Mock runs for storage task assignments

What is a mock run?

A mock run simulates the execution of a storage task assignment without actually performing any operations on your blobs. When you create a mock run, Azure Storage Actions scans and evaluates your blobs against the task conditions just as it would during a real run, but it doesn't execute any operations. Instead, it generates a detailed report that shows which blobs matched the conditions and what operations would have been performed.

Mock runs are useful when you want to:

• Preview the impact of a task before running it at scale, especially when operations are irreversible (such as delete or immutability policy)
• Validate conditions on the full set of blobs in your account, not just a small preview sample.
• Generate audit-ready reports that show which blobs would be affected, without making any changes.
• Estimate cost by understanding how many blobs would be targeted and how many operations would be invoked.

How mock runs work

Create a mock run as a storage task assignment with the trigger type set to MockRun. Like other assignment types, a mock run targets a specific storage account, uses optional prefix filters to scope the blobs, and generates execution reports in a designated export container.

When a mock run executes:

1. Azure Storage Actions enumerates all blobs that match the assignment's scope and prefix filters.
2. It evaluates each blob against the storage task's conditions.
3. For blobs that match, it records the operations that would be performed - but it doesn't execute any operations.
4. It generates a detailed report that lists the matched blobs and their simulated operations.

Mock run vs. condition preview

Both mock runs and the condition preview feature help you validate task conditions before execution, but they serve different purposes:

Use condition preview while composing your task conditions, then use a mock run for final validation before enabling a real run.

Mock run lifecycle and states

Mock runs follow the same lifecycle as run-once assignments.

Concurrency behavior

Only one run - mock or real - can execute on a storage account at a time. This concurrency model matches the model for real task runs:

• If a real run is in progress, the system queues a new mock run until the real run finishes.
• If a mock run is in progress, the system queues a new real task run or skips it for scheduled runs.
• If another mock run is in progress, the system queues the new mock run.

This model ensures stability and prevents resource contention on the target storage account.

Mock run reports

When a mock run finishes, the process writes a detailed report to the report export container that you specify during assignment creation. You can access reports in CSV formats.

You can also view a summary of the run directly in the Azure portal on the assignment's mock run results page. The summary includes the number of objects listed, objects that met the conditions, and operations that the process would have performed.

Report columns:

Example CSV output:

The process also generates a summary JSON file alongside the report. This file contains aggregate metrics:

{
  "completionTime": "2024-10-21T17:46:59",
  "destination": "taskoutput",
  "endpoint": "https://contoso1storage1.blob.core.windows.net",
  "fileFormat": "csv",
  "fileSchema": [
    "Container",
    "Blob",
    "Operation to be performed",
    "Result",
    "Matched condition block"
  ],
  "files": [
    "<link to the reporting file>"
  ],
  "objectsListed": 1100,
  "objectsToBeOperated": 240,
  "operationType": "BlobOperation",
  "runId": "mockrun-assignment-2024-10-21T17:30:13.9121342Z",
  "startTime": "2024-10-21T17:37:12",
  "status": "succeeded"
}

Key fields in the summary:

• objectsListed: Total number of blobs scanned during the mock run.
• objectsToBeOperated: Number of blobs that matched the conditions and would have had operations performed.
• status: The outcome of the mock run (succeeded or failed).

Transition from mock run to real run

After reviewing the mock run report and confirming the results are as expected, you can transition the assignment from a mock run to a real run:

1. Go to the assignment in the Azure portal.
2. Edit the assignment and change the trigger type from Mock run to Run once or Recurring.
3. Save the updated assignment.

By using this process, you can go from validation to execution without recreating the assignment from scratch.

Pricing

Mock runs are billed similarly to real task assignment runs, with one key difference: no charge is applied for the operations meter, since no operations are actually performed on blobs.

Standard Blob Storage API costs for listing and reading blob properties during the scan still apply.

Permissions

The managed identity associated with the storage task must have the appropriate role on the target storage account to perform a mock run. Although the process doesn't perform any operations, the identity needs read access to scan and evaluate blobs:

• Minimum role: Storage Blob Data Reader
• Recommended role: Storage Blob Data Owner (if you plan to transition to a real run using the same assignment)

Both system-assigned and user-assigned managed identities support mock runs.

If the target storage account has network restrictions, ensure that the Allow trusted Microsoft services option is enabled in the account's networking configuration.

See also

• Create and use a mock run
• Storage task assignment
• Storage task runs
• Best practices for storage tasks