Restoring Soft-Deleted Blobs with multithreading in Azure Storage Using C#

Blob soft delete is an essential feature that safeguards your data against accidental deletions or overwrites. By retaining deleted data for a specified period, it ensures data integrity and availability, even in the event of human error. However, restoring data in the soft delete state can be more labor-intensive, as the undelete API must be called for each individual deleted blob. Currently, there is no option to bulk undelete all blobs.

In this blog, we provide a sample C# code that will help you restore soft-deleted data efficiently. The code leverages multiple threads to expedite the restoration process, making it particularly effective if you have a large number of blobs to restore. Additionally, this program can be configured to undelete blobs within a specific container or directory, rather than scanning the entire storage account.

To run this program, follow these steps:

Install .NET SDK: Ensure you have the .NET SDK installed on your machine.
Connect to Azure Account:

Connect-AzAccount

Add NuGet Source:

dotnet nuget add source https://api.nuget.org/v3/index.json -n nuget.org

Create a New Console Application:

dotnet new console --force

Add the following code to Program.cs.

using Azure.Core; using Azure.Identity; using Azure.Storage.Files.DataLake; using Azure.Storage.Files.DataLake.Models; var StorageAccountName = "xxxx"; var ContainerName = "xxxx"; var DirectoryPath = ""; var Concurrency = 500; var BatchSize = 500; static DataLakeServiceClient GetDatalakeClient(string accountName) { DataLakeClientOptions clientOptions = new DataLakeClientOptions() { Retry = { Delay = TimeSpan.FromMilliseconds(500), MaxRetries = 5, Mode = RetryMode.Fixed, MaxDelay = TimeSpan.FromSeconds(5), NetworkTimeout = TimeSpan.FromSeconds(30) }, }; // only works for prod. DataLakeServiceClient client = new( new Uri($"https://{accountName}.blob.core.windows.net"), new DefaultAzureCredential(), clientOptions); return client; } Console.WriteLine("Starting the program"); var client = GetDatalakeClient(StorageAccountName); var throttler = new SemaphoreSlim(initialCount: Concurrency); List<Task> tasks = new List<Task>(); List<string> containerNames = new List<string>(); if (string.IsNullOrEmpty(ContainerName)) { var containers = client.GetFileSystems(); foreach (var container in containers) { containerNames.Add(container.Name); } } else { containerNames.Add(ContainerName); } var totalSuccessCount = 0; var totalFailedCount = 0; foreach (var container in containerNames) { Console.WriteLine($"Recoverying for container {container}"); var fileSystem = client.GetFileSystemClient(container); var deletedItems = fileSystem.GetDeletedPaths(pathPrefix: DirectoryPath); var count = 0; var totalSuccessCountForContainer = 0; var totalFailedCountForContainer = 0; foreach (PathDeletedItem item in deletedItems) { await throttler.WaitAsync(); count++; try { var task = (fileSystem.UndeletePathAsync(item.Path, item.DeletionId)); var continuedTask = task.ContinueWith(t => { throttler.Release(); if (t.IsFaulted) { Interlocked.Increment(ref totalFailedCount); Interlocked.Increment(ref totalFailedCountForContainer); Console.WriteLine($"Failed count for container {totalFailedCountForContainer}, total failed count {totalFailedCount}, path {DirectoryPath + item.Path} due to {t.Exception.Message}"); } else { Interlocked.Increment(ref totalSuccessCount); Interlocked.Increment(ref totalSuccessCountForContainer); Console.WriteLine($"Success count for container {totalSuccessCountForContainer}, total success count {totalSuccessCount}"); } }); tasks.Add(continuedTask); } catch (Exception ex) { Console.WriteLine("Failed to create task: " + ex.ToString()); } finally { if (count == Math.Max(Concurrency, BatchSize)) { count = 0; await Task.WhenAll(tasks); tasks.Clear(); } } } await Task.WhenAll(tasks); Console.WriteLine($"Recover finished for container {container}"); }

Replace xxxx with your specific storage account and container name. If you need to restore a particular directory, provide the directory name; otherwise, leave it empty to scan the entire container. The code is configured to run with 500 threads by default, but you can adjust this number according to your needs.

Add Required Packages:

dotnet add package Azure.Identity dotnet add package Azure.Storage.Blobs

Build the Project:

dotnet build --configuration Release

Run the Program:

dotnet <path_to_dll>

Once the application is running, you can monitor the console window to track its progress and identify any potential issues or failures.

Published on: September 03, 2024

Learn more

Azure PaaS Blog articles

Announcing the end of support for Node.js 20.x in the Azure SDK for JavaScript

After July 9, 2026, the Azure SDK for JavaScript will no longer support Node.js 20.x. Upgrade to an Active Node.js Long Term Support (LTS) ver...

2 hours ago

MCP Apps on Azure Functions: Quickstart with TypeScript

Learn how to build and deploy MCP (Model Context Protocol) apps on Azure Functions using TypeScript. This guide covers MCP tools, resources, l...

7 hours ago

Setting up Power BI Version Control with Azure Dev Ops

In this blog post is a way set up version control for Power BI semantic models (and reports) using the PBIP (Power BI Project) format, Azure D...

6 days ago

Azure Developer CLI (azd) – March 2026: Run and Debug AI Agents Locally, GitHub Copilot Integration, & Container App Jobs

Run, invoke, and monitor AI agents locally or in Microsoft Foundry with the new azd AI agent extension commands. Plus GitHub Copilot-powered p...

6 days ago

Blog image

Azure PaaS Blog articles

Learn more

Restoring Soft-Deleted Blobs with multithreading in Azure Storage Using C#

Related posts

Announcing the end of support for Node.js 20.x in the Azure SDK for JavaScript

MCP Apps on Azure Functions: Quickstart with TypeScript

Setting up Power BI Version Control with Azure Dev Ops

Azure Developer CLI (azd) – March 2026: Run and Debug AI Agents Locally, GitHub Copilot Integration, & Container App Jobs

Writing Azure service-related unit tests with Docker using Spring Cloud Azure

Azure SDK Release (March 2026)

Specifying client ID and secret when creating an Azure ACS principal via AppRegNew.aspx will be removed

Azure Firewall integration with Microsoft Sentinel and Defender XDR

Azure Developer CLI (azd): Run and test AI agents locally with azd

Microsoft Purview compliance portal: Endpoint DLP classification support for Azure RMS–protected Office documents