Systems and Server Administration

Deferred maintenance on servers compounds quietly. Disks fill up, backup jobs fail silently, patches get skipped, and the first sign of trouble is a production outage with no clean recovery path. Subnet Works manages server environments to prevent that, and steps in when it has already happened.

Common problems businesses call about

• A server that has not been patched in months (or years)
• Backup jobs that have been failing silently
• Running out of disk space with no clear picture of what is using it
• Needing to migrate workloads off aging physical hardware
• Adding a new service or application without the in-house expertise to deploy it
• Not having a documented recovery procedure for any of the above

What the engagement covers

Operating systems

• Ubuntu and Debian Server: installation, hardening, patching, and ongoing administration
• Windows Server: Active Directory, Group Policy, DNS, DHCP, file shares, and patch management
• System hardening: disabling unused services, configuring host-based firewalls, managing SSH access

Virtualization

• Proxmox VE: hypervisor setup, KVM guest deployment, resource allocation, and cluster configuration
• VM lifecycle management: snapshots, cloning, live migration, and backup integration
• GPU passthrough for workloads requiring dedicated hardware access

Containers and services

• Docker and Docker Compose: service deployment, network configuration, volume management
• Reverse proxy setup with Caddy or Nginx: TLS termination, path-based routing, virtual hosting
• Service orchestration and dependency management across compose stacks

Storage

• ZFS: pool design (RAIDZ1, RAIDZ2, mirror), dataset management, snapshot scheduling, and scrub
• NFS share configuration with access controls tied to network topology
• Disk usage analysis and capacity planning

Backup and recovery

• Automated backup pipelines with defined retention schedules
• Off-site or off-volume backup targets to protect against local failure
• Backup monitoring with alerting on job failure
• Documented recovery procedures tested before they are needed

Monitoring and observability

• Resource utilization tracking (CPU, RAM, disk I/O, network throughput)
• Log aggregation and alerting for critical service events
• Uptime monitoring with notification on service failure

Representative scope examples

Virtualization migration. A small business in Essex County was running three aging physical servers, each a single point of failure with no backups. Subnet Works migrated all workloads to a Proxmox VE host with ZFS RAIDZ1 storage, configured automated snapshot-based backups to a separate pool, and documented the recovery procedure for each VM.

Silent backup failure recovery. A professional services firm discovered their nightly backup jobs had been failing for two months after a volume path change following a routine OS update. There was no alerting. Subnet Works audited the pipeline, corrected the configuration, implemented job monitoring with email alerting on failure, and wrote a recovery runbook.

Docker stack deployment. A small company needed to run several internal web applications without the cost of cloud hosting. Subnet Works deployed the services via Docker Compose on an existing Ubuntu server, configured Caddy as a reverse proxy with automatic TLS for each subdomain, set up log rotation, and handed over documentation covering deployment, update, and rollback procedures.