AIX, short for Advanced Interactive eXecutive, is IBM’s UNIX operating system for Power hardware. AIX monitoring is the practice of collecting availability and performance data from those systems on a repeating schedule, storing it, and alerting when a value crosses a threshold you defined. In practice that means processor and memory usage, filesystem capacity, network interface activity, and whether the processes and services the business depends on are still running.

AIX ships with strong native diagnostic tooling, including topas, nmon, vmstat, and lparstat. Those tools answer a different question. They tell an administrator who is already logged in to a specific system what is happening right now, at high resolution. Operational monitoring answers whether anyone needs to log in at all, across every AIX Host at once, with history retained for trend analysis and an audit trail of state changes. The two are complementary: native tools for diagnosis, Nagios XI for detection, notification, and reporting.

AIX infrastructures tend to be small in Host count and large in business impact. A handful of systems can carry the transaction load for an entire organization, and the people who understand them deeply are often a small team. Continuous monitoring converts their knowledge into checks that run whether or not anyone is watching.

• Catch capacity problems before they stop work: Filesystem capacity and memory pressure build gradually and are visible long before an application fails, provided something is measuring them every few minutes.
• Know when a critical process stops: Process and service checks turn a silent exit into an alert, instead of a discovery made hours later.
• Give the on-call team one place to look: AIX, Linux, Windows, network devices, and databases report into the same Nagios XI interface, with the same Host and Service model.
• Support availability commitments with evidence: Availability reporting is generated from recorded state history rather than reconstructed after an incident.
• Reduce notification fatigue: Thresholds, notification escalations, and scheduled downtime keep routine noise from burying the alerts that matter.
• Preserve institutional knowledge: Checks, thresholds, and contacts written into Nagios XI outlast the individual administrator who knew which value was normal.

AIX Server Monitoring with the NCPA Agent

NCPA is the agent-based path, and the one Nagios’ current AIX documentation recommends first for AIX 7.3.

• Single cross-platform agent: The same agent runs on AIX, Windows, Linux, macOS, and Solaris, so one deployment and one firewall rule pattern covers a mixed environment.
• One TLS port: Nagios XI polls the agent over TCP port 5693, authenticated with a token rather than a system account.
• Wizard-driven setup: The NCPA Wizard scans the target Host for drives and running services, then pre-fills the metrics it finds so you adjust thresholds instead of typing check definitions.
• Active and passive in one agent: Checks can be polled by Nagios XI, or pushed from the AIX Host to the Nagios Remote Data Processor (NRDP) when the network path only allows outbound traffic.
• Agent-side visibility: The NCPA web interface exposes a browsable check API, system information, and real time graphs, which makes validating a check far quicker than editing configuration blind.
• Experimental status on AIX: AIX 7.3 support in NCPA v3 is labeled experimental in Nagios’ installation documentation, and it was introduced in NCPA v3.3.0. Plan a pilot Host rather than an system-wide rollout on day one.

Documentation: NCPA v3 Agent Installation Instructions, which includes the AIX RPM install, service control, and token configuration steps.

Agentless AIX Monitoring with SNMP

Where installing an agent is not permitted, Nagios XI monitors AIX by polling the SNMP daemons that ship with the operating system. This is the path to use on AIX levels below 7.3, and in environments with a hard no-agent policy.

• No software installed on the AIX Host: Monitoring uses snmpd together with the aixmibd, hostmibd, and snmpmibd subsystems already present on AIX.
• Configuration file supplied by Nagios: Nagios publishes an AIX SNMP configuration file that exposes only the object identifiers the Nagios XI server needs, rather than opening the full MIB tree.
• Discovery during configuration: At Step 2 the wizard queries the AIX Host and returns a scanned disk list and scanned process list, so filesystems and processes are selected from what actually exists.
• Reuses a wizard that ships with Nagios XI: AIX is configured through the Linux SNMP Wizard. AIX is not Linux, and there is no dedicated AIX wizard, but the configurations the Linux SNMP Wizard generates work against AIX SNMP data.

Documentation: How To Monitor AIX Using SNMP In Nagios XI 2024 And 2026.

AIX Performance Monitoring Metrics

Both paths produce performance data that Nagios XI graphs and retains, so a threshold breach can be read in the context of the preceding days or weeks.

• Processor usage: CPU utilization collected on the check interval you define.
• Memory usage: Memory consumption tracked per Host, with separate Warning and Critical values.
• Filesystem and disk usage: Capacity per mounted filesystem, reported as a percentage used alongside the absolute figures, which is what makes a slow fill visible before it becomes an outage.
• Network interface activity: Traffic on the interfaces the wizard detects.
• Performance graphs and trends: Collected values feed the Nagios XI graphing layer, where a Time Stacked Performance Graph compares the current period against the previous ones, and a Multistacked Performance Graph puts metrics from several Hosts or Services on one chart.

Documentation: Generating Graphs With Nagios XI 2024 & 2026.

AIX Process and Service Monitoring

• Process count checks: Select a process from the scanned list and set the acceptable number of running instances. A single number alerts when the count falls to or below it, so a stopped daemon raises Warning or Critical rather than passing silently.
• Acceptable ranges: Entering two values defines a range, which catches both a process that has died and a process that has spawned far more instances than expected.
• Running services: On the NCPA path, the wizard scan returns running services on the AIX Host so they can be monitored alongside resource metrics.

Custom AIX Checks and Plugins

Environments with site-specific requirements are not limited to what a wizard offers.

• Any language, one contract: A check is any executable that returns exit code 0 for OK, 1 for Warning, 2 for Critical, or 3 for Unknown, with optional performance data after a pipe character. Shell, Perl, and Python are all normal choices on AIX.
• Run custom plugins through NCPA: Custom scripts placed in the agent’s plugin directory can be executed as active checks or submitted passively, which keeps AIX-specific logic on the AIX Host.
• Community-contributed AIX plugins: Nagios Exchange carries an AIX plugin category covering filesystems, error report entries, multipath devices, and similar checks. These are contributed and maintained by their authors rather than bundled with Nagios, several were written for AIX 5.3 through 7.1, and some expect NRPE or the Nagios Service Check Acceptor (NSCA) rather than a current agent. Install them separately, read the code, and test them before relying on them.
• Generic wizard for one-off checks: The Generic Plugin Wizard in Nagios XI wraps a plugin you supply into Host and Service definitions without hand-editing configuration files.

Thresholds, Notifications, and AIX Reporting

• Thresholds you specify: Every check compares its result against Warning and Critical values you set. Nagios XI does not infer what normal looks like; the threshold is the definition of normal, which makes each alert explainable.
• Notifications and escalations: Alerts route by contact and Host group, with escalations for unacknowledged problems and scheduled downtime to suppress alerts during planned maintenance.
• Availability reporting: The default report is an Availability Summary built from recorded state history, and reports can be limited to a single Host, Host group, or Service group.
• SLA reporting and Capacity Planning: Both are Enterprise Edition features rather than part of every license. Service-Level Agreement (SLA) reporting measures recorded uptime against a target percentage you set, and Capacity Planning projects future usage from performance data already collected, which is useful for AIX environments where hardware refresh cycles are long. A 60-day trial of the Enterprise features can be activated from within Nagios XI if you want to evaluate them first.
• Scheduled delivery: Reports and dashboard pages can be emailed on a schedule to people who will never log in to Nagios XI.

Documentation: Generating Reports With Nagios XI 2024 & 2026, Generating SLA Reports With Nagios XI 2024 & 2026, and How To Use Capacity Planning.

Security and Access Control for AIX Monitoring

• Token authentication on the agent path: NCPA authenticates with a token held in its configuration file, not an operating system login, so monitoring does not require a shell account on the AIX Host. Use a placeholder such as <NCPA_TOKEN> in runbooks and avoid shell metacharacters in the token itself, since they can interfere with check execution.
• Encrypted agent traffic: NCPA listens over TLS on port 5693 and generates a self-signed certificate at install time, so the first browser visit produces an expected certificate warning.
• Treat SNMP community strings as passwords: Under SNMP v1 and v2c the community string crosses the network in cleartext and grants read access to system state. Do not leave it set to public, restrict which source addresses may poll, and prefer authenticated and encrypted SNMP where your AIX level and policy allow it.
• Expose the minimum: The AIX SNMP configuration file Nagios supplies limits the visible object identifiers to those the checks need.
• Role-based access in Nagios XI: User permissions and multi-tenancy scope who can see and act on which AIX Hosts, which matters when platform teams and application teams share one monitoring server.
• Protect the monitoring server itself: A Nagios XI server holds credentials and tokens for everything it monitors. Keep the interface off the public internet and apply releases promptly.

Fewer surprises on systems nobody wants to touch. AIX Hosts often run untouched for long periods, which is a virtue operationally and a risk in monitoring terms. Scheduled checks with defined thresholds mean a filling filesystem or a stopped process produces an alert with a timestamp, rather than an incident review that starts with nobody knowing when the problem began.

One operational picture across a mixed environment. Most organizations running AIX also run Linux, Windows, network hardware, and databases. Bringing AIX into the same Nagios XI interface removes the separate console that only two people know how to read, and lets Host groups follow the service rather than the platform.

Evidence for availability conversations. Because state history is recorded continuously, availability figures come from the monitoring record. That changes discussions with application owners and auditors from recollection to data.

Choice about agents, decided by policy rather than by tooling. Some AIX environments allow an agent and some do not. Supporting both NCPA and agentless SNMP means the monitoring approach can follow your change-control reality instead of forcing an exception.

Room for what only your team knows. The plugin model means an internal check for a batch window, a queue depth, or a business-specific condition becomes a monitored Service with the same alerting, graphing, and reporting as a processor check.

Cost that tracks the environment. Nagios XI is licensed by Nodes (Hosts), so an AIX footprint of a few dozen systems is licensed as a few dozen Nodes, and monitoring depth per Host is not metered.

Most AIX monitoring Support Cases trace back to one of the items below. Confirm them before running a wizard.

For the NCPA Agent Path

• AIX level: The current NCPA v3 AIX package targets AIX 7.3 TL4. Support is documented as experimental, and it arrived in NCPA v3.3.0.
• Agent version: Deploy NCPA 3.4.3 or newer on AIX rather than an earlier 3.x release. Numerous AIX and Solaris packaging, service autostart, and configuration-preservation issues were fixed across the 3.2.3 to 3.4.3 releases, with 3.4.3 current as of August 2026.
• No in-place upgrade from NCPA v2: An inline upgrade from v2 to v3 is not supported. Remove NCPA v2 first if it is present.
• Install method: The AIX package is an RPM, installed with dnf localinstall on the AIX Host.
• Network path: TCP port 5693 reachable from the Nagios XI server to the AIX Host. On AIX the firewall is not enabled by default and permits 5693, so the constraint is usually an external firewall rather than the Host.
• Token: A token set in the agent configuration file, which the wizard validates before it will continue past Step 1.
• Root access on the AIX Host: Required to install the package and edit the configuration file, not to run the checks.

For the Agentless SNMP Path

• SNMP configured on the AIX Host: Nagios’ documented walkthrough configures SNMP v2c and sets a community string, then restarts the aixmibd, hostmibd, snmpmibd, and snmpd subsystems with stopsrc and startsrc.
• Network path: UDP port 161 reachable from the Nagios XI server to the AIX Host.
• A community string that is not the default: Replace public, treat the value as a credential, and limit which addresses may poll.
• Comfort editing AIX configuration files: The documentation assumes you can edit operating system configuration files on the AIX server.
• A successful test query first: Verify SNMP from the Nagios XI server before opening the wizard. Nagios XI includes SNMP plugins in /usr/local/nagios/libexec, and a storage check run from there confirms both SNMP configuration and firewall reachability.

On the Nagios XI Side

• No Premium Feature required: The NCPA Wizard and Linux SNMP Wizard both ship with Nagios XI and are not on the Premium Wizard list, so AIX monitoring does not depend on active Maintenance & Support benefits. Maintenance & Support remains the route to upgrades, security releases, and Case Support.
• Node capacity: Each AIX Host consumes one Node against your license, including each logical partition monitored as its own Host.

The steps below follow the NCPA path, which is the current recommendation for AIX 7.3. For the agentless route, substitute the SNMP preparation and the Linux SNMP Wizard.

1. Confirm the AIX level and remove any existing NCPA v2 installation.
2. Download the AIX NCPA RPM to the AIX Host and install it with dnf localinstall, then confirm the subsystem is running with lssrc -s ncpa.
3. Set the token in /usr/local/ncpa/etc/ncpa.cfg, then restart the agent with stopsrc -s ncpa followed by startsrc -s ncpa.
4. Verify the agent by browsing to its interface on port 5693 from the Nagios XI server and logging in with the token, which also confirms the network path.
5. In Nagios XI, navigate to Configure > Configuration Wizards, select the NCPA Wizard, and enter the AIX Host address, port, and token. The wizard validates the token before continuing.
6. At Step 2, confirm the Host name, select the metrics returned by the scan, and set Warning and Critical thresholds for each one.
7. Complete the remaining steps and click Finish, then open the status details link the wizard provides to confirm the new Host and Services are checking in.

Full walkthroughs: How To Monitor AIX With Nagios XI 2024 And 2026, NCPA v3 Agent Installation Instructions, How To Monitor Devices Using NCPA In Nagios XI, and How To Monitor AIX Using SNMP In Nagios XI 2024 And 2026. The AIX package is on the NCPA downloads page.

Nagios XI is the operational layer: is the Host Up, is each Service OK, has a measured value crossed a threshold, and who needs to know. Each check returns a value that Nagios XI compares against the figures you supplied, which is what makes AIX monitoring predictable to run and easy to explain to an auditor.

It also sits alongside the rest of the Nagios Monitoring Solutions, each licensed separately, so an AIX estate can grow into them as needs change. Centralized collection, search, and retention of AIX system logs and error report output is Nagios Log Server. Traffic flow analysis across the network your AIX systems sit on is Nagios Network Analyzer. A single view across several independent Nagios XI servers, split by site, segment, or tenant, is Nagios Fusion.

Other Operating Systems

• UNIX Monitoring
• Solaris Monitoring
• Linux Monitoring
• Operating System (OS) Monitoring

Servers, Processes, and Storage

• Server Monitoring
• UNIX Service & Process Monitoring
• Filesystem Monitoring
• Disk Monitoring

Protocols and Applications on AIX

• SNMP Monitoring
• DB2 Monitoring
• Oracle Monitoring
• WebSphere Monitoring

Reporting and Availability

• Ensuring Service-Level Agreements
• Increasing Availability
• Reducing IT Downtime

Other Nagios Monitoring Solutions

• Nagios XI
• Nagios Log Server
• Nagios Network Analyzer
• Nagios Fusion

If your AIX systems are currently monitored by habit and institutional memory, the fastest way to test the alternative is to instrument one Host. Pick a single AIX server, choose the agent or agentless path according to your change-control rules, and run the wizard. Processor, memory, filesystem, and process checks with thresholds and notifications take minutes, not a project.

Nagios XI is self-hosted software that you install and run on your own infrastructure, so an evaluation happens inside your network with your own AIX Hosts and firewall rules. Start with the free trial, or ask for a walkthrough with someone who can look at your environment with you.