VMSS stuck in "0 to 1" state for hours, no allocation, no errors

Hello Vassily

Thank you for contacting us about the VMSS issue.

When a Virtual Machine Scale Set (VMSS) is stuck scaling from 0 → 1 for several hours with no allocation and no visible errors, this typically indicates a backend allocation or control‑plane provisioning stall, not a quota or configuration issue.

Actions to perform:

1. Scale down to 0, then scale back to 1
   • This forces Azure to re‑evaluate capacity and often clears a stuck allocation.
2. Change the VM size
   • Select a similar VM size (same family if possible) and retry scaling.
   • This is one of the most reliable ways to unblock allocation when capacity is constrained.
3. Change the availability zone
   • Move from the current zone to another zone and retry scale‑out.
4. Recreate the VMSS (for new deployments)
   • If the scale set is newly created and no workloads depend on it yet, recreating it with a different size or zone is often faster than waiting for a stuck allocation.
5. Recommended solution is to redeploy the VMSS it will work.
6. “The VMSS is stuck at 0 instances because backend allocation failed due to insufficient approved regional vCPU quota, even though portal usage shows zero.”
7. Over the call user has changed the Region from west us 3 to East US 2 which fixed the issue.

https://learn.microsoft.com/en-us/troubleshoot/azure/virtual-machines/windows/allocation-failure