I recently created a new Ubuntu VM on Azure and successfully accessed it via SSH. I then ran a CI/CD pipeline that made three connections to the VM (for frontend, backend, and one other connection). After the pipeline completed, I can no longer access the VM via SSH, and even pinging the VM's IP address fails. The issue persists even after redeploying the VM.
Details:
- New Ubuntu 24.04 LTS VM on Azure
- Initially, SSH access was working fine
- NSG rule allows SSH (port 22) from any IP
- CI/CD pipeline (using GitHub Actions) connected successfully
- After pipeline completion, all connectivity to VM lost (SSH and ping)
- VM status in Azure portal shows as running
- VM agent status is reported as ready
- Redeployment of the VM did not resolve the issue
- Ping to the VM's IP address does not return any response, despite the VM showing as started
Steps Taken:
- Verified VM is shown as running in Azure portal
- Checked NSG rules - SSH still allowed
- Attempted to access via Azure Serial Console
- Reviewed CI/CD logs for potential network configuration changes
- Redeployed the VM, but the issue persists
Questions:
- What could cause a VM to become completely unresponsive after a CI/CD run, even surviving a redeployment?
- Are there any Azure-specific networking issues that might lead to this situation?
- How can I troubleshoot network connectivity when the VM appears to be running but is completely unresponsive?
- What additional diagnostics or logs should I check to identify the root cause?
- Could this be related to a broader Azure networking issue, and if so, how can I verify this?
Any insights, suggestions, or similar experiences would be greatly appreciated. Thank you!