DigitalOcean Referral Badge
cloud1
cloud2
cloud3
cloud4
cloud5
cloud6
← Back to catalog
code scan_velocity

Scan velocity

Syndu annotates all incoming traffic and extracts behavioral signals that help explain intent. This page defines the scan_velocity signal — what it means, how to interpret it, and how it will later connect to live evidence across IPs, subnets, organizations, ISPs, countries, and cities.

Signal gist High request rate and broad endpoint coverage suggest scanning or automated enumeration.
Read the definition → View live signal sections →

Definition

Canonical reference for scan_velocity behavior.
Catalog code
scan_velocity
Display name
Scan velocity
How to read this signal
This annotator represents a behavioral pattern, not a claim of identity. It’s designed to help you understand why certain traffic looks suspicious, automated, probing, or exploit-oriented — and to support consistent reporting across the Syndu system.
Explanation
Flags clients that hit many endpoints quickly, sustain a cadence inconsistent with human browsing, or enumerate routes and resources systematically. This is a tempo-and-breadth signal: requests-per-minute, unique-path count, and burst patterns tend to be the most explanatory features. Legitimate crawlers/monitors can trigger this and should be handled with allowlists or policy context.

Live sections

These panels will be wired to real metrics, enrichment context, and drill-down links.
Signal footprint over time
Rolling volume, bursts, first/last seen, and time-window slices (e.g. last hour/day/week). This will help separate chronic background noise from active campaigns.
Coming next: time series + burst markers
Top affected entities
Links to the entities where scan_velocity is most present: IPs, subnets, organizations/ASNs, ISPs, and geographies — with “why” context.
Coming next: entity leaderboards + drill-down
Enrichment context
How enrichment affects interpretation: known crawlers, monitored ranges, trusted scanners, or policy exceptions. This is where “benign but noisy” gets separated from “unknown and risky.”
Coming next: enrichment flags + allowlist context