Back to work
Feb 28, 2021
3 min read

Backup and Recovery of IT Systems

Architected and deployed an enterprise backup infrastructure using Veeam Backup & Replication for a hybrid Windows/Linux environment, implementing automated backup strategies, application-aware recovery, and bare-metal restore capabilities across 7 production servers.

Timeline and Details

Start dateEnd dateAssociated withProject URL
January 2021February 2021Algebra Bernays UniversityProject PDF

Overview

Designed and implemented a comprehensive backup and disaster recovery solution for a hybrid Windows/Linux environment. Built automated backup workflows with application-aware processing, item-level recovery, and bare-metal restore capabilities to ensure business continuity and minimize downtime.

Infrastructure Deployed

Servers (7 total)

  • Linux file server (NFS/SMB shares)
  • Linux KVM virtualization host
  • Windows AD DC with DHCP
  • Windows SQL Server with file shares
  • Windows Exchange Server
  • Veeam Backup & Replication manager
  • Veeam Recovery Environment (bare-metal restore)

Storage Architecture

  • Configured iSCSI storage (targetcli + initiator) with 350GB ReFS repository
  • Implemented 64KB block allocation for optimized Veeam performance
  • Automated synthetic full backups weekly with incremental daily backups

Technical Implementation

Backup Strategy

  • Configured protection groups for automated server discovery and monitoring (8-hour scan intervals for critical systems)
  • Implemented application-aware processing for Exchange and SQL Server with transaction log backups
  • Deployed 2-3x daily backup schedules outside business hours (critical systems: 3x/day; others: 2x/day)
  • Enabled guest file system indexing for granular item-level recovery
  • Established 8-day retention policy with weekly health checks to prevent data corruption

Recovery Capabilities

  • Bare-metal recovery: Full server restoration to dissimilar hardware using Veeam RE
  • Application-level recovery: Exchange mailbox item restore and SQL database point-in-time recovery
  • File-level recovery: NFS/SMB share granular restore
  • Successfully tested Active Directory DC recovery with <5-minute RTO

Technologies Used
Veeam Backup & Replication Enterprise Plus • iSCSI (targetcli) • CentOS 7 • Windows Server 2016 • NFS • SMB/CIFS • SQL Server • Exchange Server • ReFS • BackupPC • KVM

Results

  • Achieved 99.9% backup success rate with automated health monitoring
  • Reduced recovery time objective (RTO) to under 5 minutes for critical systems
  • Implemented synthetic full backups to minimize storage consumption (extreme compression enabled)
  • Validated disaster recovery procedures through successful bare-metal and application-level restore tests
  • Ensured business continuity with multiple daily backup points for mission-critical Exchange and SQL servers