Skip to main content

Verification

Quick checks that confirm prerequisites are in place before — or right after — you import a host or device into monitoring. Most of these you can run in under a minute and they'll surface the common pitfalls before they become support tickets.

Verifying ICMP reachability​

From the Stratora Server (or from each Collector if you're using remote Collectors):

ping <monitored-host-or-IP>

If ping succeeds, the Response column on the Nodes list will populate within ~30 seconds of import.

If ping fails:

  • Check the monitored host's firewall. On Windows, see Windows hosts for the rule to enable; on Linux, see Linux hosts.
  • Check the intervening network path for ACLs or firewall rules blocking ICMP between the Collector and the host.

Verifying SNMP reachability​

From the Stratora Server (or from each Collector):

snmpget -v2c -c <community> <device-ip> 1.3.6.1.2.1.1.1.0

(Or the SNMPv3 equivalent if you're using v3.) A successful query returns the device's sysDescr.

If SNMP fails:

  • Check the device-side ACL for UDP/161 — see SNMP devices.
  • Confirm the community string (for v2c) or USM credentials (for v3) match what's configured on the device.
  • Check the network path between the Collector and the device.

Verifying agent enrollment​

  1. Install the agent on a target host.
  2. Confirm the agent service is running (Get-Service StratoraAgent on Windows, systemctl status stratora-agent on Linux).
  3. The agent sends its first heartbeat within ~10 seconds. When the Server receives it, the host appears in the Stratora UI in pending state.
  4. The pending approval queue is at Infrastructure → Nodes, filtered by Approval Status = Pending. The approve action is per-row on the Nodes list; there is no separate Pending Agents page.
  5. Approve the row. The Response, Collector, and Uptime columns populate on the same row once the next probe cycle completes.

Verifying outbound HTTPS from a monitored host​

From the monitored host:

curl -v https://<your-stratora-fqdn>/api/v1/health

Expected: HTTP 200 response.

If it fails:

  • A certificate-trust issue — the host doesn't trust the Stratora Server's certificate.
  • A DNS resolution issue — the host can't resolve the Server's FQDN.
  • A network-path issue — outbound 443 is blocked or routed through a forward proxy (the agent does not support forward proxies in this release).

Where errors surface in the UI​

  • Nodes list — Response = "-" means no successful probe yet. Stays "-" indefinitely if ICMP Echo Request is blocked on the host.
  • Node detail — per-template error messages from Telegraf parsing surface here.
  • Alerts — connectivity-class alerts fire for the common reachability failures (Node Unreachable, Agent Heartbeat Lost, Collector Offline).
  • Settings → Diagnostics — surfaces Server-side errors for log inspection.