provider KIND installer for Vanilla × Docker LoadBalancer support #2071) can be developed in parallel (Week 2-3)
• vCluster ([feature]: Implement vCluster as a distribution, and use vind for the docker provider #2017) + Omni ([feature]: Add Omni provider for Talos #1970) can run independently in Phase 4
• Documentation ([feature]: Document LoadBalancer support matrix and configuration examples #2074) can start when first installer is complete

External Dependencies

• Hetzner account required for [feature]: Add hcloud-ccm installer for Talos × Hetzner LoadBalancer support #2070 testing
• Omni account required for [feature]: Add Omni provider for Talos #1970 (future)
• Google Go proxy recovery for Daily Test Coverage Improver - Go Module Proxy Infrastructure Failure #2079 (external, waiting)

Testing Strategy

• LoadBalancer: Each installer gets integration test before documentation phase
• Update command: Extensive dry-run mode testing to prevent cluster breakage
• Backup/Restore: Test with real workload data, document v1 limitations clearly

---

Generated by: Agentic Planner
Date: February 7, 2026
Repository: devantler-tech/ksail
Previous Plan: Discussion #2080 (Feb 6, 2026)

---

AI generated by Agentic Planner

AI generated by Agentic Planner

• expires on Feb 14, 2026, 10:09 AM UTC