How ɳSentry works

Product-layer observability on top of your ɳSelf stack. Status pages, incident management, SLOs, and RUM — all self-hosted, all yours.

Free infra monitoring stays free. Prometheus, Grafana, Loki, and the full monitoring stack ship in every ɳSelf instance at no cost. ɳSentry operates at the product layer, above infrastructure — these two work together, not instead of each other.

Status pages your customers can actually see

Your infrastructure monitoring (Prometheus, Grafana) is for you. ɳSentry adds a public-facing status page so your customers know when services are up or down. Configure which checks feed each page, set maintenance windows, and subscribe users to email or webhook alerts.

Public or private status pages
Per-service and per-region breakdowns
Subscriber notifications via email or webhook
Maintenance window scheduling

Incident management from alert to post-mortem

When something breaks, ɳSentry opens an incident, pages the right person, and tracks every action until resolution. Alerts route through your preferred channels: Slack, email, PagerDuty, or webhooks. After resolution, a post-mortem template is waiting.

Auto-open incidents from uptime or SLO breach
Multi-channel alert routing
On-call rotation management
Post-mortem templates and timelines

SLO tracking and error budgets

Define the reliability targets that matter to your users — uptime percentages, latency thresholds, error rates. ɳSentry tracks your error budget in real time and alerts before you burn through it. Engineers see exactly how much reliability headroom is left.

Availability and latency SLOs
Error budget burn rate dashboards
Alert before budget is exhausted
Rolling window calculations (7d, 28d, 30d)

Real-user monitoring without the surveillance

Measure what your actual users experience: Core Web Vitals, page load times, JS errors. ɳSentry collects performance data from real sessions without cookies, third-party scripts, or data leaving your server. You see the numbers, nobody else does.

Core Web Vitals (LCP, CLS, FID/INP)
JS error capture with stack traces
No third-party scripts or cookies required
Data stays on your server

Uptime checks from your own infrastructure

HTTP and TCP uptime checks run from your VPS on a schedule you set. No external check service has access to your internal services. Scripted browser checks (via nself-synthetic-monitor) can walk multi-step flows and report on the result.

HTTP and TCP uptime checks
Configurable intervals and thresholds
Scripted browser workflow checks
Cron job heartbeat monitoring

Want early access?

ɳSentry ships in v1.1.0. Join the waitlist and get an invite code when it launches.

Join the waitlist Self-host guide