Amazon monitoring is the practice of collecting performance and availability metrics from AWS services, applying thresholds to those metrics, and alerting when a threshold is crossed. Nagios XI reads the metrics AWS publishes to CloudWatch, stores the results as Services attached to a Host in your monitoring configuration, and presents cloud resources alongside the on-premises servers, network devices, and applications you already monitor.

• Credit exhaustion is silent: A burstable EC2 instance throttles when its CPU credit balance reaches zero, and the symptom looks like a slow application rather than a resource limit.
• Storage growth becomes a billing surprise: Bucket size and object count trend upward without a triggering event, so nothing prompts a review until the invoice arrives.
• Error rates identify the failing side: Separating 4XX from 5XX responses tells you whether a client is requesting something it cannot have or the service itself is failing.
• Latency degrades before it breaks: First byte latency climbing over weeks is a capacity signal that an availability check will never report.
• Cloud and on-premises belong in one view: Splitting an application across a data center and an AWS region splits the alert trail unless both halves report into the same monitoring server.

Amazon EC2 Instance Monitoring

The Amazon EC2 Wizard connects with an Access Key ID and Secret Access Key, lists the instances available to that Identity and Access Management (IAM) user, and creates Services for the metrics you select. Instance-level detail is covered on the Amazon EC2 monitoring page.

• CPU utilization: Measure processor load to identify instances that are saturated or oversized for their workload.
• CPU credit usage and credit balance: Track consumption and remaining balance on burstable instance types before throttling begins.
• Disk read/write operations: Watch operation counts to see when storage activity, rather than the processor, is the constraint.
• Disk read/write bytes: Measure throughput volume moving to and from instance storage.
• Network in/out: Track bytes transferred to establish a normal range and alert on departures from it.
• Network packets in/out: Separate packet rate from byte volume to distinguish many small requests from a few large transfers.
• Instance Up/Down: Confirm the instance is reachable and running.

Documentation: How To Monitor Amazon EC2 In Nagios XI | How To Troubleshoot EC2 Data Issues In Nagios XI

Amazon S3 Bucket Monitoring

The Amazon S3 Wizard lists the buckets available to your credentials and creates Services per bucket for the metrics you select. Bucket-level detail is covered on the Amazon S3 monitoring page.

• Bucket size in bytes: Trend stored volume so growth is a planned conversation rather than an invoice line.
• Number of objects: Count objects per bucket, which matters for lifecycle rules and listing performance.
• All, Get, and Put requests: Measure request volume to establish the normal working rate for each bucket.
• Delete requests: Track deletion volume against the rate your retention policy expects.
• Head, Post, and List requests: Watch metadata and listing operations, which often dominate request counts in application workloads.
• Bytes downloaded and uploaded: Measure transfer volume in each direction.
• 4XX errors: Catch client-side failures such as denied or missing object requests.
• 5XX errors: Catch service-side failures affecting the applications that depend on the bucket.
• First byte latency and total request latency: Separate time to first response from end-to-end request time to locate where delay accumulates.

Documentation: How To Monitor Amazon S3 In Nagios XI

Operating System Monitoring Inside EC2 Instances

CloudWatch reports on the instance as an AWS resource. To monitor what runs inside it, install the Nagios Cross-Platform Agent (NCPA) on the instance and configure it with the Linux Server or Windows Server Wizard. NCPA reports memory, filesystem usage, running processes, and services, giving an EC2 guest the same coverage as Linux server monitoring on hardware you own. Allow inbound TCP 5693 from your Nagios XI server in the instance security group and the host firewall.

Documentation: NCPA Agent Installation Instructions | How To Monitor Linux With Nagios XI With NCPA

Threshold Configuration and Alerting

• Per-metric thresholds: Set Warning and Critical values on each selected metric in Step 2 of either wizard.
• State-based notification: Nagios XI notifies on transitions into Warning, Critical, and back to OK, so a sustained problem does not generate repeat noise unless you configure it to.
• Contact groups and escalations: Route cloud alerts to the team that owns the account rather than to every recipient.
• Historical graphing: Every numeric metric is graphed, which is how a credit balance trending toward zero becomes visible before it lands there.

AWS Credential and Access Control

• Dedicated IAM user: Create a monitoring-only IAM user rather than supplying account root or administrator credentials.
• Read-only permission set: The EC2 Wizard needs cloudwatch:GetMetricStatistics, cloudwatch:GetMetricData, cloudwatch:ListMetrics, ec2:DescribeInstances, and ec2:DescribeRegions. Grant nothing beyond what the checks call.
• Key rotation: Rotate the Access Key and Secret Access Key on the schedule your policy requires, then update the wizard-created Host configuration.
• Nagios XI user rights: Restrict which Nagios XI users can view or edit the cloud Hosts, since the stored credentials sit in the check configuration.

One interface for cloud and on-premises. AWS resources appear as Hosts and Services next to your physical servers, switches, and databases, so an application spanning both is diagnosed from one screen instead of two consoles.

Thresholds you set, not defaults you inherit. Warning and Critical values are yours to define per metric and per instance, which means a batch-processing instance and a customer-facing instance can carry different definitions of a problem.

Trend data for capacity conversations. Stored performance data turns bucket growth, request rates, and credit consumption into a chart you can bring to a budget review.

Alerting your team already knows. Cloud checks use the same notification, escalation, and acknowledgement workflow as the rest of your monitoring, so nothing new has to be learned or maintained.

• Active Maintenance & Support benefits. Both the Amazon EC2 Wizard and the Amazon S3 Wizard are Premium features and will not function without them.
• An IAM user with an Access Key ID and Secret Access Key. Generate these in the AWS Management Console under your account’s security credentials.
• CloudWatch and EC2 describe permissions. Without the five actions listed above, the Get Available Instances button in the wizard returns an error.
• Outbound HTTPS from the Nagios XI server. The checks call AWS API endpoints on TCP 443.
• The boto3 Python library on the Nagios XI server. The Amazon S3 Wizard requires it.
• Detailed Monitoring enabled on the instance. Without it, most EC2 metric checks return no data. Enable it from the Monitoring tab of the instance details using Manage detailed monitoring.
• A CloudWatch Agent for disk metrics. If Detailed Monitoring is on and only the disk checks return no data, install the CloudWatch Agent on the instance.
• An ICMP rule in the security group. A new EC2 instance blocks ping by default, so the Host alive check fails until you add the rule.

1. Create the IAM user and key pair. Attach a policy granting the CloudWatch and EC2 describe actions listed above, then record the Access Key ID and Secret Access Key. Treat the secret as you would any other credential and store it as <API_TOKEN> in your own runbooks.
2. Prepare the instances. Enable Detailed Monitoring, add an ICMP rule to the security group if you want a ping-based Host check, and install the CloudWatch Agent where you need disk metrics.
3. Run the Amazon EC2 Wizard. Navigate to Configure > Configuration Wizards, select the Amazon EC2 Wizard, enter your keys, click Get Available Instances, and choose the instance to monitor.
4. Select metrics and thresholds. In Step 2, enter a Host Name, select the metrics you want, and set Warning and Critical values for each. Complete Steps 3 through 5 and click Finish.
5. Run the Amazon S3 Wizard. Repeat the process with the Amazon S3 Wizard, clicking Get Available Buckets in Step 1 to list the buckets your credentials can reach.

Documentation: How To Monitor Amazon EC2 In Nagios XI | How To Monitor Amazon S3 In Nagios XI | Understanding And Using Configuration Wizards In Nagios XI

Nagios XI checks the availability and performance of AWS resources against thresholds you define, and alerts and graphs the results. The EC2 and S3 Wizards read metrics AWS publishes to CloudWatch, so a metric that CloudWatch does not expose is not available to the check. For visibility inside an instance, install NCPA and use the operating system wizards.

Centralizing CloudTrail or application logs from AWS for search and retention is a separate function handled by Nagios Log Server, and flow analysis of Virtual Private Cloud traffic belongs to Nagios Network Analyzer. Both are licensed separately from Nagios XI. Nagios XI does not act on AWS resources on your behalf, so it will not resize, restart, or terminate an instance in response to a threshold breach.

Related Solutions

Automation and Orchestration

• Chef monitoring
• Kubernetes monitoring
• Red Hat OpenShift monitoring
• Docker monitoring

Systems Ansible Commonly Manages

• Linux monitoring
• Red Hat Enterprise Linux (RHEL) monitoring
• Windows Server monitoring
• Server monitoring

Operational Outcomes

• Linux Service and process monitoring
• Reducing IT downtime
• Application monitoring
• System and application log monitoring

Nagios XI ships with the Amazon EC2 and Amazon S3 Wizards, so a working set of cloud checks is a credential and a few clicks away. Download the free trial to configure them against your own account, or request a demo and we will walk through the wizards with you.