Find the line, then understand the pattern.
You found the one error. Is it isolated, or is it happening on every host? Log Insights groups by any field - status_code, host.name, http.route - so you go from one record to its blast radius in one click.
- Search by time range, service, environment, and host
- Severity filtering and display
- Structured field conditions
- Log detail inspector for full context
- Log Insights / Analyze for grouped field exploration
From a log line to an alert and a board.
The investigation does not end at search. The same field that filtered your view becomes the basis for an alert, a saved view, or a dashboard count - so "this happened once" turns into "we get paged before it happens again."
- Create alert rules from grouped log fields
- Custom log table columns through Saved Views
- Log count trend and latest-logs dashboard widgets
- JSONL log export on paid plans
Pair it with.
Questions about logs.
Are logs full-text searchable?
Yes, across services, hosts, and severity. You can also filter by structured attribute equality and use the log inspector to drill into a single record.
Do you preserve OpenTelemetry log attributes?
Yes. Resource attributes (service.name, host.name, deployment.environment) and log attributes are indexed and available as filters and group-by fields in Log Insights.
Can I jump from a log to its trace?
Yes. If the log carries a trace_id, the inspector links straight to the matching trace and span. Add OpenTelemetry's log appender or use the Collector's transform processor to attach trace context.
How is JSON parsed?
Send pre-parsed structured logs via OTLP and the original keys become indexed attributes. For raw text, use Data Shaper rules to extract fields at ingest.
How long are logs retained?
Free: 7 days. Solo: 30 days. Team: 60 days. Older logs are permanently deleted; use exports to archive specific result sets.
Search your first logs in minutes.
Send OTLP logs and start exploring - no proprietary SDK required.