An Azure service that provides a general-purpose, serverless container platform.
There is no documented customer-facing API for querying historical Microsoft.App backend telemetry by correlation ID; because this failure stopped in provider preflight and exposed no nested error, the supported way to obtain the precise rejection is to ask Azure Support to correlate the existing ID without rerunning the deployment.
Q1–Q2: What remains retrievable
ARM preflight calls the resource provider without creating or modifying resources. If preflight fails, the deployment never starts; the error appears in the Activity Log but not deployment history.^1^ That matches the absent environment and deployment record.
Since the Activity Log, CLI validation, and direct REST validation have already returned only ValidationForResourceFailed, no additional documented ARM read operation can expand that historical response. A correlation ID is intended to let Support identify related backend operations.^2^
A new validation or what-if request would be a new operation with new telemetry, not a query of the original operation. What-if makes no resource changes, but it performs preflight checks and therefore might return the same generic provider response.^3^ Since both validation routes have already been tested, repeating them is unlikely to reveal the historical nested error.
Do not enable responseContent merely for this investigation. ARM explicitly warns that request or response debug logging can expose sensitive data retrieved during deployment operations.^4^
Q3: reference() and listKeys()
reference() and listKeys() are supported ARM resource functions. listKeys() can be used in resource properties, but:
- The target resource must exist when evaluated.
- The supplied resource ID and API version must be valid.
- The deployment identity must be authorized for the corresponding list-keys operation.
- Sensitive list output must not be placed in template outputs because outputs are retained in deployment history.^5^
Therefore, these expressions are a potential validation failure point if evaluation or authorization fails, but their presence alone does not establish the cause. Assuming the caller has the effective listKeys permission and the expressions resolve correctly, there is no evidence here that using them is invalid.
The Container Apps schema treats the Log Analytics shared key as sensitive and recommends passing it as a secure parameter.^6^ That is a security recommendation, not proof that inline listKeys() caused this failure.
Q4: Safe, non-creating prerequisite checks
The requested combination is documented: workload-profile environments support Consumption, custom VNet integration, internal access, and disabled public network access.^7^ The 2025-07-01 schema also accepts publicNetworkAccess values Enabled and Disabled.^8^
For a workload-profile environment, /24 exceeds the /27 minimum and delegation to Microsoft.App/environments is correct. Check that neither the infrastructure subnet nor relevant peered address spaces overlap these reserved ranges:
-
[IP Address]/16 -
[IP Address]/16 -
[IP Address]/16 -
[IP Address]/24 -
[IP Address]/17 -
[IP Address]/19 -
[IP Address]/19 -
[IP Address]/19^9^
Run these read-only checks:
az containerapp env workload-profile list-supported -l swedencentral
az network vnet subnet show -g <vnet-rg> \
--vnet-name <vnet> -n <subnet> \
--query "{prefix:addressPrefix,prefixes:addressPrefixes,delegations:delegations[].serviceName,nsg:networkSecurityGroup.id,routeTable:routeTable.id}"
az network vnet show -g <vnet-rg> -n <vnet> \
--query "{addressSpace:addressSpace.addressPrefixes,dnsServers:dhcpOptions.dnsServers}"
The first command verifies current regional workload-profile availability; it does not retrieve the original provider error.^10^
If custom DNS, an NSG, firewall, or route table is attached, inspect its existing rules without changing them. Consumption profiles must be able to use Azure DNS through [IP Address]/AzurePlatformDNS; required outbound dependencies include Microsoft Container Registry, Azure Front Door first-party services, and—when used—Microsoft Entra ID and Azure Monitor.^11^
These are potential causes only. None is confirmed by the generic error.
Q5: Escalation
In Azure Portal:
- Select ? → describe the issue → Create a support request.
- Choose Technical and the affected subscription/service.
- Supply the UTC timestamp, region, resource type/API version, operation
Microsoft.Resources/deployments/validate/action, correlation ID, and sanitized template hash. - Explicitly request correlation with
Microsoft.Appprovider-side preflight telemetry and state that no deployment record or resource exists. - Select No for Advanced diagnostic information if consent-based collection is not acceptable. This limits that collection request; it does not guarantee what historical service telemetry exists or whether it remains correlatable.
Creating a support request requires subscription-level Owner, Contributor, Support Request Contributor, or equivalent Microsoft.Support/* permission. Microsoft documents that technical support normally requires a support plan; free subscription-management support does not cover this provider-level technical investigation.^12^
At present, the confirmed conclusion is the diagnostic limitation—not a template, networking, Log Analytics, or regional root cause.
References
- Preflight: Server validation before deployment
- View deployment history with Azure Resource Manager
- ARM template deployment what-if operation
- DebugSetting interface-Property Details
- Resource functions for ARM templates
- Microsoft.App managedEnvironments 2025-01-01 (deployment-language-arm-template)
- Use Azure Front Door Premium with a custom virtual network and Private ...
- Microsoft.App managedEnvironments 2025-07-01 (deployment-language-terraform)
- Virtual network configuration
- az containerapp env workload-profile | Microsoft Learn
- Network security groups for configuring a virtual network in Azure Container Apps
- Create an Azure support request