site1.erralert.com
System & uptime
Uptime, memory and CPU snapshots, Windows event-log error counts. The "is this machine generally healthy" category.
Event log error count (last hour)
Counts Error-level events in the System and Application logs over the past hour. A sharp jump usually means something is broken.
Memory + CPU snapshot
Posts current free RAM (%) and a 1-second CPU sample. Run frequently (every minute) for 'find machines under sustained load' alerts.
Uptime (Linux / macOS)
Posts /proc/uptime as days. Use with capture.value to nag for reboots after N days.
Uptime (Windows)
Posts uptime in days. Useful for "find machines that have not rebooted in N days" alerts.
Windows host agent — all metrics (one payload)
The all-in-one agent: posts a single JSON blob with uptime, CPU, memory, every disk (size / free / SMART health / SSD-HDD-NVMe-USB type), selected services, pending-reboot state, pending Windows Updates, and 24h event-log error counts. Create ONE capture probe per host and point this at it — the probe page renders a full <strong>Host metrics</strong> dashboard automatically. Pair with <code>capture.freshness</code> (alert if the agent stops) plus <code>capture.value</code> checks on paths like <code>cpu.load_pct</code>, <code>memory.used_pct</code>, <code>min_disk_free_pct</code>, <code>updates.pending</code>, or <code>any_disk_unhealthy</code>.