InfoFounding Members35 lifetime licenses available — view the offer > https://uptriva.com/foundingInfoMCP launch announcementConnect Claude, Cursor or any MCP client with your API token — let an agent create monitors, manage incidents and configure alerts. Read the docs → https://uptriva.com/docs/api/mcp
All check types
Apps & Workloads

Internal Agent Monitoring

Agent monitoring puts a lightweight, self-updating binary inside your network — reporting process deaths, systemd failures, OOM kills, SSH logins, file-integrity changes, port changes and resource pressure on hosts no external probe can reach.

See what external probes physically cannot

What it checks

  • Process and systemd unit state
  • OOM kills and kernel events
  • SSH logins, failed-auth bursts and new accounts
  • File integrity on configs and authorized_keys
  • CPU, memory, disk, load and network interface state

Set it up

  1. 1Install the agent binary on the host — one command
  2. 2Watch the host appear in the dashboard on first heartbeat
  3. 3Create event rules and monitor types for the signals you need

FAQ

Which platforms does the agent run on?

Linux (amd64/arm64), macOS and Windows — systemd units on Linux, log show on macOS, the Security event log on Windows.

Is it resource-heavy?

A single Go binary designed for minimal footprint — it reads /proc, journals and sockets on an interval, not a metrics daemon.

Related check types

Heartbeat / Cron MonitoringThe check for jobs that must call in — or something is wrongAPI MonitoringEndpoints that answer are not endpoints that workGame Server MonitoringPlayers see 'server offline' before your dashboard does — unless you watch the query port

Start monitoring in minutes

External probes plus internal agents, escalation chains and status pages — free tier, no card required.

Start free Read the docs