Softadastra Cloud

Softadastra Cloud is the backend for Softadastra Reliability Cloud.

It helps teams answer one critical question:

Does your application lose data when the network fails?

Softadastra Cloud receives reliability test data from local agents, stores the evidence, computes a reliability score, and exposes the results to the dashboard on softadastra.com.

What Softadastra Cloud does

Softadastra Cloud is a reliability testing control plane. It is designed to collect and analyze what happens when an application faces real-world failure conditions:

• network timeout
• retry failure
• unstable connection
• duplicate operation
• failed recovery
• possible data loss

The first MVP focuses on HTTP APIs. A local agent runs tests against an API, records events, sends them to Softadastra Cloud, and the dashboard shows the result.

Product promise

Softadastra shows whether your application loses data when the network fails.

Why this matters

Many applications work well when the network is stable. But real users often face: bad mobile networks, unstable Wi-Fi, server timeouts, failed retries, browser refreshes, app restarts, partial writes.

These failures can create hidden problems: lost orders, duplicate payments, missing updates, broken sync, incomplete reports, silent data loss.

Softadastra Cloud makes those failures visible.

How it works

1. Create a project
2. Create an agent
3. Run reliability tests locally
4. Submit runs and events to the cloud
5. Compute a reliability score
6. View the result in the dashboard

High-level flow:

Agent CLI / runtime
        |
        | X-Agent-Key
        v
Softadastra Cloud API
        |
        v
Projects, Agents, Runs, Events, Reports
        |
        v
Reliability Score
        |
        v
Dashboard on softadastra.com

Core concepts

Project

A project represents an application or API being tested.

PulseGrid API
https://pulsegrid.softadastra.com

Agent

An agent is a local runtime or CLI process allowed to submit reliability data for a project. Agents authenticate with API keys.

The plain API key is shown only once. Only the hash is stored by the backend.

Run

A run is one reliability test session containing: target URL, HTTP method, scenario, status, start time, finish time.

Event

An event is one observed fact during a run. Examples: agent_started, request_timeout, retry_started, recovery_succeeded, recovery_failed, duplicate_detected, data_loss_risk_detected.

Report

A report is the final summary of a run. It stores a human-readable summary and the raw JSON output from the agent.

Reliability score

The reliability score turns events into a risk profile.

100 = excellent
0   = dangerous

Risk flags: data loss risk, duplicate risk, timeout risk, recovery risk.

MVP scope

Included:

• projects
• agents
• runs
• events
• reports
• reliability scoring
• agent API key authentication
• dashboard API
• SQLite/MySQL-ready backend through Vix DB

Not included yet:

• billing
• teams
• organizations
• Kubernetes
• mobile SDK
• distributed agents
• advanced alerting
• P2P telemetry

Backend architecture

Softadastra Cloud follows a clean modular backend architecture.

Routes
  -> Controllers
  -> Services
  -> Repositories
  -> Database

Each domain has its own module: health, auth, projects, agents, runs, events, reports, scoring.

Example module structure:

modules/projects/
├── Project.hpp
├── ProjectController.hpp
├── ProjectController.cpp
├── ProjectService.hpp
├── ProjectService.cpp
├── ProjectRepository.hpp
├── ProjectRepository.cpp
├── ProjectRoutes.hpp
└── ProjectRoutes.cpp

This keeps HTTP logic, business rules, and SQL separated.

Technology

Softadastra Cloud is built with:

• C++20
• Vix.cpp
• Vix DB
• Vix JSON
• Vix middleware
• Vix tests
• SQLite for local development

Vix provides the runtime-oriented C++ backend foundation. Softadastra Cloud provides the product logic.

Repository structure

cloud/
├── LICENSE
├── README.md
├── CMakeLists.txt
├── CMakePresets.json
├── vix.json
├── .gitignore
├── .env.example
├── src/
├── sql/
├── tests/
└── docs/

Directory	Purpose
src/	Backend source code
sql/	SQL schema files
tests/	Unit and integration tests
docs/	Architecture, API, database docs

Configuration

Configuration is loaded through CloudConfig.

Main values:

Key	Description
service_name	Service identifier
environment	development or production
server_host	Bind address
server_port	Bind port
cors_allowed_origin	Allowed CORS origin
dev_mode	Enables dev helpers
auto_migrate	Run migrations on startup
jwt_secret	Dashboard token secret
agent_api_key_secret	Agent key secret
max_request_body_bytes	Max request body size

Example .env:

CLOUD_SERVICE_NAME=softadastra-cloud
CLOUD_ENVIRONMENT=development

SERVER_HOST=127.0.0.1
SERVER_PORT=8080

CLOUD_CORS_ALLOWED_ORIGIN=http://localhost:5173
CLOUD_DEV_MODE=true
CLOUD_AUTO_MIGRATE=true

CLOUD_JWT_SECRET=change-me
CLOUD_AGENT_API_KEY_SECRET=change-me
CLOUD_MAX_REQUEST_BODY_BYTES=1048576

API groups

GET  /health

POST /api/auth/register
POST /api/auth/login
GET  /api/auth/me

GET  /api/projects
POST /api/projects
GET  /api/projects/:id
PUT  /api/projects/:id
DEL  /api/projects/:id

GET  /api/agents
POST /api/projects/:project_id/agents
GET  /api/agents/:id
POST /api/agents/:id/rotate-key
DEL  /api/agents/:id

GET  /api/runs
POST /api/runs
GET  /api/runs/:id
POST /api/runs/:id/status
POST /api/runs/:id/finish
DEL  /api/runs/:id

GET  /api/events
POST /api/runs/:run_id/events
GET  /api/events/:id

GET  /api/reports
POST /api/runs/:run_id/report
GET  /api/reports/:id
GET  /api/runs/:run_id/report
DEL  /api/reports/:id

GET  /api/runs/:run_id/score

Response format

Success:

{
  "success": true,
  "message": "Resource loaded",
  "data": {}
}

Error:

{
  "success": false,
  "error": "validation_failed",
  "message": "Invalid input"
}

List:

{
  "success": true,
  "message": "Resources loaded",
  "data": [],
  "meta": {
    "total": 10,
    "limit": 20,
    "offset": 0
  }
}

Scoring model

Start from 100, then subtract:

Event	Penalty
data_loss_risk_detected	-35
duplicate_detected	-20
request_timeout	-10
recovery_failed	-25
critical severity	-15 each, max -30
error severity	-8 each, max -24
warning severity	-3 each, max -15

Score clamped between 0 and 100.

Grades:

Grade	Range
A	90 – 100
B	75 – 89
C	60 – 74
D	40 – 59
F	0 – 39

Development

Configure and build:

vix build

Run the backend:

./build-ninja/softadastra-cloud

Run tests:

vix tests

Documentation

File	Content
docs/architecture.md	Backend architecture and module design
docs/api.md	Full API reference
docs/database.md	Database schema and design

Final mental model

Agent observes failures locally.
Cloud stores the evidence.
Scoring turns evidence into risk.
Dashboard makes the risk visible.

Softadastra Cloud is the control plane for reliability testing.