DataFrugal™

What is DataFrugal?

DataFrugal is a cloud backup and data control platform designed for engineering, infrastructure, and operations teams.

It provides a structured, policy-driven control plane for protecting, retaining, and accessing data stored on secure object storage backends — without exposing raw storage SDKs or requiring teams to manage low-level cloud primitives.

DataFrugal can be used independently or integrated into existing application and system workflows through APIs, the df-agent, and a centralized web interface.

Common use cases include Database backups, VM backups, log archival, compliance snapshots, and long-term data retention. Across these scenarios, DataFrugal provides:

• Usage-aligned pricing based on protected data volume, not per-user licenses or individual API calls
• Centralized visibility and control using metadata-based insights, without inspecting or processing data content
• Policy-driven versioning and retention for reliable recovery and governance
• Enterprise-grade security controls including encryption at rest and in transit, role-based access, and network restrictions
• Large object support for handling individual objects up to 5 TB through platform-managed transfers

With a role-aware UI for operational visibility and REST APIs plus command-line tooling for automation, DataFrugal enables teams to manage backup and retention workflows efficiently — without turning storage management into an application concern.

Key Features & Benefits

• Unified Visibility Across Protected Data: Centralized, metadata-based visibility into protected data across systems, services, and automated workflows — without inspecting data content.
• Usage-Aligned Pricing Model: Pricing scales with the volume of data protected rather than per-user licenses or individual API calls.
• Enterprise-Grade Security Controls: Encryption at rest and in transit, role-based access control, organization- scoped boundaries, and network-level restrictions.
• Role-Aware Administrative Control: Centralized management of users, roles, ownership, and usage visibility through a single policy-driven control plane.
• Secure, Controlled Access Across Environments: Browser-based UI for operational visibility combined with APIs and df-agent support for automation and system-level workflows.
• Policy-Driven Versioning and Retention: Automatic retention of historical object versions based on defined policies, enabling reliable recovery and governance.
• Metadata-Based Search and Discovery: Search and organize protected data using indexed metadata, labels, and source context — without content inspection.
• Built-In Data Protection and Access Controls: Preventive controls that restrict external sharing and reduce exposure through architectural enforcement rather than reactive monitoring.
• Large Object Support: Reliable handling of individual objects up to 5 TB through platform-managed transfers, without exposing underlying storage SDKs.

What is WORM (Write Once, Read Many)?

WORM (Write Once, Read Many) is an immutable storage technology that allows data to be written once and read any number of times without being modified or deleted until the configured retention period expires.

Once WORM protection is applied, DataFrugal prevents any changes to the protected object for the entire retention duration, ensuring the original data remains authentic, tamper-proof, and recoverable.

WORM is commonly used for:

• Backup protection
• Long-term archival
• Regulatory compliance
• Financial records
• Healthcare data
• Legal evidence
• Business-critical documents

How WORM Works in DataFrugal

When uploading a file, users can enable WORM Retention directly from the DataFrugal Dashboard by selecting a retention period before completing the upload.

For automated workflows, the same protection can be applied using the DataFrugal Agent (df-agent) during file upload.

Once uploaded with WORM enabled:

• The object becomes immutable.
• It cannot be modified.
• Protection remains active until the retention period expires.

These rules are enforced automatically by DataFrugal, regardless of how the object is accessed.

Key Benefits of WORM

• Protects data from accidental deletion
• Prevents unauthorized modifications
• Preserves original versions of important files
• Creates tamper-proof audit evidence
• Improves disaster recovery readiness
• Helps meet regulatory and compliance requirements
• Ensures long-term integrity of archived data

Understanding Retention Policies

A Retention Policy defines how long an object must remain protected.

While retention is active:

• Files cannot be deleted.
• Files cannot be modified.
• Protected versions remain immutable.
• Every access follows organizational security policies.

After the retention period expires, the object can be managed according to your organization's standard lifecycle or deletion policies.

What is Ransomware?

Ransomware is malicious software that encrypts, corrupts, or deletes an organization's data to prevent access until a ransom is paid.

Modern ransomware attacks frequently target:

• Production data
• Backup repositories
• Snapshots
• Archive storage

If attackers successfully delete or encrypt backups, organizations may lose their ability to recover critical information.

How WORM Protects Against Ransomware

DataFrugal WORM provides an additional layer of protection by making protected data immutable.

Even if an attacker gains access to user credentials or administrative systems, WORM-protected objects cannot be changed or deleted before retention expires.

Benefits against ransomware include:

• Immutable backup copies remain intact.
• Clean recovery points are always available.
• Backup deletion attempts are automatically blocked.
• Insider threats have limited impact.
• Compromised accounts cannot remove protected data.

What is Break Glass?

Break Glass is DataFrugal's secure, approval-based exception process for removing WORM retention from a protected object before its retention period expires.

Normally, WORM protection cannot be bypassed.

However, approved business requirements may occasionally require retention to be removed before it expires. Instead of allowing direct deletion, DataFrugal requires an authorized approval process that is fully documented and audited.

Break Glass ensures exceptional requests are carefully controlled without compromising overall data protection.

Why is Break Glass Needed?

Strong security should never prevent valid business operations.

Break Glass provides a controlled mechanism for exceptional situations while preserving governance and accountability.

It helps organizations:

• Prevent unauthorized retention overrides
• Ensure administrative review before removing protection
• Maintain complete audit records
• Reduce insider risk
• Meet compliance requirements
• Preserve trust in immutable storage

When Should Break Glass Be Used?

Break Glass should only be used in exceptional business situations, including:

• Sensitive information uploaded accidentally
• Incorrect files requiring approved removal
• Legal or regulatory instructions requiring early deletion
• Administrative cleanup approved by authorized personnel
• Business exceptions documented through organizational policy

Break Glass Workflow

When someone attempts to delete a WORM-protected object before its retention period expires, DataFrugal automatically blocks the operation and starts the governed Break Glass process.

Step 1 — Delete Request

A user attempts to delete a protected object.

Since WORM retention is active:

• DataFrugal blocks the deletion.
• The user is informed that the object is under retention.
• The user may submit a Break Glass request.

The request includes:

• Business justification
• Target file or object version
• Reason for removal
• Two designated System Administrator approvers

Step 2 — Administrative Approval

The selected System Administrators receive the Break Glass request.

Each administrator independently reviews:

• Business justification
• Object information
• Compliance considerations
• Security impact

Possible outcomes:

• Approve
• Reject

Both administrators must approve before retention can be removed.

If either administrator rejects the request:

• The request is closed.
• WORM protection remains active.
• No changes are made.

Step 3 — Retention Update

After receiving dual approval:

• DataFrugal removes WORM retention only for the approved object version.
• All other protected objects remain unchanged.
• The approved object becomes eligible for deletion if no additional retention rules apply.

This ensures the approval affects only the requested resource and never the entire storage environment.

Break Glass at a Glance

• Secure approval-based retention override
• Two System Administrator approvals required
• Object version-specific protection removal
• No impact on other protected data
• Complete audit trail
• Integrated with WORM governance
• Role-based access control
• Enterprise-grade security and compliance

Uploading Data

Data can be uploaded securely using the web interface or the command-line agent. Uploads are encrypted in transit and handled using resumable, chunked transfers where applicable.

1. Upload via Web Interface

Use the DataFrugal dashboard to upload files or folders using drag-and-drop or file selection. Upload progress and errors are clearly surfaced in the interface.

2. Upload via Command-Line Agent

The df-agent enables uploads from the terminal and is suitable for automation, batch transfers, and scheduled jobs.

3. Upload Failure Handling

Failed uploads are reported with descriptive error messages. Partial uploads can be retried without restarting the transfer, depending on the operation and file size.

Downloading Data

Data can be downloaded through the UI, the command-line agent, or generated cURL commands, depending on access permissions.

1. Download via Web Interface

Users can download permitted files or folders directly from the dashboard. Version awareness and integrity checks are applied automatically.

2. Download via Command-Line Agent

Infrastructure and DevOps teams can download data using the df-agent for automation and system-level workflows.

3. Download via cURL

Secure, time-bound cURL commands can be generated from the UI to allow direct file downloads from a terminal. Each command includes a pre-validated, short-lived access token and is intended for one-time, user-initiated execution rather than automated or unattended workflows.

Controlled Data Access

DataFrugal supports controlled access delegation within an organization, allowing users or services to access specific folders based on assigned permissions.

1. Access Delegation via UI

Administrators and authorized users can grant access to folders through the dashboard. Access is scoped and auditable.

2. Access Delegation via CLI

The command-line agent can be used to grant or consume scoped access programmatically, enabling system-to-system workflows.

Common cURL Actions

DataFrugal provides pre-generated cURL templates for common operations such as uploads, downloads, and authenticated access. These commands can be copied directly from the UI for scripting and automation.

df-agent (Command-Line Agent)

df-agent is a Linux-compatible CLI tool that interacts with DataFrugal using authenticated APIs. It is designed for automation, system-to-system workflows, and scheduled backup operations.

• Upload, download, delete, and share files and folders — including full directory trees
• Create folders, list dashboard resources, replace versions, and apply WORM retention
• Enables secure service-to-service access with scoped read or write permissions
• Designed for CI/CD pipelines, scheduled jobs, and infrastructure automation
• Authenticates via HTTPS using Service Identity credentials (-sid / -secret) or interactive user login
• Built-in help available via ./df-agent --help

Installation

Build and run the agent from source using the following prerequisites:

• Java 8 or later
• Maven
• df-core-api SDK

Command Examples

Dashboard

Lists dashboard resources.

./df-agent -sid <service-id> -secret '<secret>' -fd

Upload File

Uploads a single file to the specified folder.

./df-agent -sid <service-id> -secret '<secret>' -up -f file.pdf -l <folder-id>

Create Folder

Creates a new folder under the specified parent location.

./df-agent -sid <service-id> -secret '<secret>' -up -folder -f "Images" -l <parent-folder-id>

Upload Directory

Uploads an entire directory while preserving its structure.

./df-agent -sid <service-id> -secret '<secret>' -up -f ./Documents -l <folder-id>

Upload with WORM Retention

Uploads a file and applies WORM retention.

./df-agent -sid <service-id> -secret '<secret>' -up -f report.pdf -crt 7y

Download

Downloads a resource using its ID.

./df-agent -sid <service-id> -secret '<secret>' -d -id <resource-id> -f ~/Downloads

Delete

Deletes a file or folder.

./df-agent -sid <service-id> -secret '<secret>' -x -id <resource-id>

Share

Shares a resource with another user.

./df-agent -sid <service-id> -secret '<secret>' -share -id <resource-id> -user colleague@example.com

For machine-to-machine authentication, use a Service Identity with -sid and -secret instead of personal login credentials.

Service Identity

Service Identity is the recommended authentication model for machine-to-machine communication — including agents, cron jobs, background services, and CI/CD pipelines. Service identities are scoped credentials that can be rotated, revoked, and audited independently of human user accounts.

Managing Credentials

Administrators create and manage service identities from the DataFrugal dashboard:

1. Log in as a System Administrator.
2. Open Settings → Service Identity & API Credentials.
3. Assign an owner responsible for the identity.
4. Enter a Service Identity name — this becomes the -sid value in df-agent.
5. Click Create credentials — copy the secret immediately; it is shown once only.
6. Use Rotate to issue a new secret (old secret stops working immediately).
7. Revoke identities that are no longer required.

UI → CLI Mapping

Credentials created in the dashboard map directly to df-agent flags:

Always pass -sid and -secret together when authenticating.

Machine Authentication API

Under the hood, df-agent authenticates service identities via the machine-user auth endpoint:

POST /ext/v1/df/machine-user/auth

{
  "username": "<sid>",
  "dfSecret": "<secret>"
}

See the df-agent guide for command examples and the API documentation for full endpoint details.

REST API Access

DataFrugal exposes a RESTful API for integrating backup, retention, and controlled data access workflows into applications and systems.

• Endpoints for object operations, metadata management, and access delegation
• Supports scoped access between services using role- and policy-based controls
• Interactive Swagger UI for exploration and testing
• Secure token-based authentication
• Designed for automation and system-to-system integration

Identity Provider Integration

DataFrugal integrates with external identity providers for authentication, allowing organizations to manage access using existing identity systems.

• Supports Google and Microsoft OAuth-based authentication
• Organization-scoped and role-based access boundaries
• Suitable for both human users and service identities

Backup Tool Integration

DataFrugal can be integrated with external backup and scheduling tools by invoking APIs or df-agent as part of existing workflows.

• Commonly used with tools such as Bacula for orchestrated backups
• Enables secure, HTTPS-based data transfer between systems
• Supports controlled access where downstream systems can read protected data without broad permissions

Security and Access Controls

• Role- and policy-based access delegation for users and service accounts
• Network-level restrictions such as IP allowlists
• Encryption at rest and in transit using industry standards
• Support for customer-managed encryption key configurations where applicable