Skip to content
rahadbhuiyaPublic

Repository files navigation

CNSL

Correlated Network Security Layer

CI PyPI Python 3.10+ MIT License

A self-hosted SIEM for Linux.
Detects attacks that span SSH, web, database, and cloud logs simultaneously -- then blocks them automatically.


What it does

Most tools watch one log and count failures. CNSL watches everything at once.

When an attacker scans your web server, probes your database, then tries SSH with stolen credentials -- CNSL sees all three as one coordinated attack and responds before the breach completes.

Web scan      from 45.33.32.1  --+
SSH brute     from 45.33.32.1  --+--->  HIGH alert + auto-block
DB auth fail  from 45.33.32.1  --+

It also tracks how far each attacker has progressed through the kill chain, learns new attack patterns automatically, and shares threat intelligence across multiple servers in real time.

Beyond core detection: multi-node federation with a unified hub view, STIX 2.1 export and a built-in TAXII 2.1 server for sharing IOCs, Wazuh/OSSEC integration, PostgreSQL support with a migration tool, a Kubernetes Helm chart, attacker fingerprinting (spots the same actor rotating IPs), graph-based campaign correlation, and opt-in predictive blocking that reacts to an attack's trajectory before any single rule's threshold fires.


Quick start

pip install cnsl[full]
sudo python -m cnsl --dashboard --no-tcpdump
# Open http://127.0.0.1:8765
# Default login: admin / cnsl-change-me

Or from source:

git clone https://github.com/rahadbhuiya/cnsl.git
cd cnsl
python3 -m venv venv && source venv/bin/activate
pip install -e ".[full]"
sudo venv/bin/python -m cnsl --dashboard --no-tcpdump

Start in dry-run mode (default) -- no real blocks until you add --execute.

Or on Kubernetes (one DaemonSet pod per node, protecting every node in the cluster):

helm install cnsl ./helm/cnsl --namespace cnsl --create-namespace

See docs/kubernetes.md and helm/cnsl/README.md for the full walkthrough (federation, the multi-node hub view, enabling real blocking).


Dashboard

Enable with --dashboard. Runs at http://127.0.0.1:8765.

Tabs: Overview, Incidents, Blocks, Live Feed, Kill Chain, Graph, Correlation, Hub, Campaigns, Cases, UEBA, ML, Honeypot, FIM, Rules, Rate Limit, Settings.

For remote access use an SSH tunnel:

ssh -L 8765:127.0.0.1:8765 user@yourserver

Configuration

Fastest path -- the setup wizard (under 5 minutes):

sudo python -m cnsl --init

Asks a handful of questions (allowlist IP, dashboard on/off, notification channels), auto-detects which log files actually exist on this machine (nginx, apache, mysql, ufw, syslog, auth.log vs secure), and writes a working, validated config.json -- you don't need to read or understand the full reference config to get running.

Full reference, once you want to turn on optional features (Sigma rules, OIDC SSO, cloud identity, ATT&CK mapping, retention, case SLA, ...) -- all ~30 sections are documented individually in docs/configuration.md and default to off/sensible, so nothing there is required reading up front:

cp config/config.example.json /etc/cnsl/config.json

Key sections: thresholds, actions, dashboard, notifications, redis, cloud_identity, zero_trust, siem, federation.


Documentation

Document What it covers
docs/installation.md Full install, systemd, Docker
docs/configuration.md Every config option explained
docs/features.md Complete feature list
docs/architecture.md Module structure and design
docs/api.md Full REST API reference
docs/kubernetes.md Helm chart deployment (DaemonSet, hub view, federation)
docs/kill-chain.md Kill chain tracker (predictive blocking: see docs/api.md)
docs/federation.md Multi-node setup and the hub view
docs/cloud-identity.md AWS + Azure AD + GCP integration
docs/sigma-rules.md Sigma detection rule import
docs/attack-mapping.md MITRE ATT&CK technique tagging and coverage
docs/oidc-sso.md OIDC single sign-on setup
docs/source-health.md Log source silence detection
docs/retention.md Data retention and archival
docs/case-sla.md Case SLA tracking and escalation
docs/zero-trust.md Trust score engine
docs/siem-connectors.md Splunk, Sentinel, Webhook push
docs/pattern-learning.md Automated rule discovery
docs/rules.md Detection rule tuning (correlation-rule tuning: see docs/api.md)
docs/ueba.md User/entity behavior analytics
docs/cases.md Case management
docs/threat-feed.md External threat feed ingestion
docs/2fa.md Two-factor auth for the dashboard
docs/agent.md Remote log-shipping agent
docs/kafka.md Kafka log ingestion
docs/zeek.md Zeek log integration
docs/ot-iot.md OT/ICS protocol support (Modbus, DNP3, SCADA)
docs/country-blocking.md Geo-based blocking
docs/rate-limiting.md API rate limiting
docs/tenants.md Multi-tenant setup
docs/huddle.md Incident huddle/collaboration
docs/notifications.md Email/Slack/Telegram alerts
docs/changelog.md Version history
docs/ Browse all 33 guides

Requirements

  • Linux (Ubuntu 20.04+ / Debian 11+ / RHEL 8+)
  • Python 3.10+
  • Root or CAP_NET_ADMIN for iptables blocking

Optional: Redis (distributed blocklist + federation), MaxMind GeoIP database.


License

MIT. See LICENSE.