Insights
Insights is where you read your telemetry back. It ships with the tenant, so there is nothing to connect: the tenant you are looking at is the data source.
Open Insights in the xScaler portal.
The pages
| Page | What you do there |
|---|---|
| Metrics | Search for a metric, break it down by label, read the result |
| Logs | Filter a stream by label, read lines, open the trace a line belongs to |
| Traces | Span waterfall, critical path, service graph, exceptions |
| Explorer | Several queries at once, each with its own signal |
| Dashboards | Build panels, import Grafana JSON, keep version history |
| Snapshots | Freeze what is on screen and share the link |
| Alerting | Alert rules, contact points, notification policies, silences |
Moving between signals
Every result view carries the same two actions.
Add to dashboard takes the query you are looking at, metrics, logs or traces, and writes it into a dashboard panel. Pick an existing dashboard or create one from the modal.
Create alert opens the rule editor with the query already filled in. It appears on metric and log queries. The alert model covers Prometheus and Loki queries, so trace queries do not offer it.
A log line carrying a trace_id links straight to that trace. A span links back
to the logs recorded while it ran.
Saved queries
Any query can be saved and starred. Starred queries stay until you delete them. Unstarred ones can be cleared in bulk once the list gets long.
Your own Grafana still works
The Prometheus-, Loki- and Tempo-compatible query APIs stay open. Existing dashboards keep reading, and new ones can be built anywhere that speaks those APIs. See Connect Grafana datasources, or let xScaler run the Grafana for you with Managed Grafana.