Observability & Logging

Overview

Cloud Orbiter provides comprehensive observability for your Kubernetes clusters — monitoring health, performance, and security in real time. Observability is achieved through a combination of metrics collection, centralized logging, alerting, and audit trails.

Observability is critical for:

• Identifying and resolving issues quickly before they impact applications
• Predicting usage patterns and planning for growth
• Protecting clusters from potential threats
• Meeting compliance and audit requirements

Prometheus-Based Monitoring

Cloud Orbiter uses Prometheus — the industry-standard open-source monitoring system — for metrics collection and alerting in Kubernetes environments.

Enabling Prometheus

Prometheus is available as a cluster Add-On. Without it enabled, real-time time-series charts are unavailable, though basic resource usage information (CPU/RAM by namespace, network I/O) remains accessible.

To enable Prometheus:

1. Log in to Cloud Orbiter and navigate to the target cluster
2. If not enabled, click the Go to Add-on button shown in the monitoring section
3. On the Add-Ons page, locate Prometheus
4. Click the three-dot menu on the right → click Enable

Once enabled, full real-time monitoring and alerting capabilities are available.

Key Metrics

Metric	What It Shows	Why It Matters
Node Ready vs Total Nodes	Number of ready nodes vs total nodes	Ensure all nodes are operational and ready to serve requests
Node RAM Usage vs Capacity	RAM used by nodes vs available capacity	Prevent memory exhaustion; detect memory leaks
Node CPU Usage vs Capacity	CPU utilization vs available capacity	Identify overloaded nodes; plan capacity scaling
Pod Usage vs Capacity	Running pods vs cluster pod limit	Identify if cluster is approaching maximum capacity
CPU Use by Namespace	CPU consumption per namespace	Identify resource-hungry namespaces; optimize allocation
RAM Use by Namespace	Memory consumption per namespace	Optimize resource allocation per team or workload
Network I/O Pressure	Network input-output pressure on the cluster	Detect network bottlenecks; optimize cluster network performance

Access Logs

Access logs are a crucial aspect of monitoring cluster activity, identifying security threats, and troubleshooting operational issues.

Cluster-Level Access Logs

Cloud Orbiter captures access logs per cluster:

• Every API call to the cluster is recorded
• Log entries include: user identity, timestamp, IP address, HTTP method, API endpoint

To access cluster logs:

1. Navigate to the cluster dashboard
2. Select the Access Logs tab

Audit Logs

Audit logs capture all administrative actions within Cloud Orbiter:

• Tenant-level audit trail for compliance and security reviews
• Per-entry data: username, operation, IP address, API endpoint, timestamp

Example audit log entry:

Time	Username	Operation	IP Address	API
<timestamp>	<user-email>	<HTTP method>	<client-ip>	<api-endpoint>

To access Cloud Orbiter audit logs:

1. Log in as Tenant Administrator
2. Navigate to Settings → Access Logs

Regular review of audit logs is recommended to detect anomalies and potential security threats.

Live Logs

Cloud Orbiter provides real-time log streaming from running containers:

• View live output from any running pod directly in the dashboard
• No kubectl port-forwarding or SSH access required
• Useful for real-time debugging and deployment validation

To access live logs:

1. Navigate to the cluster
2. Select Access Logs → Live Logs
3. Select the namespace and pod to stream logs from

Test Suites

Cloud Orbiter includes pre-packaged test suites for proactive cluster validation:

Built-In Test Suites

Pre-configured test cases that verify cluster health and performance — run before application deployment or after infrastructure changes.

Bring Your Own Test Suites

Organizations can add custom test suites to Cloud Orbiter:

1. Navigate to Test Suites on the cluster dashboard
2. Click Bring Your Own Test Suite
3. Upload or reference your custom test configuration
4. Run the test suite on the target cluster

Downloading Logs and Reports

After test suite execution:

1. Select the completed test run
2. Click Download Logs or Download Report
3. Logs and results are available as downloadable files for sharing or archiving

Notification System

Cloud Orbiter sends real-time dashboard notifications for key infrastructure events:

Events That Trigger Notifications

Resource	Trigger Event	Recipients
Cluster	Created / Imported / Deleted	Project Admin, Default User
Cluster	Connected / Disconnected	Project Admin, Default User
Cluster	Scaled up / Scaled down	Project Admin, Default User
Application	Deployment success / failure	Project Admin
Test Suite	Execution started / completed	Project Admin, Default User
Roles	User role updated or changed	Affected user
User Membership	Membership granted or revoked	Project Admin, Default User

Managing Notifications

Notifications appear as pop-up windows in the bottom-right corner of the dashboard. From the notification bell icon, you can:

• View last 10 notifications without navigating away from the current page
• Filter to show unread only — focus on items requiring attention
• Mark as read / unread — track which notifications have been reviewed
• Navigate to the resource — click a notification to jump directly to the affected resource
• Delete notifications — remove notifications no longer needed
• Automatic deletion — notifications expire and are deleted automatically after 15 days