AI-Powered Fleet Monitoring

Your Platform Fleet, One Dashboard, Zero Blind Spots

Addax is an AI-powered fleet monitoring solution for infrastructure and Quicksand Platform deployments. Centralized visibility, intelligent diagnostics, and automated remediation — deployed on your own infrastructure, under your control.

Your Installations Are Running. Who’s Watching Them?

You deploy Quicksand to protect your customers. Five installations become twenty. Twenty become a hundred. Each one runs on different hardware, different networks, different configurations.

A certificate expires at 2 AM. A disk fills up over the weekend. A container crashes and nobody notices until a customer calls. Replication lag creeps in. A protection rule silently fails. By the time you know, the damage is done.

Generic monitoring tools don’t understand Quicksand Platform workloads. They see CPU and memory but miss protection effectiveness, container health cascades, and configuration drift. You end up with dashboards that show green while your security posture degrades.

Addax was built for exactly this problem.

See Everything. Fix It Before It Breaks.

Addax connects to every installation in your fleet and builds a complete operational picture — from infrastructure health to protection effectiveness. When something goes wrong, it doesn’t just alert you. It tells you why, and can fix it automatically.

Fleet Dashboard

Real-time status of every installation at a glance. OK, Warning, Critical, Unreachable — with one-click drill-down into full telemetry. CPU, memory, disk, container health, database metrics, protection stats, and error tracking. Auto-refreshing, mobile-friendly, always current.

Deep Telemetry

Goes far beyond basic system metrics. Addax understands Platform element-specific health: blocked traffic counters, protection rule effectiveness, request pipeline latency, database replication status, container image freshness, pending security updates, and application-level error rates. Everything you need to assess whether an installation is truly healthy.

Certificate & Endpoint Monitoring

Tracks SSL/TLS certificate expiry across every protected domain. Configurable alert thresholds — know weeks in advance, not hours. HTTP health checks verify that every endpoint is reachable, responding correctly, and performing within expected latency. Automatic domain discovery means new sites are monitored the moment they’re deployed.

Auto-Remediation

Common issues resolved in seconds, not hours. Restart failed containers. Clean up disk space. Rotate logs. Reload configurations. Clear caches. Remove zombie processes. Every action is fully audited — who triggered it, what ran, what happened, timestamped and traceable.

Multi-Channel Alerting

Notify the right people through the right channel. Webhooks, email, Slack, Nextcloud Talk, and syslog — with per-install filtering, severity-based routing, maintenance window awareness, and 30-minute deduplication to prevent alert fatigue. Escalation policies ensure critical issues reach on-call engineers when initial alerts go unacknowledged.

Configuration Drift Detection

Pin a known-good configuration as your baseline. Addax automatically detects when live configurations deviate — whether from unauthorized changes, failed deployments, or silent corruption. Know immediately when an installation is no longer running what you expect.

AI-Powered Diagnostics — On Your Infrastructure

Addax includes an integrated AI analysis engine that runs entirely on your own hardware. No data leaves your network. No cloud API calls. No third-party dependencies.

The AI examines each telemetry snapshot, compares it against the previous collection cycle, and produces structured diagnostics: problem classification, confidence score, business impact assessment, and specific recommended actions. It understands Platform-specific patterns — distinguishing a post-deployment container restart from an actual failure, recognizing attack-mode resource spikes as expected behavior, and correlating related events into a single root cause.

What the AI adds: context-aware analysis that turns raw metrics into actionable intelligence. Instead of “memory usage 92%”, you get “Memory pressure caused by ClickHouse query cache growth after bulk import — recommend cache eviction or scheduled restart during maintenance window.”

How It Works

1. Register — Add installations manually or receive them automatically from the Partner Portal provisioning pipeline. Each installation gets secure, isolated access credentials.

2. Collect — Addax reaches into each installation on a configurable interval and gathers comprehensive telemetry: system resources, container state, application metrics, protection stats, database health, and certificate status.

3. Analyze — AI diagnostics process every snapshot, compare against previous state, and flag anomalies. Rule-based checks catch known failure patterns. Capacity forecasting predicts when resources will be exhausted.

4. Alert — Notifications route through configured channels with deduplication and escalation. Maintenance windows suppress noise during planned work. OK-transition alerts confirm when issues resolve.

5. Remediate — Common issues are resolved automatically or with one-click approval. Every action is logged for compliance. Runbooks codify your team’s operational knowledge into repeatable, auditable procedures.

Use Cases

MSP Fleet Operations

Manage dozens or hundreds of customer Quicksand Platform installations from a single pane of glass. Standardize monitoring across heterogeneous environments. Reduce mean time to resolution with AI diagnostics and automated remediation. Generate SLA compliance reports showing uptime, incident count, and resolution times per customer.

Enterprise Security Operations

Centralize visibility across distributed Platform deployments — branch offices, data centers, edge locations. Feed security events into your existing SIEM via syslog or webhook. Detect protection degradation before attackers exploit it. Maintain a complete audit trail for incident response and regulatory review.

Proactive Capacity Planning

Historical telemetry stored in a time-series database enables trend analysis and capacity forecasting. Know which installations will run out of disk space next month. Identify memory pressure trends before they cause outages. Plan upgrades based on data, not guesswork.

Compliance & Audit

Every monitoring event, every remediation action, every configuration change — logged with operator identity, timestamp, and outcome. Certificate expiry tracking ensures continuous TLS compliance. Configuration drift detection catches unauthorized changes. Generate evidence for NIS2, ISO 27001, and SOC 2 controls directly from operational data.

Deployment

Self-Hosted
Deploy on your own VM or container host. Full data sovereignty — telemetry, credentials, and AI analysis never leave your network. Minimal footprint: runs comfortably on 4 vCPU / 4 GB RAM.

Automated Provisioning
Integrated with the Quicksand Partner Portal. New installations are registered automatically when provisioned through the standard deployment pipeline. No manual setup required.

Secure Connectivity
Mutual TLS for all control-plane communication. Encrypted remote access with bastion/gateway support for installations behind firewalls. Per-installation isolated credentials.

Graceful Degradation
Continues operating when components are unavailable. AI analysis, alert channels, or time-series storage can fail independently without affecting core monitoring. No single point of failure.