Overview
This article provides diagnostic steps to verify that an Aembit Agent Controller running on a Linux Virtual Machine can successfully resolve and connect to the Aembit Cloud. This guide is essential for troubleshooting "Unhealthy" or "Not Connected" states on VM-based deployments.
Relates To
Deployment Type: Linux Virtual Machine (Ubuntu, RHEL, Amazon Linux, etc.)
Component:
aembit_agent_controller.servicePath: VM --> Aembit Cloud (
*.aembit.io) on Port 443.
Cause (Common Symptoms)
Use this guide if the Agent Controller is running but encounters the following:
Service Logs:
journalctlshowstcp connect error: Connection refused (os error 111).Service Logs:
Registration failed. Error: Connection Timeout.Tenant UI: The Agent Controller status remains "Unhealthy" or "Yellow" in the UI.
Solution
Unified Connectivity Diagnostic
1. Identify your Tenant URL
Your endpoint will typically follow this format: <TENANT_ID>.aembit.io.
2. Run the Diagnostic Command
Execute the following as a user with sudo privileges:
sudo apt update && sudo apt install -y bind9-host openssl && \
host -v <TENANT_ID>.ec.us-east-2.useast2.aembit.io && \
echo Q | openssl s_client -showcerts -connect <TENANT_ID>.ec.us-east-2.useast2.aembit.io:443
(Note: If on RHEL/CentOS, use sudo yum install bind-utils openssl instead).
3. Evaluate the Outcome
| Test Layer | Expected Success Signal | Failure Indication |
| DNS Resolution | Status: NOERROR + Aembit IP Address | NXDOMAIN or Connection Timeout. Check /etc/resolv.conf |
| TLS Handshake | Verification: OK | Handshake Failure or Connection Refused. Check local iptables or VPC Security Groups. |
| Cert Chain | Terminating in Amazon Root CA 1 | self signed certificate in certificate chain. Indicates an intercepting proxy |