> ## Documentation Index
> Fetch the complete documentation index at: https://docs.condense.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Azure Blob Storage (Output, Store)

**Connector Type:** Output Connector

**Description**

Stores processed data from Condense into Azure Blob containers for secure, long-term retention.

**Brief Description**

The Azure Blob Storage Store Output Connector allows seamless integration of real-time event streams from Condense into Azure Blob Storage. It supports secure, scalable, and structured streaming of data into cloud-based object storage, making it ideal for archival, batch analytics, data lake pipelines, compliance snapshots, and downstream machine learning workflows.

This connector consumes data from a Kafka-native topic and writes each message into blob files (typically JSON, NDJSON, or CSV) in the specified Azure Blob container.

**Links to Relevant Documents**

* \[[https://learn.microsoft.com/en-us/azure/storage/blobs/storage-blobs-introduction](https://learn.microsoft.com/en-us/azure/storage/blobs/storage-blobs-introduction)]
* \[[https://learn.microsoft.com/en-us/azure/storage/common/storage-auth](https://learn.microsoft.com/en-us/azure/storage/common/storage-auth)]
* \[[https://learn.microsoft.com/en-us/rest/api/storageservices/](https://learn.microsoft.com/en-us/rest/api/storageservices/)]
* \[[https://learn.microsoft.com/en-us/azure/storage/common/storage-auth-aad-sas](https://learn.microsoft.com/en-us/azure/storage/common/storage-auth-aad-sas)]
* \[[https://learn.microsoft.com/en-us/azure/storage/blobs/lifecycle-management-overview](https://learn.microsoft.com/en-us/azure/storage/blobs/lifecycle-management-overview)]

**Connector Overview**

This output connector enables event streaming into Azure Blob Storage using structured Kafka messages published in Condense. It is ideal for building:

A. Cost-effective long-term archives.

B. Time-partitioned batch analytics pipelines.

C. Cold data lakes that integrate with Azure Synapse, Azure Databricks, or Power BI.

D. Compliance snapshots of streaming data.

Data is written as individual blob objects and can be grouped by time window or custom partitioning logic defined in the Condense transform layer.

**Core Functionality**

1. Converts Condense events into blob objects in Azure Blob Storage.
2. Supports Shared Key and SAS Token authentication.
3. Supports partitioning based on time, topic, or custom attributes.
4. Supports JSON, NDJSON, CSV, and custom output formats.
5. Provides reliable delivery with write acknowledgements and logging.

**How It Works in Condense**

1. The connector subscribes to a Kafka-native topic inside Condense.
2. Streaming messages are consumed in real time.
3. The connector authenticates with Azure Blob Storage using either a Shared Key or SAS Token.
4. Messages are converted into the configured file format.
5. Blob files are created inside the configured container based on batching policies such as time window, record count, or file size.

**Configuration**

| **Category** | **Field Name**          | **Description**                    | **Required** |
| :----------- | :---------------------- | :--------------------------------- | :----------: |
| General      | Title                   | Unique connector name              |      Yes     |
| Storage      | Storage Account Name    | Azure Storage Account name         |      Yes     |
| Storage      | Storage Container Name  | Azure Blob container name          |      Yes     |
| Storage      | Storage Access Key      | Shared Key or SAS Token            |      Yes     |
| Storage      | Storage Credential Type | `shared_key` or `sas_token`        |      Yes     |
| Condense     | Topic (Input)           | Kafka-native topic to consume from |      Yes     |

**Field-by-Field Explanation**

**1. Title**

Internal reference name for the connector.

**Examples:**

* azure-archive-2025
* telemetry-backup
* audit-trail-writes

**2. Storage Account Name**

Name of the Azure Storage Account.

Do not include the protocol or `blob.core.windows.net`.

**Examples:**

* zeliotdatastore
* iotanalyticsprod

**3. Storage Container Name**

Blob container where files will be written.

Container names must be lowercase.

**Examples:**

* vehicle-trips
* archive2025
* realtime-dumps

**4. Storage Access Key**

Credential used to authenticate with Azure Blob Storage.

Supported credential types:

* Shared Key
* SAS Token

For better security, Microsoft recommends using short-lived SAS Tokens.

Credentials can be obtained from:

* Azure Portal → Storage Account → Access Keys
* Azure Portal → Shared Access Signature

**5. Storage Credential Type**

Defines how Condense authenticates with Azure Blob Storage.

Supported values:

* `shared_key`
* `sas_token`

**6. Topic (Input)**

Kafka-native Condense topic from which streaming records are consumed.

Messages should already be formatted as JSON, CSV, or another supported output format.

**Sample Configuration Example**

```json theme={null}
{
  "title": "azure-lake-writer",
  "storageAccountName": "iotanalyticsprod",
  "storageContainerName": "vehicle-trips",
  "storageAccessKey": "REDACTED",
  "storageCredentialType": "shared_key",
  "inputTopic": "trip-summary-events"
}
```

**Troubleshooting and Common Issues**

**1. Unauthorized Access (403)**

**Cause:** Invalid or expired credentials.

**Solution:** Verify the Shared Key or regenerate the SAS Token and ensure it has not expired.

**2. Blob Not Created**

**Cause:** Incorrect topic mapping or empty payload.

**Solution:** Verify the input topic and inspect the payload before publishing.

**3. Container Not Found**

**Cause:** The configured container does not exist.

**Solution:** Create the container manually or ensure the connector has permission to create it.

**4. Slow Upload Performance**

**Cause:** Network latency or inefficient batching.

**Solution:** Increase batch size where appropriate and verify connectivity to the Azure Storage endpoint.

**Advanced Considerations**

**1. File Format and Naming**

* Files can be generated using time-based or record-count batching.
* File names typically include timestamps, topic names, and partition metadata.
* Custom transforms can generate custom file names and payload structures.

**2. Storage Access Tier**

Azure Lifecycle Management can automatically move blobs between Hot, Cool, and Archive tiers after upload.

**3. Replication**

Replication (LRS, GRS, RA-GRS, etc.) is managed by the Azure Storage Account configuration and is not controlled by Condense.

**Best Practices**

1. Use SAS Tokens instead of Shared Keys whenever possible.
2. Organize containers by business domain or retention policy.
3. Integrate Azure Blob Storage with Azure Synapse, Azure Data Factory, or Databricks for downstream analytics.
4. Apply schema validation and transformations before writing data.
5. Configure Azure Lifecycle Management policies to archive or delete old data automatically.