feat: basic health-checker #218

New issue

Open

opened 2026-03-28 04:22:26 +00:00 by mfreeman451 · 0 comments

mfreeman451 commented

2026-03-28 04:22:26 +00:00

Owner

Imported from GitHub.

Original GitHub issue: #605
Original author: @mfreeman451
Original URL: https://github.com/carverauto/serviceradar/issues/605
Original created: 2025-04-14T13:43:26Z

Monitors HTTP or gRPC health endpoints (e.g., /health, /livez, /readyz) on services, checking availability and parsing JSON status (e.g., {"status": "healthy"}). Reports health (true/false) and key details (e.g., component statuses) to agent.

Value Proposition:

Unique Niche: Targets modern microservices (e.g., Kubernetes apps, APIs) with health endpoints, which sysmon (system-level) and rperf (network) don’t cover. Unlike snmp (devices) or dusk (blockchain), it’s application-focused.
Lightweight: Fetches a small JSON response (100 bytes) every 30s, storing minimal data (e.g., healthy: true, details: {"db": "ok"}). Fits SQLite’s ~24 GB/day for 100,000 hosts (0.3 MB/day/host).
Proxmox Fit: Monitors containerized apps or APIs on Proxmox (e.g., LXC containers), complementing sysmon’s ZFS/CPU metrics.
Security: Supports mTLS for gRPC to agent and HTTP basic auth/TLS for endpoints, aligning with tls-security.md.
Ease of Use: Simple HTTP GET or gRPC call, no complex parsing (unlike Prometheus metrics). Configurable endpoints and expected status fields.

Imported from GitHub. Original GitHub issue: #605 Original author: @mfreeman451 Original URL: https://github.com/carverauto/serviceradar/issues/605 Original created: 2025-04-14T13:43:26Z --- Monitors HTTP or gRPC health endpoints (e.g., /health, /livez, /readyz) on services, checking availability and parsing JSON status (e.g., {"status": "healthy"}). Reports health (true/false) and key details (e.g., component statuses) to agent. Value Proposition: * Unique Niche: Targets modern microservices (e.g., Kubernetes apps, APIs) with health endpoints, which sysmon (system-level) and rperf (network) don’t cover. Unlike snmp (devices) or dusk (blockchain), it’s application-focused. * Lightweight: Fetches a small JSON response (100 bytes) every 30s, storing minimal data (e.g., healthy: true, details: {"db": "ok"}). Fits SQLite’s ~24 GB/day for 100,000 hosts (0.3 MB/day/host). * Proxmox Fit: Monitors containerized apps or APIs on Proxmox (e.g., LXC containers), complementing sysmon’s ZFS/CPU metrics. * Security: Supports mTLS for gRPC to agent and HTTP basic auth/TLS for endpoints, aligning with tls-security.md. * Ease of Use: Simple HTTP GET or gRPC call, no complex parsing (unlike Prometheus metrics). Configurable endpoints and expected status fields.

Rows
Columns