Sentinet³ Core

Console and maps

The state of systems in the foreground, from summary views to detailed maps: from the geographic distribution of sites, with successive drill-downs, down to the data center, the rack and the individual component.

• Enterprise Console with host and service states, operational categories, filters and groupings.
• Tactical Overview: a custom page with the priority checks and reports.
• Multi-level, georeferenced maps with visual indicators of host state.
• Link services that summarise the state of the connection between two points on the map.
• Responsive web interface, accessible from desktop, smartphone and tablet.

Time Machine

Go back to a past moment to understand the fault in the context in which it occurred. Time Machine takes the console and maps back to the date and time of interest, reconstructing in a synchronised way the states of systems and services observed at that moment.

• Start from the moment of the outage indicated by the notification or the log.
• See together the hosts and services that were critical.
• Follow maps and relationships to find the area of greatest impact.
• Compare states before and after the anomaly to reconstruct how it evolved and was resolved.
• Ideal when the incident occurred during an unattended period.

Behavioural Learning Engine and event correlation (Root Cause Analysis)

Reading how metrics behave, proposing better-fitting thresholds and tracing a service outage back to its cause: two complementary functions to configure more useful checks and focus troubleshooting on the systems responsible.

• Behavioural Learning Engine: Behavioral Thresholding, Dependency Model detection and Trend Prediction.
• Recurring conditions distinguished from anomalies, and configurations that no longer fit highlighted.
• Event correlation with rules based on the AND and OR operators: several correlated events generate proactive alerts and events.
• Correlation identifies the root cause, which can be viewed in the Root Cause Analysis console.
• Hierarchical view to separate symptoms from the origin of the outage.

Situation Analysis

A statistical view of the infrastructure: what is happening now, what persists and what is getting worse. A specialised dashboard, with widgets that the operator selects, arranges and removes, and drill-down to the operational views.

• Three perspectives: general, by host and service groups, historical.
• KPIs on host states (UP, DOWN, UNREACHABLE) and service states (OK, WARNING, CRITICAL, UNKNOWN).
• Absolute quantities, percentages and durations.
• Top 10 worst-performing systems and persistent issues.
• Comparisons with the previous day and different observation windows.

SLA and reports

Measuring service continuity and making uptime and downtime verifiable. For each check the Core calculates the percentage distribution of states over the selected period.

• Percentages linked to a verifiable timeline: every state change with date, time and duration.
• From the aggregate result to the intervals in which the outage occurred.
• Planned maintenance windows handled without altering the availability measurements.
• Time charts, state change tables and data export.

Alerts, escalation and automation

Events become timely communications, defined responsibilities and possible corrective actions. When a metric exceeds a threshold or a service changes state, the Core notifies the people responsible with the technical context needed for diagnosis.

• Notifications by email, SMS and voice; for environmental scenarios also SNMP traps, HTTP posts and mobile devices.
• On-call groups and hierarchies with escalation to the next level.
• Anti-flooding: a limit on notifications per event, with the excess sent in aggregate form.
• Preconfigured commands and sequences when specific conditions occur.
• Traceability of state changes, notifications, traps, logins and system logs.

Discovery and topology

Automatically discovering hosts, services and relationships, and keeping knowledge of the infrastructure up to date. Autodiscovery identifies the elements to be managed, topology shows where they sit and what they depend on.

• Layer 2 scans (MAC, LLDP, SNMP traps) and layer 3 scans (IP, SSH, SNMP), in normal, fast or thorough mode.
• Finds “ghost” or unregistered devices, updates the inventory and notifies every new or unexpected resource.
• Scheduled searches to catch subsequent changes.
• Checks applied through Patterns and templates.
• Assets and connections reconstructed with SNMP, LLDP and CDP: devices, segments, VLANs and dependencies.

Multi-site and federation

Monitoring sites, data centers and remote systems while keeping a coordinated view of the whole, without losing the detail needed to work on the individual system.

• Host groups, maps and roles that reflect the territorial and technical structure.
• From the overall geographic view to the individual site, rack and service.
• Federated reading of domains belonging to different sites or management perimeters.
• Physical or virtual appliances distributed according to volumes, topology and operating requirements.

Native security: designed for high-security environments

Sentinet³ is designed to work in high-security environments, including classified ones. The Core integrates self-protection measures that limit the exposure of the monitoring system and govern access to the information collected, so that the observation tool does not itself become a point of exposure.

• Access: two-factor authentication (2FA) and an anti brute force mechanism.
• Web applications: internal Web Application Firewall (WAF).
• Network perimeter: integrated firewall.
• Data and secrets: advanced credential protection and encrypted file system.
• Configurations: advanced protection of configuration files and hardening against privilege escalation.
• Isolation: advanced application isolation.