API Server VNet Integration unavailable

Question

We're seeing issue with AKS API Server VNet Integration. This started suddenly last week in the centralus region, although we've seen it in other regions as well.

HTTP 400 on azurerm_kubernetes_cluster.CreateOrUpdate:
{
  "code": "AKSCapacityHeavyUsage",
  "message": "API Server VNet Integration is currently unavailable in region
              centralus due to high demand and limited capacity. AKS is
              actively expanding support for this feature. Check for other
              available regions at aka.ms/AksVnetIntegration."
}
}

Is this something we need adjusted for our account or is there an estimated time when this will be restored? This is actively blocking our ability to run testing and has not been an issue for us over the past months+ so it seems like something suddenly degraded here.

Answer 1

Hello Micah Nagel,

Thank you for sharing the error details. The AKSCapacityHeavyUsage error you’re encountering indicates that API Server VNet Integration is currently unavailable in the centralus region due to high demand and limited capacity.

API Server VNet Integration is generally available across all public cloud regions, including centralus (with the exception of qatarcentral). However, in rare cases, temporary capacity constraints in a region can prevent new requests from being fulfilled. This is what you’re experiencing. This condition is transient and originates from platform-side capacity limits, not from your subscription, account, or cluster configuration.

At this time, there is no active public incident or estimated time for when capacity will become available again in centralus.

What you should know about capacity behavior

When enabling API Server VNet Integration on an existing AKS cluster, the platform checks for regional capacity at the time of the operation. This validation does not reserve capacity. If there is a delay before completing required follow-up actions (such as a restart), capacity might no longer be available, which can prevent the API server from starting.

In your case, whether this is a new cluster creation or an update operation, the error you’re seeing reflects a temporary capacity limitation rather than a configuration issue.

Recommended next steps

• Retry the az aks create or az aks update operation after some time. These capacity conditions are typically short-lived.

If possible, run your workload in a different region where API Server VNet Integration is available.

If the issue continues for more than 24 hours and is blocking progress, please reach out to us so we can investigate potential underlying platform constraints.

Important notes

• Once API Server VNet Integration is enabled, it cannot be disabled.
• Enabling this feature on an existing cluster changes the API server IP address and requires a cluster restart.
• No changes are required on your subscription or account-this issue is not caused by tenant-level configuration.

We understand this is blocking your testing. While there’s currently no ETA for capacity availability in centralus, retrying the operation or using an alternate region are the supported paths forward at this time. References:

• https://learn.microsoft.com/en-us/azure/aks/api-server-vnet-integration#convert-an-existing-aks-cluster-to-api-server-vnet-integration

Answer 2

The AKSCapacityHeavyUsage error for API Server VNet Integration in centralus is caused by regional capacity limits on the AKS platform, not by anything specific to the subscription or account configuration.

When this error appears, AKS is experiencing heavy usage in that region and cannot allocate additional capacity for clusters (or for specific features such as API Server VNet Integration). There is no per-account switch or quota that support can adjust to immediately restore availability, and there is no published ETA for when capacity will be restored in a given region.

The documented mitigations are:

1. Deploy to a different region
   Use a nearby region where API Server VNet Integration is available and not capacity-constrained. For example, if centralus is impacted, try another supported region listed under API Server VNet Integration availability.
2. Change cluster settings to use a different capacity pool
   AKS maintains different capacity reservations for different cluster types. If the failure is occurring for a specific configuration, try:
   • Creating a public cluster instead of a private one (or vice versa), as capacity may differ between them.
3. Use an Enterprise Agreement subscription if available
   When capacity is tight, non-EA subscriptions are throttled first. If there is access to an EA subscription, use it to create the AKS clusters.
4. Retry later
   Capacity is reclaimed as other customers delete or stop clusters. Periodically retrying the deployment may succeed once capacity becomes available again.

Because capacity is validated but not reserved, even existing clusters being converted to API Server VNet Integration can be blocked if the region is under heavy usage. There is no configuration change on the subscription that can bypass this; the only supported options are to adjust region/cluster settings or wait for capacity to be increased in that region.

---

References:

• Troubleshoot the AksCapacityHeavyUsage error code
• Create an Azure Kubernetes Service cluster with API Server VNet Integration