Dashboard Tabs

Dashboard Tabs

The dashboard has eleven main tabs plus Docs, Settings, and Help. This page explains what each one is for and when to use it.

---

At a glance

Tab	Use this when…	Main data source
Overview	You want a quick health summary	Live + last report
Live	You are watching performance right now	Live charts
Insights	You want patterns over days/weeks	Minute history + report
Session	You want to see who is online	Live + last report
Sources	You wonder “when did this last update?”	All layers
Issues	You want a prioritized fix list	Full report
Crashes	A player crashed or the server died	Background scan + report
Mods	Mod errors or conflicts	Background scan + report
Backups	You need to know if backups are recent	Background scan + report
Activity	You want a timeline of events	Background scan + report
Spark	You profiled lag with Spark	Full report

---

Overview

Your home screen after at least one health report.

• Overall health (ok / warning / critical)
• Key numbers: speed, CPU, memory, players
• Short list of recent problems with link to Issues
• Teaser for Insights patterns
• Storage and recent activity snippets
• Run full report — refreshes everything that depends on a deep check

Tip: Run a report first for the full fix list. Many cards still update from background scans before that.

---

Live

Watch the server right now.

• Right now alerts — backup running, restart scheduled, stale logs, mod file changes
• Game: TPS, tick lag (MSPT), memory, players — with charts
• Host: CPU, RAM, disk, network when available
• Change refresh speed and chart time range

Data is recorded every second (configurable) and kept up to 90 days.

---

Insights

See patterns over time — busy hours, lag vs player count, week comparisons.

Sub-areas:

Area	What it shows
Patterns	Heatmaps, averages, sticky lag episodes, outliers
Mod changes	Mods added, removed, or updated since last report
Storage	Disk jumps and world size breakdown

Needs some uptime history (ideally 24h+) for meaningful charts. Overview only shows a one-line teaser — full analysis is here. CSV export is on this tab.

---

Session

Who is on the server and who was active recently.

• Live list: online players, ping, dimension
• Search, sort, copy player UUID
• Playtime and session history from last report
• 24h player count sparkline

Requires server online for live status; historical stats come from reports.

---

Sources

Understand what updates how.

• When live charts, background scan, and last full report ran
• Plain explanation of the three update layers
• Table of which features need live vs scan vs report
• Link to Settings → Monitoring

---

Issues

Your fix list from the latest health report.

Three groups:

1. Needs attention now — crashes, out-of-memory, server down signals
2. Worth fixing when you can — lag, stale backups, mod warnings
3. Historical — older or acknowledged items

Each item has severity, what happened, and suggested steps.

---

Crashes

Crash reports in plain English.

• Headline, time, likely cause, suspected mod
• Steps to try
• What happened in the 10 minutes before the crash
• Mark as reviewed when handled
• Filters: all, needs review, mod-related, host/resource

Crash folder is checked in the background; Refresh rescans without a full report.

---

Mods

Mod health — log errors update in the background; full mod list and conflict analysis need a report.

Pages: Overview, Update conflicts, Client-only mods, Log errors.

Search and jump between sections. Badge shows count of mods with log errors.

---

Backups

Are your backups recent and configured?

Watchtower does not guess backup locations — you set them in Settings → Backups or Choose backup folder on this tab.

Shows panel backup status, local archive list, rescan, and plain “what to do” steps when backups are missing or old.

See Backups for setup.

---

Activity

Timeline of server events — joins, leaves, crashes, reboots, commands, backups, and more.

Search and filter by type. Not the same as Insights (which shows patterns over time). Enable Scheduled Reports so older events stay populated.

---

Spark

Lag breakdown from a Spark CPU profile — pick any saved .sparkprofile from the dropdown (on-demand parse). No full report required to view; optional Run report adds Spark to Overview and brief.txt.

How to use Spark workflow at the top: capture while lagging → pick profile → read advice (collapses after first load). Five sub-views — Summary (verdict, KPIs, findings, recommendations), Mods & code (mod usage, signals, hot methods), World (entities, hotspots, dimension breakdown), Capture window (timeline), Advanced (performance chart, method table, JVM/config details). Choice persists per browser. See Using-Spark-with-Watchtower.

---

Admin — Docs

All guides inside the dashboard — search, categories, diagrams on key pages. Same content as this wiki. Help (?) links here for full articles.

---

Technical appendix

Internal data layers and APIs for maintainers:

Tab	Technical notes
Overview	L0 vitals + ops peek; master refresh = full report
Live	L0 + ops-cache Right now feed
Insights	L1 rollups + ops-cache; tab id performance; GET /api/performance/dashboard
Session	GET /api/players + player_directory / window_stats from facts
Sources	GET /api/data-sources
Issues	Full queue from facts; live peek on Overview
Crashes	OPS_LOG_SCAN_SEC folder scan + facts pre_crash
Mods	Ops-cache log errors + facts manifest
Backups	backups_live ops-cache merge + facts
Activity	Ops ledger + facts backfill; subtitle references OPS_LOG_SCAN_SEC
Spark	On-demand profile list/parse (/api/spark/*); report embeds newest for Overview

Feature details and UI notes are in Changelog.

---

See also

• Live Charts
• Health Reports
• Quick Start Checklist
• Understanding-Data-Sources