Cloud & Virtualization Power Cheat Sheet for Sysadmins

☁️ Cloud & Virtualization Power Cheat Sheet for Sysadmins

Quick commands and notes for real-world hybrid environments (Proxmox, OpenNebula, KVM, AWS, GCP).

---

🔹 Proxmox VE (PVE)

Cluster & Node Management

pvecm status                         # Check cluster status
pvecm nodes                          # List cluster nodes
pvecm add <node-ip>                  # Join new node to cluster
pvecm expected <number>              # Adjust quorum expectation

VM & Container Management

qm list                              # List all virtual machines
pct list                             # List all LXC containers
qm start <vmid>                      # Start VM
qm stop <vmid>                       # Stop VM
qm shutdown <vmid> --timeout 30      # Graceful shutdown

Live Migration & Snapshots

qm migrate <vmid> <target-node> --online  # Live migrate VM
qm snapshot <vmid> preupdate              # Create snapshot
qm rollback <vmid> preupdate              # Rollback snapshot

Backup & Restore

vzdump <vmid> --storage local --mode snapshot --compress zstd --bwlimit 20000 --ionice 5
qmrestore /var/lib/vz/dump/vzdump-qemu-123.vma.zst <vmid> --unique true

Flags Explained:

• --mode snapshot: No downtime (uses QEMU guest agent)
• --compress zstd: Fast compression
• --bwlimit: Throttle bandwidth for stability
• --ionice: Prioritize I/O scheduling

Cloud-Init Templates

qm create 9000 --name "ubuntu-template" --memory 2048 --cores 2 --net0 virtio,bridge=vmbr0
qm importdisk 9000 ubuntu-22.04.qcow2 local-lvm
qm set 9000 --scsihw virtio-scsi-pci --scsi0 local-lvm:vm-9000-disk-0
qm set 9000 --ide2 local-lvm:cloudinit
qm set 9000 --boot c --bootdisk scsi0
qm set 9000 --serial0 socket --vga serial0
qm template 9000

💡 Tip: Automate template cloning:

qm clone 9000 101 --name web01 && qm start 101

Performance Tuning

• Enable NUMA and CPU pinning for DB workloads:

  qm set <vmid> --numa 1 --cpulimit 2 --cpu host
  

• Ballooning off for latency-sensitive workloads.
• Use VirtIO SCSI single for optimal disk I/O.

---

🔹 OpenNebula (CLI for oneadmin)

Cluster Overview

onehost list                         # List hosts
onedatastore list                    # Show datastores
onevm list                           # List all VMs

VM Lifecycle

onevm create template.yaml
onevm show <vmid>
onevm shutdown <vmid>
onevm terminate <vmid>
onevm migrate <vmid> <hostid> --live

Use --live for live migration.

Templates & Datastores

onetemplate list
onetemplate instantiate <template_id> --name "web01"
onedatastore show <id>

Image & Snapshot Management

oneimage create --name "ubuntu22" --path /var/lib/one/images/ubuntu.qcow2 --datastore default
onevm disk-snapshot-create <vmid> <diskid> "before-update"

Hooks & Automation

cat /etc/one/hooks.d/backup_vm.yaml
# Trigger backups on terminate

Tips:

• Use persistent images for databases, non-persistent for stateless workloads.
• Integrate with Terraform/OpenNebula provider for IaC.
• Use oneacct for billing and resource usage reports.

---

🔹 KVM / QEMU / Libvirt

VM Management

virsh list --all
virsh start <vmname>
virsh shutdown <vmname>
virsh destroy <vmname>               # Force stop

Create VM

virt-install --name ubuntu24 --ram 4096 --vcpus 2 \
  --disk path=/var/lib/libvirt/images/ubuntu24.qcow2,size=20,bus=virtio \
  --network bridge=br0,model=virtio \
  --cdrom /iso/ubuntu24.iso

Snapshots

virsh snapshot-create-as <vm> snap1 "Before patching"
virsh snapshot-list <vm>
virsh snapshot-revert <vm> snap1

Storage & Network

virsh pool-list
virsh net-list
virsh net-start default
virsh net-destroy default

Advanced Tuning

virsh edit <vmname>   # Edit XML directly

Add:

<cpu mode='host-passthrough'/>
<numatune>
  <memory mode='strict' nodeset='0'/>
</numatune>

💡 Automation Tip: Use virt-sysprep before distributing base images:

virt-sysprep -a /var/lib/libvirt/images/base.qcow2 --enable customize

---

🔹 AWS CLI Essentials

Configure

aws configure

EC2 Instance Inventory

aws ec2 describe-instances \
--query "Reservations[].Instances[].{Name:Tags[?Key=='Name']|[0].Value,PrivateIP:PrivateIpAddress,PublicIP:PublicIpAddress,State:State.Name}" \
--output table

Volumes, Snapshots, and AMIs

aws ec2 describe-volumes --output table
aws ec2 create-snapshot --volume-id vol-123 --description "Backup before update"
aws ec2 describe-images --owners self

Networking

aws ec2 describe-vpcs
aws ec2 describe-security-groups

Cost & Usage

aws ce get-cost-and-usage --time-period Start=2025-10-01,End=2025-10-31 \
--granularity MONTHLY --metrics "UnblendedCost" --output table

💡 Tip: Use jq to extract just what you need:

aws ec2 describe-instances | jq -r '.Reservations[].Instances[] | [.InstanceId, .State.Name]'

---

🔹 GCP (gcloud CLI)

Setup

gcloud config set project my-project
gcloud auth login

Compute Engine Management

gcloud compute instances list
gcloud compute instances start web01 --zone=us-central1-a
gcloud compute instances stop web01 --zone=us-central1-a

Networking & Storage

gcloud compute networks list
gcloud compute firewall-rules list
gcloud storage buckets list
gcloud storage cp file.txt gs://mybucket/

Snapshots & Disks

gcloud compute disks snapshot mydisk --snapshot-names=snap-before-patch
gcloud compute snapshots list

IAM & Billing

gcloud projects get-iam-policy my-project
gcloud billing accounts list

💡 Pro Tip: Simplify SSH access:

gcloud compute config-ssh
ssh my-vm

---

🔹 Hybrid Cloud Ops & Automation

Unified Inventory (Ansible)

ansible-inventory -i inventory_aws_ec2.yml --graph
ansible-inventory -i inventory_proxmox.yml --list

Terraform Multi-Provider Example

provider "proxmox" {
  pm_api_url = "https://pve.local:8006/api2/json"
  pm_user    = "root@pam"
  pm_password = var.pm_password
}

provider "aws" {
  region = "eu-central-1"
}

resource "proxmox_vm_qemu" "web01" {
  name        = "web01"
  target_node = "pve01"
  cores       = 2
  memory      = 2048
  disk {
    size = "20G"
  }
}

resource "aws_instance" "web_backup" {
  ami           = "ami-0abcdef1234567890"
  instance_type = "t3.micro"
  tags = {
    Name = "web-backup"
  }
}

So when you run:

  terraform init
  terraform plan
  terraform apply

Terraform creates:

• A VM called web01 on Proxmox
• An EC2 instance called web-backup on AWS

---

Monitoring Stack

• Prometheus: node_exporter, libvirt_exporter, aws_exporter
• Grafana: dashboards per environment
• Alertmanager: unified alerting

---

🔹 Tips & Best Practices

Networking

• Use separate VLANs for management, storage, and guest.
• Implement bonded NICs for redundancy.

Performance

• Enable CPU passthrough & NUMA for DB workloads.
• Always enable VirtIO drivers in guests.

Security

• Use cloud-init with SSH keys only — no passwords.
• Rotate cloud credentials regularly.

Automation

• Centralize IaC in Git + CI/CD.
• Schedule backups and test restores monthly.

Cost Optimization

• AWS: Use spot instances where possible.
• GCP: Use committed use discounts.
• PVE/OpenNebula: Use thin-provisioned storage.