Nagios Trainer

Getting a working Nagios you understand rather than one you inherited. Dependencies and prerequisites, manual installation from source versus package, directory layout and where each configuration fragment lives, the web interface and its authentication, and a safe upgrade path for an existing installation.

Topics: Nagios dependencies and prerequisites · Manual installation from source · Package-based installation and its trade-offs · Directory layout: etc, libexec, var, objects · Web interface setup and authentication · Verifying configuration before restart · Upgrading an existing installation safely · Nagios Core versus Nagios XI — what actually differs

• Assignments: (1) Install Nagios from source and from package, then diff the two layouts; (2) Break a configuration file and recover using the verification command
• Capstone: Build the lab Nagios you keep for the rest of the course, with its configuration under version control

The configuration structure that decides whether an estate scales. Hosts and services, hostgroups and servicegroups, contacts and contactgroups, commands and timeperiods — and above all templates and inheritance, which is the difference between a hundred-line configuration and a ten-thousand-line one.

Topics: Nagios terminology and the object types · Host definitions and their directives · Service definitions and service dependencies · Hostgroups and servicegroups · Contacts and contactgroups · Command definitions and macro expansion · Time periods and exclusions · Templates, inheritance and object reuse · Splitting configuration across files and directories · Custom object variables

• Assignments: (1) Refactor a flat configuration into templates with a single host definition per host; (2) Model a service dependency so a database outage does not page for every dependent app
• Capstone: Design an object structure for a multi-environment estate where adding a host is a three-line change

How Nagios actually decides something is broken. Host checks versus service checks, the plugin exit-code contract, SOFT and HARD state types and max_check_attempts, active versus passive checks, the scheduler and check interleaving, flapping detection, and runtime modification of parameters through the external command file.

Topics: Host checks and service checks: the difference that surprises people · Plugin exit codes: OK, WARNING, CRITICAL, UNKNOWN · State types: SOFT and HARD, and max_check_attempts · check_interval, retry_interval and notification_interval · Active and passive checks; freshness checking · Check scheduling, interleaving and spreading load · Flap detection and its thresholds · Scheduled downtime and acknowledgements · Runtime modification of Nagios parameters via the external command file · Reading nagios.log and the status file

• Assignments: (1) Trace one check from schedule to HARD state to notification using the logs; (2) Tune check and retry intervals for a flapping service and prove the improvement
• Capstone: Produce a check-scheduling design for a thousand-service estate that spreads load evenly

The output of the whole system, and the part most estates get wrong. Notification preferences on hosts, services and contacts, the filters a notification must pass before it is sent, escalations, timeperiod-based routing, customising message content, email and mobile delivery, and how to make on-call quiet enough to be trusted.

Topics: Notification logic: every filter a notification must pass · Configuring notification preferences on contacts, hosts and services · notification_options and when each state notifies · Notification methods: email, SMS, webhook, chat integrations · Sending notifications via e-mail and the mail command chain · Customising notification messages with macros · Customising mobile providers and gateways · Escalations: first_notification, last_notification, escalation periods · Scheduling notifications with timeperiods · Reducing alert volume: dependencies, downtime, acknowledgement discipline

• Assignments: (1) Build an escalation chain from team to lead to manager with correct timing; (2) Rewrite a notification template so the message is actionable without opening the console
• Capstone: Deliver a notification model for a 24x7 service with escalation, quiet hours and documented ownership

Where Nagios gets its reach. The standard plugin set, the plugin contract that lets any executable become a check, writing custom plugins in shell and Python, packaging and distributing them, addons that extend the core, and event handlers that take corrective action on a state change instead of merely reporting it.

Topics: The Nagios plugin contract: exit codes, output format, performance data · The standard plugin set and what each covers · Writing a custom plugin in shell and in Python · Plugin arguments, thresholds and the range syntax · Performance data and graphing addons · Addons: NRPE, NSCA, NDOUtils and their roles · Managing plugins across a fleet · Event handlers: acting on a state change · Global versus per-object event handlers and their risks

• Assignments: (1) Write a custom plugin including threshold parsing and performance data; (2) Build an event handler that restarts a service on a HARD CRITICAL and logs the action
• Capstone: Deliver a plugin library for your estate, packaged and deployed by configuration management

Getting inside a machine rather than just pinging it. NRPE for Linux — installation, configuration, command definitions, allowed hosts and the arguments debate — and NSClient++ for Windows desktops and servers. Then the agentless alternatives, and the monitoring wizards that speed up repetitive host onboarding.

Topics: Installing and configuring the Linux agent (NRPE) · nrpe.cfg, allowed_hosts and command definitions · check_nrpe and the argument-passing security decision · Using the Linux monitoring wizard · Installing NSClient++ on Windows machines · Monitoring a Windows desktop · Monitoring a Windows server: services, event log, disks, performance counters · Agentless options: SSH checks, WMI, SNMP · Passive check submission with NSCA · Deploying agents at scale with configuration management

• Assignments: (1) Onboard a Linux and a Windows host with agents and a common service template; (2) Compare an NRPE check with the equivalent agentless SSH check and justify a choice
• Capstone: Bring a mixed Linux and Windows fleet under monitoring with one reusable template set

Everything you monitor from the outside. Network devices over SNMP and ICMP, routers and switches with interface and bandwidth checks, websites and URLs including content and certificate validation, DNS, FTP, and the common TCP and UDP ports — plus hostgroup and servicegroup configuration for publicly enabled services.

Topics: Monitoring network devices: ICMP, SNMP, interface status · Monitoring a router or switch: bandwidth, errors, port state · SNMP walks, OIDs and check_snmp · Monitoring a website and specific URLs · Content checks, redirects and TLS certificate expiry · Monitoring DNS resolution and authority · Monitoring FTP servers · Monitoring common TCP and UDP ports · Publicly enabled services: HTTP, FTP and friends · Hostgroup and servicegroup configuration for network estates · Parent-child host relationships and network outage suppression

• Assignments: (1) Model a network topology with parent hosts so a switch failure reports one outage, not forty; (2) Alert on certificate expiry thirty days ahead across every public endpoint
• Capstone: Deliver network monitoring for a site with correct parent-child topology and SNMP-based interface checks

Running Nagios for an organisation. Checking system status, managing users and permissions, cloning accounts and masquerading for support, managing components and plugins, then the Core Config Manager as an alternative to editing files by hand, and dashboards that give each audience the view it needs.

Topics: Checking system status and performance of the Nagios process itself · Managing users, roles and permissions · Cloning user accounts and masquerading as a user for support · Managing components and managing plugins · Accessing and navigating the Core Config Manager · CCM workflows: creating objects, applying configuration, rollback · Understanding and using dashboards · Customising dashboards per audience · Reports: availability, SLA, state history, alert histogram · Backup, restore and configuration change control

• Assignments: (1) Create a role that can acknowledge alerts but not change configuration; (2) Build two dashboards — one for on-call, one for a service owner
• Capstone: Deliver an administration model with roles, dashboards, reporting and a documented change process

Growing the estate without growing the manual work, and knowing when to stop. Auto-discovery jobs and what to do with their output, distributed monitoring with NSCA and mod_gearman, performance tuning for large configurations, integration with configuration management, and an honest assessment of when a check-based system should be complemented or replaced.

Topics: Running auto-discovery jobs and reviewing the results · Bulk host onboarding from an inventory source · Generating Nagios configuration from configuration management · Distributed monitoring: NSCA, mod_gearman, and worker patterns · Performance tuning: large configurations, check latency, reaper intervals · Integrating with ticketing and chat tooling · Nagios alongside Prometheus, Zabbix or a hosted platform · Auditing an inherited configuration: which checks still matter · Planning a migration with a parallel-run period

• Assignments: (1) Generate host and service definitions from an inventory file automatically; (2) Audit an inherited configuration and classify every check as keep, fix or retire
• Capstone: Produce a scaling and modernisation plan for an existing Nagios estate, with a parallel-run migration path