VM Linux is automatically shutdown unexpectedly due to agent issue. So, can it resolve.

Question

VM Linux automatically shuts down unexpectedly due to an agent issue. Then, we need to restart manually and be on standby at all times, which may not always be feasible. So, I can suggest how to resolve it.

Answer 1

Dear Bimala,

Thank you for bringing this issue to our attention. Unexpected shutdowns due to the Azure Linux agent can indeed disrupt operations, and I understand the need for a reliable solution to minimize manual intervention. Am I right? if so pls check the Azure Linux Agent Status Connect to the VM via SSH and run

systemctl status waagent

If the agent is inactive, restart it

systemctl restart waagent

next review Agent Logs for Errors Investigate the logs to identify the root cause

sudo apt update && sudo apt upgrade -y walinuxagent

Update the Azure Linux Agent Ensure the latest version is installed for exampple for Ubuntu/Debian

sudo apt update && sudo apt upgrade -y walinuxagent

for RHEL/CentOS

sudo yum update -y WALinuxAgent

Enable Auto-Recovery Features in Azure

• Configure Auto-shutdown (if applicable) in the Azure Portal under VM > Settings > Auto-shutdown.
• Set up Azure Monitor Alerts to notify you of unexpected shutdowns.

Automate VM Recovery

• Use Azure Automation to trigger automatic restarts if the agent fails.
• Implement a custom watchdog script to monitor and restart the agent if needed.
• Schedule regular maintenance windows to update the VM and agent.
• Ensure backups are configured via Azure Backup for quick recovery if needed.

If the issue persists after these steps, may be need to reinstall the VM agent or consider redeploying the VM while preserving the disks.

Best regards,

Alex

P.S. If my answer help to you, please Accept my answer

Answer 2

Dear Bimala Shrestha

Okay ) lets deeg in ))) long read if u ready. So to ensure your Linux VM project continues running without user disruption I'm with u will do some an implementation plan for automated recovery, so correct me me if i will write somethink wrong okay?

Set Up Azure Automation (Cloud-based monitoring)

1. Create Automation Account

• Azure Portal > Automation Accounts > Create
• Name: VM-Recovery-Automation
• Enable Managed Identity

1. Create PowerShell Runbook

Schedule. Set to run every 5 minutes

Alerts > Create alert rule for failed executions

On-VM Watchdog Script (Immediate fallback)

Create /usr/local/bin/agent-watchdog.sh

So like taht.... but in general if its really critical instead all of it u can deploy a secondary VM in Availability Set, configure Load Balancer to shift traffic if primary fails and Enable Azure Site Recovery for VM replication, that would be really best.

Hope that would help u.

Best regards,

Alex

P.S. If my answer help to you, please Accept my answer