RousselTMLEARNING

Observability

Cribl: the data pipeline for observability and security

Discover Cribl: why a telemetry data pipeline, the Cribl suite (Stream, Edge, Lake, Search, Cribl.Cloud), the sources, routes, pipelines and destinations model, licensing and comparison with OpenTelemetry Collector, Bindplane, Vector, Logstash and Fluent Bit.

1. Why a data pipeline? #

The problem Cribl solves.

The situation #

Companies send their logs, metrics and traces directly to their tools: SIEM, observability platform, archive. Each source is wired to each tool, each tool bills the volume it receives, and switching tools means reconfiguring hundreds of agents. A data pipeline sits between sources and destinations to route, reduce, enrich and normalize data in one place.

Without and with a pipeline

What you gain #

  • Cost: do not index in the SIEM what is useless there (allowed traffic, empty fields, duplicates); cutting 30 to 50% of the volume sent is common.
  • Freedom: change or add a tool without touching sources; send the same data to two tools during a migration.
  • Quality: normalize formats, add context (team, geolocation, criticality), mask personal data before it leaves the company.
  • Retention: keep 100% of raw data cheaply on object storage and replay it on demand.

2. The Cribl suite #

Five complementary products.

The products #

The Cribl suite
ProductRoleWhere it runs
Cribl Streamthe central pipeline: sources, routes, pipelines, destinations, replayyour servers, Kubernetes or Cribl.Cloud
Cribl Edgelightweight agent: files, journald, Windows events, metrics, Kuberneteson each host or node
Cribl Lakemanaged dataset storage, retention per datasetCribl.Cloud
Cribl Searchfederated search where data lives (S3, Lake, APIs), without re-ingesting itCribl.Cloud
Cribl.Cloudmanaged control plane, workspaces, Copilot (AI assistant)SaaS
Good to know : This course focuses on Cribl Stream, the core of the suite, and covers Edge, Lake and Search in their collection, storage and search roles.

3. The processing model #

Sources, routes, pipelines, destinations.

An event's journey #

Event journey
ConceptDefinition
Sourcedata entry point: listening (syslog, HEC, OTLP) or collecting (S3, REST API)
Routeordered rule: a JavaScript filter, a pipeline, a destination
Pipelinesequence of functions applied to events
Functionunit processing: Parser, Eval, Drop, Mask, Lookup, Aggregations…
Destinationoutput: Splunk, Elastic, Datadog, S3, Kafka, Loki, Cribl Lake…
Packreusable bundle of routes, pipelines and reference data

The Cribl event #

An event is a JSON object. Some fields have special meaning:

FieldMeaning
_rawthe original event text
_timetimestamp in seconds since the Unix epoch
host, source, sourcetype, indexmetadata inherited from the Splunk world, used to filter and route
__inputId, __*internal fields (double underscore): usable in filters, never sent to destinations
cribl_*information added by Cribl (pipeline, breaking)

4. Licensing and versions #

Starting for free.

The offerings #

  • Free: up to 1 TB per day, on Cribl.Cloud or self-hosted. Self-hosted, the free licence requires sending anonymized telemetry to Cribl and is limited to a single Worker Group.
  • Paid (Standard, Enterprise): higher volumes, multiple Worker Groups, RBAC, single sign-on, support; on Cribl.Cloud, usage is counted in credits.
  • Versions: a minor release about every two months (4.17 in March 2026, 4.19 in July, then 4.20); on Cribl.Cloud, a “Regular” or “Slow” release channel per workspace.
Best practice : Cribl University offers free courses and certification tracks (Cribl Certified User, then Admin Stream or Edge, then Engineer): a good complement to this course.

5. Cribl and alternatives #

Where Cribl fits.

Comparison #

ToolStrengthDifference with Cribl Stream
OpenTelemetry Collectoropen standard, free, OTLP-nativeYAML configuration, no UI or data preview, thinner security ecosystem
Bindplanefleet management for OpenTelemetry collectorsbuilt on OpenTelemetry; Cribl has its own engine and event format
Vectorvery fast, open source (Datadog)file-based configuration, no graphical control plane
Logstash, Fluent Bithistorical, free, widespreadno centralized multi-tool routing or built-in replay
Cribl StreamUI with before/after preview, replay, 100+ integrations, SIEM-orientedproprietary software, paid beyond 1 TB per day

Commands to know #

Commands to know
  • docker run -d -p 9000:9000 cribl/cribl:latest : run Cribl Stream locally
  • /opt/cribl/bin/cribl status : service status
  • /opt/cribl/bin/cribl version : installed version
  • sourcetype=='pan:traffic' : example route filter

Sources #

Checked in November 2026 (Cribl 4.20): Cribl ships a minor release about every two months, check the release notes before relying on a detail.

Related articles

Installing and administering Cribl Stream

Installing Cribl Stream with Docker or on Linux, running it as a service, understanding $CRIBL_HOME, default …

Sources, destinations and routes in Cribl Stream

Receiving and sending data with Cribl Stream: push, pull, collector and internal sources, event breaking, des…

Pipelines and functions: parse, enrich, mask

Building Cribl Stream pipelines: pre-processing, processing and post-processing, the function catalogue (Pars…

Keep learning

Support us

♥ SUPPORT US

Support RousselTM Learning

Courses, articles, labs and certificates stay free with no intrusive ads. A voluntary contribution, of any amount you choose, pays for hosting, lab environments and new content.

Contribute with PayPal Any amount · card or PayPal accountLearn more
QR code PayPal
Or scan this QR code with your phone

Thank you, every contribution counts!