Azure Event Hub Connector

Overview

The Azure Event Hub connector streams event data from Azure Event Hubs into Edge Delta Pipelines. It collects application logs, telemetry, and events from Azure services and makes it available for AI teammates to query through the Edge Delta MCP connector.

When you add this streaming connector, it appears as a Azure Event Hub source in your selected pipeline. AI teammates access this data by querying the Edge Delta backend with the Edge Delta MCP connector.

Add the Azure Event Hub Connector

To add the Azure Event Hub connector, you select a pipeline to receive the event data and configure Azure Event Hub access settings.

Prerequisites

Before configuring the connector, ensure you have:

• Azure Event Hub namespace with an Event Hub created
• Connection string with Listen permission (obtain from Event Hub Shared Access Policies)
• Azure Storage account for checkpoint management
• Azure Storage account access key
• Consumer group name (default is $Default)

Configuration Steps

1. Navigate to AI Team > Connectors in the Edge Delta application
2. Find the Azure Event Hub connector in Streaming Connectors
3. Click the connector card
4. Select the pipeline (environment) to receive this data
5. Configure Event Hub and Storage options (see below)
6. Click Save

The connector is now streaming Event Hub data into your pipeline.

Configuration Options

Connector Name

Name to identify this Azure Event Hub connector instance.

Connection String

Connection string to the Azure Event Hub with Listen permission. The connection string contains the endpoint, shared access key name, shared access key, and entity path needed to connect to your Event Hub.

To retrieve the connection string from Azure Portal:

1. Navigate to your Event Hubs Namespace
2. Select Shared access policies under Settings
3. Choose a policy with Listen claims (or create a new one)
4. Copy the Connection string–primary key value

For detailed instructions, see Get an Event Hubs connection string.

Format: Endpoint=sb://namespace.servicebus.windows.net/;SharedAccessKeyName=PolicyName;SharedAccessKey=KEY;EntityPath=eventhubname

Consumer Group

Consumer group to use for the Azure Event Hub. Each consumer group maintains its own offsets independently.

Default: $Default

Storage Account Name

Azure Storage account name used for Event Processor Host checkpointing and lease management. The connector uses this to track progress and coordinate partition consumption.

Format: Storage account name only (e.g., mycompanystorage), not the full URL

Storage Account Key

Access key for the Azure Storage account. Retrieve from Azure Portal: Storage Account → Access Keys → Copy key1 or key2.

Storage Container Name

Blob container name for storing checkpoint and lease data. The container is automatically created if it doesn’t exist.

Format: Lowercase letters, numbers, and hyphens (3-63 characters)

Partition IDs

Specific Event Hub partition IDs to consume. Leave empty to automatically balance all partitions across consumers (Event Processor Host mode).

Default: Empty (automatic partition balancing)

Format: Comma-separated partition numbers (e.g., 0,1,2) or empty

Checkpoint Directory

Local filesystem path for persisting checkpoint data as backup to Azure Storage.

Default: /var/lib/edgedelta/checkpoints

Metadata Level (Resource Attributes)

Resource attributes attached to ingested events for filtering and grouping.

Default attributes: host.name, host.ip, service.name, ed.source.name, ed.source.type

Metadata Level (Attributes)

Additional attributes to attach to events, such as ed.env.id for environment identification.

Rate Limit

Rate limit configuration to control maximum data ingestion volume and manage processing capacity.

Target Environments

Select the Edge Delta pipeline (environment) where you want to deploy this connector.

How to Use the Azure Event Hub Connector

The Azure Event Hub connector integrates seamlessly with AI Team, enabling analysis of streaming event data from Azure services. AI teammates automatically leverage the ingested data based on the queries they receive and the context of the conversation.

Use Case: Azure Application Diagnostics

When Azure App Services or Functions stream diagnostic logs to Event Hub, AI teammates can analyze application errors, performance issues, and usage patterns. For example, when investigating a production incident, teammates can correlate application exceptions from Event Hub with infrastructure metrics to identify root causes.

Use Case: IoT Device Monitoring

Azure IoT Hub routes device telemetry to Event Hub for downstream processing. AI teammates can analyze device health patterns, identify failing sensors, and detect anomalies across device populations. This is valuable when combined with alert event connectors like PagerDuty—teammates can investigate IoT alerts by querying recent telemetry from affected devices.

Use Case: Azure Resource Activity Analysis

Azure Monitor diagnostic settings export resource activity logs to Event Hub. AI teammates can track configuration changes, analyze resource health trends, and correlate operational events across your Azure infrastructure. When integrated with ticketing systems like Jira, teammates can automatically document infrastructure changes and their impacts.

Troubleshooting

Connection errors: Verify the connection string is complete and correctly formatted, including endpoint, shared access key name, key, and entity path. Ensure the Event Hub exists and is not paused.

Authentication failures: Confirm the shared access policy has Listen permission on the Event Hub. Verify the storage account key is valid and hasn’t been rotated.

Checkpoint errors: Ensure the storage container name follows Azure naming requirements (3-63 lowercase characters, numbers, hyphens). Verify the storage account key provides write access.

No events received: Check that events are being sent to the Event Hub (verify in Azure Portal metrics). Confirm the consumer group name matches exactly (case-sensitive).

High partition lag: Increase Edge Delta consumer instances to distribute partition processing load. Optimize downstream pipeline processing to reduce per-event processing time.

Next Steps

• Learn about creating custom teammates that can use Event Hub data
• Explore the Edge Delta MCP connector for querying event data

For additional help, visit AI Team Support.