proxmox/docs/10-best-practices/PROXMOX_FINAL_RECOMMENDATIONS.md

Proxmox VE Final Recommendations and Summary

Last Updated: 2026-05-09
Document Version: 1.0
Status: Active Documentation

---

Live cluster (2026-05-09): PROXMOX_CLUSTER_ARCHITECTURE.md — PVE 9.1.7, 136 running guests, ml110 0; r630-01 57, r630-02 41, r630-03/04 19 each. The sections below retain historical migration and audit notes; do not use them as the current placement source of truth.

Date: 2025-01-20
Status: Complete Review with Actionable Recommendations

---

✅ Completed Tasks Summary

1. Hostname Migration - COMPLETE ✅

• r630-01 (192.168.11.11): Successfully renamed from pve to r630-01
• r630-02 (192.168.11.12): Successfully renamed from pve2 to r630-02
• All services operational after migration
• /etc/hosts updated on both hosts

2. IP Address Audit - COMPLETE ✅

• Cluster (2026-05-09): 136 running LXC/QEMU; static IP inventory: ALL_VMIDS_ENDPOINTS.md
• IP Conflicts: 0 ✅ (re-verify when adding CTs)

3. Proxmox Configuration Review - COMPLETE ✅

• All hosts reviewed
• Storage configurations analyzed
• Issues identified and documented

---

🔴 Critical Issues and Fixes

Issue 1: Storage Node References Outdated

Problem: Storage configuration files reference old hostnames (pve, pve2) instead of new hostnames (r630-01, r630-02)

Impact: Storage may show as disabled or inaccessible

Fix Applied:

# On r630-01
sed -i 's/nodes pve$/nodes r630-01/' /etc/pve/storage.cfg
sed -i 's/nodes pve /nodes r630-01 /' /etc/pve/storage.cfg

# On r630-02
sed -i 's/nodes pve2$/nodes r630-02/' /etc/pve/storage.cfg
sed -i 's/nodes pve2 /nodes r630-02 /' /etc/pve/storage.cfg

Status: ✅ Fixed

---

📊 Host Configuration Summary

ml110 (192.168.11.10)

• Status: ✅ Operational (cluster member)
• CPU: 6 cores
• Memory: ~63 GiB reported; ~3.5% used
• Guests: 0 (2026-05-09)
• Recommendation: Use r630-* for new workloads unless intentionally placing on ml110.

r630-01 (192.168.11.11) - Previously "pve"

• Status: ✅ Operational
• CPU: 32 cores @ 2.40GHz
• Memory: ~125 GiB reported; ~66% used
• Guests: 57
• Recommendation: Monitor thin1/data utilization before large disks.

r630-02 (192.168.11.12) - Previously "pve2"

• Status: ✅ Operational
• CPU: 56 cores @ 2.00GHz
• Memory: ~125 GiB reported; ~62% used
• Guests: 41 (includes infra CTs such as NPMplus — see inventory docs)
• Recommendation: Watch thin1-r630-02 … thin6 pool %.

---

🎯 Critical Recommendations

1. Enable Storage on r630-01 and r630-02 🔴 CRITICAL

Priority: HIGH - Required before starting new VMs

Actions:

1. ✅ Update storage.cfg node references (DONE)
2. ⏳ Enable local-lvm storage on r630-01
3. ⏳ Enable thin1-thin6 storage on r630-02
4. ⏳ Verify storage is accessible

Commands:

# On r630-01
ssh root@192.168.11.11
pvesm set local-lvm --disable 0
pvesm set thin1 --disable 0
pvesm status

# On r630-02
ssh root@192.168.11.12
for storage in thin1 thin2 thin3 thin4 thin5 thin6; do
    pvesm set "$storage" --disable 0
done
pvesm status

2. Verify Existing VMs on r630-02 ⚠️ HIGH PRIORITY

Issue: VMs found on r630-02 storage (VMIDs: 100, 101, 102, 103, 104, 105, 130, 5000, 6200, 7800)

Actions:

1. List all VMs/containers on r630-02
2. Verify they're accessible
3. Check their IP addresses
4. Update IP audit if needed

Commands:

ssh root@192.168.11.12
pct list
qm list
# Check each VMID's IP configuration

3. Distribute VMs Across Hosts ⚠️ RECOMMENDED

Current: All 34 VMs on ml110 (overloaded)

Recommendation:

• Migrate some VMs to r630-01 and r630-02
• Balance workload:
  • ml110: Keep management/lightweight VMs
  • r630-01: Medium workload VMs
  • r630-02: Heavy workload VMs (best CPU)

Benefits:

• Better performance (ml110 CPU is slower)
• Better resource utilization
• Improved redundancy

4. Update Cluster Configuration ⚠️ RECOMMENDED

Issue: Cluster may still reference old hostnames

Actions:

1. Verify cluster status
2. Check if hostname changes are reflected
3. Update cluster configuration if needed

Commands:

# On any cluster node
pvecm status
pvecm nodes
# Verify hostnames are correct

5. Storage Performance Optimization ⚠️ RECOMMENDED

Current State:

• ml110: Using local-lvm (good performance)
• r630-01: Only local (directory) - slower
• r630-02: thin1-thin6 available but need activation

Recommendation:

• Enable LVM thin storage on both r630-01 and r630-02
• Use thin provisioning for space efficiency
• Monitor storage usage

---

📋 Detailed Recommendations by Category

Storage Recommendations

Immediate Actions (Before Starting VMs)

1. Enable local-lvm on r630-01

   • Thin pools already exist (pve/data, pve/thin1)
   • Just need to activate in Proxmox
   • Will enable efficient storage

2. Enable thin storage on r630-02

   • 6 volume groups available (thin1-thin6)
   • Each ~230GB
   • Enable all for maximum flexibility

3. Verify storage after enabling

   • Test VM creation
   • Test storage migration
   • Monitor performance

Long-term Actions

1. Implement storage monitoring

   • Set alerts for >80% usage
   • Monitor thin pool usage
   • Track storage growth

2. Consider shared storage

   • For easier migration
   • For better redundancy
   • NFS or Ceph options

Performance Recommendations

ml110

• CPU: Older/slower - Reduce workload
• Memory: High usage (75%) - Monitor closely
• Action: Migrate some VMs to r630-01/r630-02

r630-01

• CPU: Good (32 cores) - Ready for workloads
• Memory: Excellent (99% free) - Can handle many VMs
• Action: Enable storage, start deploying VMs

r630-02

• CPU: Excellent (56 cores) - Best performance
• Memory: Excellent (98% free) - Can handle many VMs
• Action: Enable storage, verify existing VMs, deploy new VMs

Network Recommendations

Current Status

• Flat network (192.168.11.0/24)
• All hosts accessible
• Gateway: 192.168.11.1

Recommendations

1. VLAN Migration (Planned)

   • Segment by service type
   • Improve security
   • Better traffic management

2. Network Monitoring

   • Monitor bandwidth
   • Track performance
   • Alert on issues

Security Recommendations

1. Update Passwords

   • Some hosts use weak passwords ("password")
   • Consider stronger passwords
   • Use SSH keys where possible

2. Firewall Configuration

   • Review firewall rules
   • Restrict access where needed
   • Document firewall policies

3. Access Control

   • Review user permissions
   • Implement least privilege
   • Audit access logs

---

🚀 Action Plan

Phase 1: Storage Configuration (CRITICAL - Do First)

1. ✅ Update storage.cfg node references
2. ⏳ Enable local-lvm on r630-01
3. ⏳ Enable thin storage on r630-02
4. ⏳ Verify storage is working

Estimated Time: 15-30 minutes

Phase 2: VM Verification

1. ⏳ List all VMs on r630-02
2. ⏳ Verify VM IP addresses
3. ⏳ Update IP audit if needed
4. ⏳ Test VM accessibility

Estimated Time: 15-30 minutes

Phase 3: Cluster Verification

1. ⏳ Verify cluster status
2. ⏳ Check hostname references
3. ⏳ Update if needed
4. ⏳ Test cluster operations

Estimated Time: 10-15 minutes

Phase 4: VM Distribution (Optional)

1. ⏳ Plan VM migration
2. ⏳ Migrate VMs to r630-01/r630-02
3. ⏳ Balance workload
4. ⏳ Monitor performance

Estimated Time: 1-2 hours (depending on number of VMs)

---

📝 Verification Checklist

Pre-Start Verification

• Hostnames migrated correctly
• IP addresses audited (no conflicts)
• Proxmox services running
• Storage enabled on r630-01
• Storage enabled on r630-02
• VMs on r630-02 verified
• Cluster configuration updated

Post-Start Verification

• All VMs accessible
• No IP conflicts
• Storage working correctly
• Performance acceptable
• Monitoring in place

---

🔧 Quick Fix Commands

Enable Storage on r630-01

ssh root@192.168.11.11
pvesm set local-lvm --disable 0
pvesm set thin1 --disable 0
pvesm status

Enable Storage on r630-02

ssh root@192.168.11.12
for storage in thin1 thin2 thin3 thin4 thin5 thin6; do
    pvesm set "$storage" --disable 0
done
pvesm status

Verify Cluster

# On any node
pvecm status
pvecm nodes

List All VMs

# On each host
pct list
qm list

---

📊 Resource Summary

Host	CPU	Memory (reported)	Guests (2026-05-09)	Notes
ml110	6	~63 GiB · ~3% used	0	Quorum / idle
r630-01	32	~125 GiB · ~66% used	57	Highest count
r630-02	56	~125 GiB · ~62% used	41	Thin pools active
r630-03	32	~125 GiB · ~62% used	19
r630-04	32	~125 GiB · ~17% used	19

Totals: 158 logical cores in cluster; ~562 GiB reported maxmem sum (API); 136 running guests. Storage is node-local thin — see pvesm status per host.

---

🎯 Priority Actions

Historical (migration era)

The checklist below applied when r630 storage names and hostname references were being repaired. For current capacity actions, use PROXMOX_CLUSTER_ARCHITECTURE.md and live pvesm status.

🔴 CRITICAL (historical)

1. Enable storage on r630-01 — validate pools if UI shows disabled stubs
2. Enable storage on r630-02 — thin pools are active on r630-02 (verify utilization)

⚠️ HIGH PRIORITY

1. Continue distributing new CTs across r630-01 … r630-04 using thin pool headroom
2. Reconcile any disabled storage rows that reference foreign node names (cosmetic in pvesm)

📋 RECOMMENDED

1. Monitor thin pool % on busy nodes (r630-01, r630-02)
2. Implement / maintain Prometheus/Grafana per operational stack
3. Keep ALL_VMIDS_ENDPOINTS.md updated when adding VMIDs

---

Last Updated: 2025-01-20
Status: Review Complete - Storage Configuration Needed