Full-Stack Correlation Board
The "war room" dashboard — host health, service performance, log errors, and Davis problems correlated in one view.
Row 1: Infrastructure
Tile: CPU by Host (Line Chart)
timeseries avg(dt.host.cpu.usage), by:{dt.entity.host}
Tile: Memory by Host (Line Chart)
timeseries avg(dt.host.memory.usage), by:{dt.entity.host}
Row 2: Services
Tile: Response Time (Line Chart)
timeseries avg(dt.service.request.response_time), by:{dt.entity.service}
Tile: Error Rate (Line Chart)
timeseries avg(dt.service.request.failure_rate), by:{dt.entity.service}
Row 3: Logs & Problems
Tile: Log Errors Over Time (Line Chart)
fetch logs
| filter loglevel == "ERROR"
| makeTimeseries errors=count(), by:{log.source}
Tile: Active Problems (Table)
fetch events, from:now()-24h
| filter event.kind == "DAVIS_PROBLEM"
| fields display_id, event.name, event.status, timestamp
| sort timestamp desc
| limit 5
The Correlation Pattern
When you see this... Check this...
────────────────────────────── ──────────────────────────────
CPU spike on host → Service RT on same host
Service RT spike → Log errors from that service
Log errors spike → Davis problems (auto-correlated)
Davis problem opened → Affected entities in Smartscape
💡 Set all tiles to the same time range. When you see a spike in one tile, check the same timestamp in other tiles — that's full-stack correlation.
🛠 Try it: Build this dashboard tile by tile. Start with host health, add service performance, then logs, then Davis problems. The power is in seeing all four signal types on one screen — when a problem fires, you can instantly see which layer caused it.