Public Preview: Zone Support for Standby Pools in Azure Container Instances
May 15, 2025First Fridays June – An insider’s view: Global Executive & Employee Communications at Microsoft
May 15, 2025We’re excited to share a new enhancement to standby pools in both Azure Virtual Machine Scale Sets (VMSS) and Azure Container Instances (ACI): the introduction of a Degraded Mode health state. This feature provides deeper visibility into the operational health of your standby pools and enables proactive monitoring and troubleshooting.
What Is Degraded Mode?
Standby pools are designed to reduce scale-out latency by pre-provisioning compute capacity. With the addition of health state monitoring, you can now track the real-time status of your pool using the runtime view API, which reports on:
- The number of instances in each state (e.g., Running, Creating, Deallocated)
- Provisioning status
- Overall health classification: Healthy, Degraded, or Failed
Health State Categories
State | Description |
---|---|
Healthy | All instances are in the desired state and ready to serve requests. |
Degraded | The pool is experiencing issues provisioning instances. Common causes include quota limits, permission errors, or extension failures. |
Failed | A critical issue prevents the pool from functioning. Intervention is required. |
How Degraded Mode Works
When a standby pool enters the Degraded state:
- Instance creation is paused for 30 seconds to allow for issue resolution.
- After the pause, the system automatically retries provisioning.
- Health state transitions are tracked and exposed via the runtime view API.
This behavior helps prevent runaway retries and gives you time to address underlying issues such as:
- Quota exhaustion
- Misconfigured permissions
- Extension or image failures
Here’s a simplified example of the API response:
{ “properties”: {
“instanceCountSummary”: [ {
“instanceCountsByState”: [
{“state”: “creating”, “count”: 5 },
{ “state”: “running”, “count”: 5 },
{ “state”: “deleting”, “count”: 10 }
]
}
],
“provisioningState”: “Succeeded”, “status”: {
“code”: “HealthState/degraded”,
“message”: “The pool is degraded.”
}
},
“name”: “myStandbyPool”,
“type”: “Microsoft.StandbyPool/standbyVirtualMachinePools/runtimeViews”
}
Recommended Actions
If your pool enters a degraded state:
- Monitor for prolonged “creating” or “deleting” states.
- Use Azure Monitor or Log Analytics to investigate quota, permission, or provisioning issues.
- Review extension configurations and ensure all required roles and limits are properly set.
Why It Matters
This update gives you the tools to:
- Detect and respond to provisioning issues faster
- Maintain high availability and performance
- Reduce operational guesswork with clear health signals