Timeline and Details
oVirt Implementation
| Start date | End date | Associated with | Project URL |
|---|---|---|---|
| November 2020 | December 2020 | Algebra Bernays University | Project PDF |
vSphere Implementation
| Start date | End date | Associated with | Project URL |
|---|---|---|---|
| February 2021 | March 2021 | Algebra Bernays University | Project PDF |
Overview
Designed and deployed production-ready virtualization infrastructure demonstrating enterprise features: automated failover, live migration, and storage multipathing. Implemented 2-node ESXi cluster with redundant storage across dual protocols (iSCSI + NFS), achieving zero-downtime operations. Automated infrastructure provisioning through Bash scripting, reducing deployment time by 80%.
oVirt Implementation
Architecture Design
Infrastructure Components
- Deployed KVM hypervisor with oVirt Engine for centralized management
- Configured 4 virtual machines: Engine node + 3 compute nodes
- Implemented NFS and iSCSI storage backends
Automation
- Developed Bash scripts for VM lifecycle management (start, stop, clone)
- Automated network and storage configuration
- Scripted node provisioning for rapid deployment
Technical Stack
oVirt • KVM • CentOS • Bash Scripting • NFS • iSCSI
vSphere Implementation
Architecture Design
Cluster Infrastructure
- 2 ESXi hosts (VMware ESXi 6.5.0) managed by vCenter Server
- 2 CentOS storage servers (iSCSI target + NFS server)
- Network segmentation: Management, storage (dual paths), HA, FT, vMotion
- Total shared storage: 120 GB across 6 datastores
Storage Configuration
- iSCSI: 3 LUNs (20 GB each) with targetcli on CentOS 01
- NFS: 3 shares (19.75 GB each) on CentOS 02
- MPIO Round Robin: 6-path configuration per LUN for load balancing and failover
- Dual storage networks (192.168.20.0/24 and 192.168.30.0/24) for redundancy
High Availability & Advanced Features
- vSphere HA: Automatic VM restart on host failure (60-second restart time)
- vMotion: Zero-downtime live migration between hosts
- Storage vMotion: Live storage migration between iSCSI and NFS
- Fault Tolerance: Dedicated network for mission-critical workloads
- Cluster: MojKlaster62 with 2 nodes
Automation Scripts
Developed Bash automation for CentOS storage servers:
- Network configuration: Dual NICs with static IPs
- Disk provisioning: Automated partitioning, formatting (ext4), mounting
- iSCSI target: targetcli configuration with LUN creation and ACL management
- NFS exports: Automated share creation with persistent fstab entries
Deployment Impact: Reduced provisioning from 2+ hours to under 5 minutes per server
Testing & Validation
Successfully tested:
- Compute migration: ESXi 01 ↔ ESXi 02 with zero downtime
- Storage migration: iSCSI ↔ NFS with zero downtime
- HA failover: VM automatic restart within 60 seconds
- MPIO failover: Automatic path switching on network failure
Technical Stack
VMware vSphere 6.5 • VMware ESXi • vCenter Server • CentOS 7 • iSCSI (targetcli) • NFS v3 • VMFS 6 • vMotion • vSphere HA • Fault Tolerance • MPIO Round Robin • Bash
Results
oVirt Outcomes
- Successfully deployed KVM-based private cloud with centralized management
- Automated VM operations through custom scripting
- Demonstrated dual-protocol storage integration
vSphere Outcomes
- Achieved zero-downtime migrations for both compute and storage
- 6-path MPIO configuration with automatic failover
- 80% reduction in deployment time through automation
- Production-ready HA cluster with 60-second failover capability
Key Learnings
- Storage multipathing: MPIO Round Robin essential for I/O distribution and redundancy
- Network segmentation: Dedicated VLANs for storage, vMotion, HA, FT critical for performance
- Automation impact: Scripting eliminates manual errors and accelerates provisioning
- vMotion capabilities: Enables maintenance windows without service interruption