Unit
- Role scope: A system administrator ensures availability, performance, security, and capacity of computing resources. This includes servers, storage, networking, and application services across on-premise, cloud, and hybrid environments.
- Operational models: Organizations adopt various models including dedicated sysadmins, shared operations teams, and DevOps fusion roles. Each model defines escalation paths, change windows, and ownership boundaries.
- ITIL alignment: The Information Technology Infrastructure Library provides a framework for service management. Key processes include incident management, problem management, change management, and service request fulfillment. ITIL helps standardize procedures and reduce downtime.
- Environment types: Production, staging, development, and disaster recovery environments have distinct requirements. Production demands strict change control and high availability, while development allows experimentation. Segregation prevents unintended impact.
- Documentation standards: A runbook documents standard operating procedures (SOPs) for common tasks. It includes startup/shutdown sequences, troubleshooting steps, and escalation contacts. Well-maintained runbooks reduce mean time to recovery (MTTR).
- Monitoring expectations: Administrators rely on dashboards showing CPU, memory, disk, and network metrics. Alerting thresholds should trigger notifications before resources become exhausted. Proactive monitoring prevents outages.
- Change management: Changes follow a formal process: request, review, approval, implementation, and post-implementation review. This reduces risk of misconfiguration and provides an audit trail for compliance.
- Communication channels: Incident bridges, chat platforms, and ticketing systems coordinate response. Clear communication during outages keeps stakeholders informed and reduces confusion. Regular status updates are essential.
- Backup and recovery: Administrators implement scheduled backups with retention policies to protect against data loss. Recovery testing ensures backups are restorable. RTO and RPO define acceptable downtime and data loss limits. Regular validation prevents surprises during actual disasters.
- Patch management: Regular patching addresses security vulnerabilities and software bugs. A patch cycle includes testing in staging, scheduled maintenance windows, and rollback plans. Missing patches are a common root cause of breaches. Automation tools help streamline deployments across the fleet.
- Automation and scripting: System administrators automate repetitive tasks using Bash, PowerShell, or configuration management tools like Ansible. Automation reduces human error, enforces consistency, and frees time for complex troubleshooting. Infrastructure as code further standardizes deployments.
- Capacity planning: Administrators analyze usage trends to forecast future resource needs. This includes CPU, memory, storage, and network bandwidth. Proactive capacity planning prevents performance degradation and avoids emergency upgrades. Historical data and growth projections drive decisions.
Code Example
Bash
# Example runbook snippet: verify system health
echo "=== System Health Check ==="
uptime
df -h / | tail -1
free -m | grep "Mem:"
systemctl is-active sshd
journalctl -n 5 --no-pagerThis simple health check script outputs uptime, disk usage, memory, service status, and recent logs. System administrators run such checks before applying patches or during incident triage.
System Architecture
Pattern: Three-tier operational model
Components:
- Layer 1 - Infrastructure: Physical or virtual servers, network switches, storage arrays. Administrators manage hardware lifecycle and firmware updates.
- Layer 2 - Platform: Operating systems, hypervisors, container runtimes. Configuration management ensures consistent OS builds across the fleet.
- Layer 3 - Application: Web servers, databases, middleware. Administrators collaborate with developers to deploy and troubleshoot application services.
Quick Revision
- A system administrator manages availability, performance, security, and capacity of IT resources.
- ITIL processes like incident, problem, and change management bring structure to operations.
- Runbooks document standard procedures and reduce mean time to recovery.
- Environments must be segregated to prevent production disruptions during testing.