DiunDashboard

What This Is

A web-based dashboard that receives DIUN webhook events and presents a persistent, visual overview of which Docker services have available updates. Built for self-hosters who use DIUN to monitor container images but need something better than dismissable push notifications — a place that nags you until you actually update.

Core Value

Reliable, persistent visibility into which services need updating — data never disappears, and the dashboard is the one place you trust to show the full picture.

Requirements

Validated

✓ Receive and store DIUN webhook events — existing
✓ Display all tracked images with update status — existing
✓ Acknowledge/dismiss individual updates — existing
✓ Manual tag/group organization via drag-and-drop — existing
✓ Tag CRUD (create, delete, assign, unassign) — existing
✓ Optional webhook authentication via WEBHOOK_SECRET — existing
✓ Docker deployment with volume-mounted SQLite — existing
✓ Auto-polling for new updates (5s interval) — existing
✓ Service icon detection from image names — existing

Active

Fix SQLite foreign key enforcement (PRAGMA foreign_keys = ON)
Fix INSERT OR REPLACE data loss (switch to proper UPSERT)
Add PostgreSQL support alongside SQLite (dual DB, user chooses)
Bulk acknowledge (dismiss all, dismiss by group)
Filtering and search across updates
In-dashboard new-update indicators (badge/counter/toast)
Request body size limits on webhook and API endpoints
Improve error handling in tests (replace silent returns with t.Fatal)
Data persistence resilience (survive container restarts reliably)

Out of Scope

DIUN bundling / unified deployment — future milestone, requires deeper DIUN integration research
Auto-grouping by Docker stack/compose project — future milestone, requires Docker socket or DIUN metadata research
Visual DIUN config management UI — future milestone, depends on DIUN bundling
Notification channel management UI — DIUN already handles this; visual config deferred to DIUN integration milestone
OAuth / user accounts — single-user self-hosted tool, auth beyond webhook secret not needed now
Mobile app — web-first, responsive design sufficient

Context

User hosts services on a VPS using Coolify (Docker-based PaaS)
DIUN monitors container images for new versions and sends webhooks
Previous approach (Gotify push notifications) failed because notifications were easy to dismiss and forget
Dashboard was a daily driver but data loss (likely volume misconfiguration + SQLite bugs) eroded trust
Coolify doesn't show available updates — this fills that gap
Target audience: self-hosters using DIUN, not limited to Coolify users
Existing codebase: Go 1.26 backend, React 19 + Tailwind + shadcn/ui frontend, SQLite via modernc.org/sqlite

Constraints

Tech stack: Go backend + React frontend — established, no migration
Database: Must support both SQLite (simple deploys) and PostgreSQL (robust deploys)
Deployment: Docker-first, single-container with optional compose
No CGO: Pure Go SQLite driver (modernc.org/sqlite) — must maintain this for easy cross-compilation
Backward compatible: Existing users with SQLite databases should be able to upgrade without data loss

Key Decisions

Decision	Rationale	Outcome
Dual DB (SQLite + PostgreSQL)	SQLite is fine for simple setups, Postgres for users who want robustness	— Pending
Fix SQLite bugs before adding features	Data trust is the #1 priority; features on a broken foundation waste effort	— Pending
Defer auto-grouping to future milestone	Requires research into Docker socket / DIUN metadata; don't want to slow down stability fixes	— Pending
Defer DIUN bundling to future milestone	Significant scope; need stability and UX improvements first	— Pending

Evolution

This document evolves at phase transitions and milestone boundaries.

After each phase transition (via /gsd:transition):

Requirements invalidated? → Move to Out of Scope with reason
Requirements validated? → Move to Validated with phase reference
New requirements emerged? → Add to Active
Decisions to log? → Add to Key Decisions
"What This Is" still accurate? → Update if drifted

After each milestone (via /gsd:complete-milestone):

Full review of all sections
Core Value check — still the right priority?
Audit Out of Scope — reasons still valid?
Update Context with current state

Last updated: 2026-03-23 after initialization

4.6 KiB Raw Blame History