| ID | product.platform.observability.log_pipeline |
|---|---|
| Description | Loki + Vector log aggregation. |
| Key | log_pipeline |
| Type | component |
| Team | Platform Engineering |
| Owner | platform@iotgateway.example |
| Status | active |
| Technologies | Grafana Loki, OpenTelemetry Collector |
| Attributes | ingest_rate: 200k lines/s, retention: 30d |
| Tags |
The Log Pipeline collects structured logs from every service, ships them through Vector for parsing and enrichment, and stores them in Grafana Loki for fast, label-indexed search.
Index labels, not lines
Loki indexes only a small set of labels (service, level, region) and stores the log body compressed. That keeps ingest cheap at 200k lines/s while still letting you grep within a label stream.
trace_id for correlation.Every log line carries the trace_id so a noisy error can jump straight into
the Trace Pipeline
for the full request story. Hot logs are retained 30d.
1 incoming link
| Source object | Label | Link type | Relation | Raw id | Required | Description | |
|---|---|---|---|---|---|---|---|
| correlates with | — | links | product.platform.observability.log_pipeline | — | — |