Mongostat

What it is

A workspace tab with a chart at the top, a side panel of filters / metric pickers, and the operations capture below. Each sample stores opcounters, network I/O, lock states, page faults, and (when "capture ops" is enabled) the current db.currentOp().inprog snapshot — so you can click any point in the past and see exactly which queries were running.

When to use

Open before maintenance windows to baseline the cluster. Keep open during incident response to correlate the spike with what was running. Replay later for root-cause analysis. Auto-trigger source for Alerting rules.

How to open

1. In the NoSqlStudio desktop app, open the Tools menu → Dashboards → Mongostat, or press Ctrl+Alt+Shift+M.
2. The workspace opens with a chart of opcounters by default. The side panel on the right lets you toggle metrics and adjust the time window.
3. Click "Enable capture ops" to record the db.currentOp() snapshot at each sample — adds load but unlocks replay.

Working with it

• Metric pickers — opcounters (insert / update / delete / query / getmore / command), network in/out, locks, page faults, connections.
• Filter presets — slow > 100ms, COLLSCAN-heavy, low-selectivity, write-heavy, ineffective. Each is a preset combination of capture filters.
• Replay — click any point on the chart; the captured ops snapshot for that moment appears below with the full operation JSON.
• Storage — samples persisted in userData/Mongostat/. Gear icon in the toolbar configures retention + mirror to MongoDB (best-effort secondary storage).

Limitations (v1)

• Most operations finish in <1ms — they rarely appear in db.currentOp() within a single tick. Click Pause + Resume to widen the sampling window.
• Mirror to MongoDB is best-effort. If the secondary storage is unreachable, samples still persist to local file and back-fill when the connection returns.
• Free tier — some fields (network I/O, page faults) are hidden by Atlas; corresponding chart lines are flat. The Realtime Dashboard banner explains.