Troubleshoot resource allotment failures when you create, restart, or resize VMs in Azure

  • Article
  • 7 minutes to read

When you lot create a virtual machine (VM), start stopped (deallocated) VMs, or resize a VM, Microsoft Azure allocates compute resources to your subscription. We are continually investing in additional infrastructure and features to make sure that nosotros always have all VM types available to support client demand. However, yous may occasionally feel resource allocation failures because of unprecedented growth in demand for Azure services in specific regions. This problem can occur when y'all try to create or start VMs in a region while the VMs brandish the following mistake code and message:

Error lawmaking: AllocationFailed or ZonalAllocationFailed

Fault message: "Allocation failed. Nosotros do not accept sufficient capacity for the requested VM size in this region. Read more than nigh improving likelihood of resource allotment success at https://aka.ms/allocation-guidance"

Note

If you are troubleshooting a virtual machine calibration set (VMSS), the process is the same equally a standard VM. To resolve the issue, you should follow the directions in this commodity.

Error message: "Allotment failed. If yous are trying to add a new VM to a Virtual Car Scale Ready with a unmarried placement grouping or update/resize an existing VM in a Virtual Motorcar Scale Set with a single placement grouping, please note that such allotment is scoped to a single cluster, and it is possible that the cluster is out of capacity. Please read more about improving likelihood of allotment success at http://aka.ms/allocation-guidance."

This article explains the causes of some of the common allocation failures and suggests possible remedies.

If your Azure consequence is not addressed in this article, visit the Azure forums on MSDN and Stack Overflow. You tin can post your consequence on these forums or to @AzureSupport on Twitter. Also, you can file an Azure back up request by selecting Get support on the Azure back up site.

Until your preferred VM type is available in your preferred region, nosotros advise customers who encounter deployment problems to consider the guidance in the following tabular array equally a temporary workaround.

Identify the scenario that best matches your instance, and then retry the allocation asking by using the respective suggested workaround to increase the likelihood of allocation success. Alternatively, yous can e'er retry later. This is because enough resources may have been freed in the cluster, region, or zone to accommodate your asking.

Resize a VM or add VMs to an existing availability set

Cause

A asking to resize a VM or add a VM to an existing availability set must be tried at the original cluster that hosts the existing availability gear up. The requested VM size is supported by the cluster, but the cluster may non currently take sufficient capacity.

Workaround

If the VM can be part of a different availability set, create a VM in a unlike availability set (in the same region). This new VM can then be added to the same virtual network.

Stop (deallocate) all VMs in the same availability set, and so starting time all applicable VMs in batch. To stop: Click Resources groups > [your resource grouping] > Resources > [your availability fix] > Virtual Machines > [your virtual automobile] > Cease. After all VMs stop, navigate to Home > Virtual machines, add an Availability set filter so select Group by availability set. Select all applicable VMs and click Commencement. This step makes sure that a new allocation attempt is run and that a new cluster can be selected that has sufficient capacity.

Beginning partially stopped (deallocated) VMs

Cause

Partial deallocation ways that you stopped (deallocated) i or more, but not all, VMs in an availability set. When you deallocate a VM, the associated resources are released. Starting VMs in a partially deallocated availability set is the same as calculation VMs to an existing availability prepare. Therefore, the allotment request must be tried at the original cluster that hosts the existing availability set that may not have sufficient chapters.

Workaround

End (deallocate) all VMs in the same availability fix, then beginning all applicable VMs in batch. To stop: Click Resources groups > [your resource grouping] > Resources > [your availability prepare] > Virtual Machines > [your virtual machine] > Terminate. After all VMs finish, navigate to Dwelling house > Virtual machines, add together an Availability prepare filter and and then select Group by availability set. Select all applicable VMs and click Start. This will brand sure that a new allotment endeavour is run and that a new cluster can be selected that has sufficient capacity.

Start fully stopped (deallocated) VMs

Crusade

Full deallocation means that you stopped (deallocated) all VMs in an availability set up. The resource allotment request to beginning these VMs will target all clusters that support the desired size within the region or zone. Alter your allocation asking per the suggestions in this article, and retry the asking to improve the take chances of allocation success.

Workaround

If you use older VM series or sizes, such as Dv1, DSv1, Av1, D15v2, or DS15v2, consider moving to newer versions. See these recommendations for specific VM sizes. If you don't have the choice to employ a different VM size, endeavour deploying to a unlike region within the same geo. For more than information about the available VM sizes in each region at https://aka.ms/azure-regions

If you lot are using availability zones, endeavour another zone inside the region that may have bachelor capacity for the requested VM size.

If your allocation asking is large (more than than 500 cores), see the guidance in the following sections to break upwardly the asking into smaller deployments.

Allocation failures for older VM sizes (Av1, Dv1, DSv1, D15v2, DS15v2, etc.)

As nosotros expand Azure infrastructure, we deploy newer-generation hardware that's designed to support the latest virtual automobile types. Some of the older series VMs do not run on our latest generation infrastructure. For this reason, customers may occasionally feel resource allotment failures for these legacy SKUs. To avoid this problem, we encourage customers who are using legacy series virtual machines to consider moving to the equivalent newer VMs per the following recommendations: These VMs are optimized for the latest hardware and will let you have reward of better pricing and functioning.

Legacy VM-series/size Recommended newer VM-serial/size More information
Av1-serial Av2-serial https://azure.microsoft.com/blog/new-av2-series-vm-sizes/
Dv1 or DSv1-series (D1 to D5) Dv3 or DSv3-series https://azure.microsoft.com/blog/introducing-the-new-dv3-and-ev3-vm-sizes/
Dv1 or DSv1-series (D11 to D14) Ev3 or ESv3-series
D15v2 or DS15v2 If you are using theResource Manager deployment model in social club to accept advantage of the larger VM sizes, consider moving to D16v3/DS16v3 or D32v3/DS32v3. These are designed to run on the latest generation hardware. If you are using the Resource Manager deployment model to make sure your VM case is isolated to hardware dedicated to a single client, consider moving to the new isolated VM sizes, E64i_v3 or E64is_v3, which are designed to run on the latest generation hardware. https://azure.microsoft.com/blog/new-isolated-vm-sizes-at present-available/

Allotment failures for big deployments (more 500 cores)

Reduce the number of instances of the requested VM size, and then retry the deployment performance. Additionally, for larger deployments, you may desire to evaluate Azure virtual machine scale sets with multiple placement groups. The number of VM instances can automatically increase or subtract in response to demand or a defined schedule, and you take a greater chance of resource allotment success considering the deployments tin can exist spread across multiple clusters when deployed as a multi-placement group. Learn more about working with large virtual machine calibration sets and how to convert an existing scale fix to span multiple placement groups. Note that y'all tin can alter a calibration set from supporting a single placement group to supporting multiple placement groups, but you cannot perform a conversion in the other direction.

Groundwork information

How allocation works

The servers in Azure datacenters are partitioned into clusters. Usually, an allocation request is attempted in multiple clusters, but it's possible that certain constraints from the allocation request strength the Azure platform to try the request in only 1 cluster. In this article, we'll refer to this every bit "pinned to a cluster." Diagram 1 below illustrates the instance of a normal allotment that is attempted in multiple clusters. Diagram 2 illustrates the case of an resource allotment that's pinned to Cluster 2 because that'south where the existing Cloud Service CS_1 or availability set is hosted. Diagram 1 shows allocation attempted in multiple clusters and Diagram 2 shows allocation pinned to one cluster.

Why allotment failures happen

When an allocation asking is pinned to a cluster, there's a college chance of failing to observe free resources since the bachelor resource pool is smaller. Furthermore, if your allotment request is pinned to a cluster but the blazon of resource you requested is non supported past that cluster, your request will neglect fifty-fifty if the cluster has gratuitous resource. The post-obit Diagram three illustrates the instance where a pinned resource allotment fails because the merely candidate cluster does not have free resource. Diagram four illustrates the case where a pinned allocation fails because the only candidate cluster does non support the requested VM size, even though the cluster has costless resources.

Diagram 3 shows allocation failed at pinned cluster: No free resource available and Diagram 4 shows allocation failed at pinned cluster: Size not supported.