Microsoft Entra ID is Azure’s identity and access management (IAM) solution, which is used to authenticate and authorize users and services. It’s built on top of open standards such OAuth 2.0, which allows services (applications) to obtain access tokens to make requests to Azure services, including Azure Storage, Azure Service Bus, Azure Key Vault, Azure Cosmos DB, Azure Database for Postgres, Azure SQL, etc.
Applications can authenticate with Microsoft Entra ID and obtain an access token to make requests to Azure services through several methods:
If you are just getting started, it is recommended to use workload identity federation.
With Managed Identities (MI), your application can authenticate with Microsoft Entra ID and obtain an access token to make requests to Azure services. When your application is running on a supported Azure service (such as Azure VMs, Azure Container Apps, Azure Web Apps, etc), an identity for your application can be assigned at the infrastructure level. You can also setup Microsoft Entra ID to federate trust to your Dapr application identity directly by using a Federated Identity Credential. This allows you to configure access to your Microsoft resources even when not running on Microsoft infrastructure. To see how to configure Dapr to use a federated identity, see the section on Authenticating with a Federated Identity Credential. This is done through system or user assigned managed identities, or workload identity federation.
Once using managed identities, your code doesn’t have to deal with credentials, which:
While some Dapr Azure components offer alternative authentication methods, such as systems based on “shared keys” or “access tokens”, you should always try to authenticate your Dapr components using Microsoft Entra ID whenever possible. This offers many benefits, including:
It’s recommended that applications running on Azure Kubernetes Service leverage workload identity federation to automatically provide an identity to individual pods.
When using Azure Role-Based Access Control (RBAC) with supported services, permissions given to an application can be fine-tuned. For example, you can restrict access to a subset of data or make the access read-only.
Using Microsoft Entra ID provides an improved auditing experience for access. Tenant administrators can consult audit logs to track authentication requests.
While Microsoft Entra ID allows you to use MI, you still have the option to authenticate using certificates.
By default, Dapr components are configured to interact with Azure resources in the “public cloud”. If your application is deployed to another cloud, such as Azure China or Azure Government (“sovereign clouds”), you can enable that for supported components by setting the azureEnvironment
metadata property to one of the supported values:
"AzurePublicCloud"
"AzureChinaCloud"
"AzureUSGovernmentCloud"
Support for sovereign clouds is experimental.
To authenticate with Microsoft Entra ID, you will need to add the following credentials as values in the metadata for your Dapr component.
Depending on how you’ve passed credentials to your Dapr services, you have multiple metadata options.
Field | Required | Details | Example |
---|---|---|---|
azureTenantId | Y | ID of the Microsoft Entra ID tenant | "cd4b2887-304c-47e1-b4d5-65447fdd542b" |
azureClientId | Y | Client ID (application ID) | "c7dd251f-811f-4ba2-a905-acd4d3f8f08b" |
azureClientSecret | Y | Client secret (application password) | "Ecy3XG7zVZK3/vl/a2NSB+a1zXLa8RnMum/IgD0E" |
When running on Kubernetes, you can also use references to Kubernetes secrets for any or all of the values above.
Field | Required | Details | Example |
---|---|---|---|
azureTenantId | Y | ID of the Microsoft Entra ID tenant | "cd4b2887-304c-47e1-b4d5-65447fdd542b" |
azureClientId | Y | Client ID (application ID) | "c7dd251f-811f-4ba2-a905-acd4d3f8f08b" |
azureCertificate | One of azureCertificate and azureCertificateFile | Certificate and private key (in PFX/PKCS#12 format) | "-----BEGIN PRIVATE KEY-----\n MIIEvgI... \n -----END PRIVATE KEY----- \n -----BEGIN CERTIFICATE----- \n MIICoTC... \n -----END CERTIFICATE----- |
azureCertificateFile | One of azureCertificate and azureCertificateFile | Path to the PFX/PKCS#12 file containing the certificate and private key | "/path/to/file.pem" |
azureCertificatePassword | N | Password for the certificate if encrypted | "password" |
When running on Kubernetes, you can also use references to Kubernetes secrets for any or all of the values above.
Field | Required | Details | Example |
---|---|---|---|
azureClientId | N | Client ID (application ID) | "c7dd251f-811f-4ba2-a905-acd4d3f8f08b" |
Using Managed Identities, the azureClientId
field is generally recommended. The field is optional when using a system-assigned identity, but may be required when using user-assigned identities.
When running on Azure Kubernetes Service (AKS), you can authenticate components using Workload Identity. Refer to the Azure AKS documentation on enabling Workload Identity for your Kubernetes resources.
You can use a Federated Identity Credential in Microsoft Entra ID to federate trust directly to your Dapr installation regardless of where it is running. This allows you to easily configure access rules against your Dapr application’s SPIFFE ID consistently across different clouds.
In order to federate trust, you must be running Dapr Sentry with JWT issuing and OIDC discovery enabled. These can be configured using the following Dapr Sentry helm values:
jwt:
# Enable JWT token issuance by Sentry
enabled: true
# Issuer value for JWT tokens
issuer: "<your-issuer-domain>"
oidc:
enabled: true
server:
# Port for the OIDC HTTP server
port: 9080
tls:
# Enable TLS for the OIDC HTTP server
enabled: true
# TLS certificate file for the OIDC HTTP server
certFile: "<path-to-tls-cert.pem>"
# TLS certificate file for the OIDC HTTP server
keyFile: "<path-to-tls-key.pem>"
issuer
value must match exactly the value you provide when creating the Federated Identity Credential in Microsoft Entra ID.Providing these settings exposes the following endpoints on your Dapr Sentry installation on the provided OIDC HTTP port:
/.well-known/openid-configuration
/jwks.json
You also need to provide the Dapr runtime configuration to request a JWT token with the Azure audience api://AzureADTokenExchange
.
When running in standalone mode, this can be provided using the flag --sentry-request-jwt-audiences=api://AzureADTokenExchange
.
When running in Kubernetes, this can be provided by decorating the application Kubernetes manifest with the annotations "dapr.io/sentry-request-jwt-audiences": "api://AzureADTokenExchange"
.
This ensures Sentry service issues a JWT token with the correct audience, which is required for Microsoft Entra ID to validate the token.
In order for Microsoft Entra ID to be able to access the OIDC endpoints, you must expose them on a public address. You must ensure that the domain that you are serving these endpoints via is the same as the issuer you provided when configuration Dapr Sentry.
You can now create your federated credential in Microsoft Entra ID.
cat > creds.json <<EOF
{
"name": "DaprAppIDSpiffe",
"issuer": "https://<your-issuer-domain>",
"subject": spiffe://public/ns/<dapr-app-id-namespace>/<dapr-app-id>",
"audiences": ["api://AzureADTokenExchange"],
"description": "Credential for Dapr App ID"
}
EOF
export APP_ID=$(az ad app create --display-name my-dapr-app --enable-access-token-issuance --enable-id-token-issuance | jq .id)
az ad sp create --id $APP_ID
az ad app federated-credential create --id $APP_ID --parameters ./creds.json
Now that you have a federated credential for your Microsoft Entra ID Application Registration, you can assign the desired roles to it’s service principal.
An example of assigning “Storage Blob Data Owner” role is below.
az role assignment create --assignee-object-id $APP_ID --assignee-principal-type ServicePrincipal --role "Storage Blob Data Owner" --scope "/subscriptions/$SUBSCRIPTION/resourceGroups/$GROUP/providers/Microsoft.Storage/storageAccounts/$ACCOUNT_NAME"
To configure a Dapr Component to access an Azure resource using the federated credentail, you first need to fetch your clientId
and tenantId
:
CLIENT_ID=$(az ad app show --id $APP_ID --query appId --output tsv)
TENANT_ID=$(az account show --query tenantId --output tsv)
Then you can create your Azure Dapr Component and simply provide these value:
apiVersion: dapr.io/v1alpha1
kind: Component
metadata:
name: azureblob
spec:
type: state.azure.blobstorage
version: v2
initTimeout: 10s # Increase the init timeout to allow enough time for Azure to perform the token exchange
metadata:
- name: clientId
value: $CLIENT_ID
- name: tenantId
value: $TENANT_ID
- name: accountName
value: $ACCOUNT_NAME
- name: containerName
value: $CONTAINER_NAME
The Dapr runtime uses these details to authenticate with Microsoft Entra ID, using the Dapr Sentry issued JWT token to exchange for an access token to access the Azure resource.
Important: This authentication method is recommended for development only.
This authentication method can be useful while developing on a local machine. You will need:
az login
commandWhen Dapr is running on a host where there are credentials available for the Azure CLI, components can use those to authenticate automatically if no other authentication method is configuration.
Using this authentication method does not require setting any metadata option.
In this example, you will set up an Azure Key Vault secret store component that uses Microsoft Entra ID to authenticate.
To use a client secret, create a file called azurekeyvault.yaml
in the components directory, filling in with the details from the above setup process:
apiVersion: dapr.io/v1alpha1
kind: Component
metadata:
name: azurekeyvault
namespace: default
spec:
type: secretstores.azure.keyvault
version: v1
metadata:
- name: vaultName
value: "[your_keyvault_name]"
- name: azureTenantId
value: "[your_tenant_id]"
- name: azureClientId
value: "[your_client_id]"
- name: azureClientSecret
value : "[your_client_secret]"
If you want to use a certificate saved on the local disk, instead, use:
apiVersion: dapr.io/v1alpha1
kind: Component
metadata:
name: azurekeyvault
namespace: default
spec:
type: secretstores.azure.keyvault
version: v1
metadata:
- name: vaultName
value: "[your_keyvault_name]"
- name: azureTenantId
value: "[your_tenant_id]"
- name: azureClientId
value: "[your_client_id]"
- name: azureCertificateFile
value : "[pfx_certificate_file_fully_qualified_local_path]"
In Kubernetes, you store the client secret or the certificate into the Kubernetes Secret Store and then refer to those in the YAML file.
To use a client secret:
Create a Kubernetes secret using the following command:
kubectl create secret generic [your_k8s_secret_name] --from-literal=[your_k8s_secret_key]=[your_client_secret]
[your_client_secret]
is the application’s client secret as generated above[your_k8s_secret_name]
is secret name in the Kubernetes secret store[your_k8s_secret_key]
is secret key in the Kubernetes secret storeCreate an azurekeyvault.yaml
component file.
The component yaml refers to the Kubernetes secretstore using auth
property and secretKeyRef
refers to the client secret stored in the Kubernetes secret store.
apiVersion: dapr.io/v1alpha1
kind: Component
metadata:
name: azurekeyvault
namespace: default
spec:
type: secretstores.azure.keyvault
version: v1
metadata:
- name: vaultName
value: "[your_keyvault_name]"
- name: azureTenantId
value: "[your_tenant_id]"
- name: azureClientId
value: "[your_client_id]"
- name: azureClientSecret
secretKeyRef:
name: "[your_k8s_secret_name]"
key: "[your_k8s_secret_key]"
auth:
secretStore: kubernetes
Apply the azurekeyvault.yaml
component:
kubectl apply -f azurekeyvault.yaml
To use a certificate:
Create a Kubernetes secret using the following command:
kubectl create secret generic [your_k8s_secret_name] --from-file=[your_k8s_secret_key]=[pfx_certificate_file_fully_qualified_local_path]
[pfx_certificate_file_fully_qualified_local_path]
is the path to the PFX file you obtained earlier[your_k8s_secret_name]
is secret name in the Kubernetes secret store[your_k8s_secret_key]
is secret key in the Kubernetes secret storeCreate an azurekeyvault.yaml
component file.
The component yaml refers to the Kubernetes secretstore using auth
property and secretKeyRef
refers to the certificate stored in the Kubernetes secret store.
apiVersion: dapr.io/v1alpha1
kind: Component
metadata:
name: azurekeyvault
namespace: default
spec:
type: secretstores.azure.keyvault
version: v1
metadata:
- name: vaultName
value: "[your_keyvault_name]"
- name: azureTenantId
value: "[your_tenant_id]"
- name: azureClientId
value: "[your_client_id]"
- name: azureCertificate
secretKeyRef:
name: "[your_k8s_secret_name]"
key: "[your_k8s_secret_key]"
auth:
secretStore: kubernetes
Apply the azurekeyvault.yaml
component:
kubectl apply -f azurekeyvault.yaml