Skip to main content

Troubleshooting TvRMM

Most TvRMM troubleshooting starts by confirming tenant visibility, endpoint identity, platform support state, and whether the action is owned by a direct agent, a host, or an appliance-native workflow.

Common issues​

Endpoint missing after install​

Generate a fresh portal bootstrap, confirm the selected organization, check outbound connectivity to the agent endpoint, and verify the agent service is running on the target host.

Action button disabled​

Check your effective role and the operating-as selector. Actions such as scripts, terminal, patch install, reboot, homelab controls, user management, and deletion require more than viewer access.

VM or container row is inventory-only​

A VM or container row may be reported by a host instead of directly managed by an in-guest agent. Install and link a real in-guest agent when direct endpoint behavior is required.

Use Guest row routing and direct-agent authority to confirm whether the visible row is an unlinked host-observed guest, linked guest context, or direct in-guest agent.

Patch scan differs by OS​

Different operating systems use different update tools, repositories, and appliance rules. If a patch result looks surprising, compare it with the endpoint's native update tool and confirm the platform support state before changing policy.

Policy did not apply​

Check the endpoint's effective policy source, assignment level, and whether an explicit No policy assignment blocks inheritance. Then confirm the endpoint has checked in since the policy changed.

Remote access controls are disabled​

Open the Remote tab and read the session state. Remote access can be blocked by policy, unsupported endpoint type, offline agent state, license state, missing endpoint helper, role limits, or relay availability.

For a host-observed guest, remote desktop requires a linked or direct real in-guest agent. Open the linked direct agent endpoint when the Remote tab says to use the linked agent.

Evidence to collect​

  • Tenant, organization, endpoint name, and whether the endpoint has a directly installed agent.
  • Operating system, architecture, appliance type, package manager, and repository state where relevant.
  • Last heartbeat time, latest audit time, command result stage, and any visible portal error message.
  • The action attempted and the role shown by the operating-as selector.
  • Effective policy name and source when the issue involves patching, logs, monitoring, scripts, agent updates, or remote access.
  • For guest rows, whether the row is host-observed only, linked to a direct agent, or the direct in-guest agent row.

Agent cleanup and removal​

When an endpoint cannot reach the portal, has duplicate local agents, points at the wrong tenant, or has stale certificate material, use the dedicated Agent cleanup and removal runbook.

For install verification, update recovery, identity recreation, or relocation boundaries, use Agent lifecycle and recovery.