Prometheus — prometheus
Takes instant-query snapshots of PromQL expressions on every poll tick — each query becomes a stream
of metric events, joinable with logs and deploys on the same key. Supports
discovery: given a url, it lists the metrics the server carries and
proposes queries.
Prerequisites: a reachable Prometheus server (its base URL, e.g. http://localhost:9090).
Set up
Agent
Ask a connected agent:
Use navflow to ingest the 5xx rate per service from the Prometheus at
localhost:9090.
The agent calls discover_source (introspects the server’s metrics, proposes queries),
test_source (dry-run — evaluates the queries once), then create_source.
Configuration
| field | type | description |
|---|---|---|
url* | string | Prometheus base URL, e.g. http://localhost:9090 |
default_key | string | entity key for series that carry no key label (default unknown) |
queries* | list | metrics to ingest — each a PromQL query mapped into the envelope |
Each queries entry:
| field | type | description |
|---|---|---|
promql* | string | PromQL (a bare metric name = raw, lossless ingest) |
event_type | string | event type for this query’s events (default metric) |
field | string | fields.<name> the numeric value is stored under (default value) |
text | string | rendered line template (default {key} {field}={val}) |
key_label | string | series label to read the entity key from, e.g. service |
Verify
Open the source in the console: Recent events shows one event per series per poll. Watch a metric move in Explore by picking the entity the series is keyed on.