`rsigma engine tail`🔗

Stream a running daemon's live detections to the terminal.

Synopsis🔗

rsigma engine tail [OPTIONS]

Description🔗

engine tail answers "what is the daemon firing right now?" without standing up a sink or scraping /metrics. It opens a streaming connection to a running engine daemon on GET /api/v1/detections/stream and prints each detection and correlation result as it fires, in the same EvaluationResult shape the sinks emit, so engine tail and a saved sink file are the same format.

rsigma engine tail --level high --rule whoami

It is the detections-out counterpart to engine tap (events-in). Where the tap records what the engine received, tail shows what it fired, after post-evaluation enrichment and regardless of which sinks are configured.

The stream is lossy by design: it can never apply backpressure to the sink task or stall the at-least-once ack-join. If a session's buffer fills under load, detections are dropped and counted, and the final summary record reports the gap.

Like engine status it is a read-only client over the admin API. It uses a synchronous HTTP client and does not need the daemon build feature, and it follows the same address convention as config reload: --addr defaults to daemon.api.addr, and wildcard binds (0.0.0.0, [::]) map to loopback.

The tail is disabled by default. Enable it on the daemon with the --enable-tail flag or daemon.tail.enabled: true in the config; otherwise the endpoint returns 503.

Filters🔗

Two optional server-side filters keep a noisy daemon's tail readable:

--level <severity>: minimum severity (informational, low, medium, high, critical). Results below it, or with no level, are excluded.
--rule <substring>: a case-insensitive substring matched against the rule title or id.

Both are applied at the sink, so filtered-out results never cross the wire.

Flags🔗

Flag	Default	Description
`--addr <HOST:PORT or URL>`	from `daemon.api.addr`	Daemon API address as `host:port` or a full URL. `https://` URLs work for TLS deployments.
`--duration <D>`	unset	Capture window (humantime). Unset streams until interrupted or `--limit` is reached.
`--limit <N>`	unset	Stop after N detections, before the duration if reached first.
`--level <severity>`	unset	Minimum severity filter.
`--rule <substring>`	unset	Case-insensitive title/id substring filter.
`-c, --config <PATH>`	discovery chain	Explicit config file used to resolve the daemon address.

Rendered through the global --output-format layer: a TTY-aware default (pretty json on a terminal, ndjson when piped) plus csv/tsv row streaming and a table view (buffered, so table suits a bounded --duration/--limit tail rather than an open-ended one). The global --quiet / --no-stats flags suppress the stderr stats line. See Output Formats.

Examples🔗

Watch high-severity detections🔗

rsigma engine tail --level high

Follow a specific rule and pipe to jq🔗

rsigma engine tail --rule "suspicious login" | jq '.matched_fields'

Capture a fixed window to a file🔗

rsigma engine tail --duration 5m --output-format ndjson > detections.ndjson

Stop after the first N detections🔗

rsigma engine tail --limit 20 --output-format table

Output🔗

The stream is NDJSON: one result per line, followed by a final summary record the client uses for the stats line and keeps out of the rendered output:

{"rsigma_tail_summary":{"streamed":42,"dropped":0}}

A non-zero dropped means a session buffer filled under load; the tail missed detections.

Exit codes🔗

Code	Meaning
`0`	The stream ended cleanly (even with zero detections).
`3`	The daemon could not be reached, returned a non-2xx status (e.g. `503` when the tail is disabled, `409` at the session cap, `400` for bad params), or sent an unreadable stream.

rsigma engine tail🔗