Governance & Acceptable Use
EphemeralNet’s bootstrap, TURN/STUN, and shared control-plane infrastructure must remain reliable without becoming an abuse vector. This policy mirrors the guidance applied to the canonical *.shardian.com services; adapt it to your jurisdiction and hosting agreements.
Scope & goals
- Covered services: Public bootstrap/control daemons, relay infrastructure, and automation that the community relies upon for discovery.
- Authoritative hosts:
bootstrap1.shardian.com,bootstrap2.shardian.com,stun.shardian.com,turn.shardian.com. Mirrors must meet the same uptime + abuse handling standards. - Audience: Operators, incident responders, and moderators coordinating access to shared infrastructure.
- Objectives: Preserve availability, protect user privacy, and minimise illegal use of TTL storage.
Acceptable use principles
- Ephemeral data only – Participants must honour TTL enforcement; attempts to persist data beyond expiry violate the policy.
- Lawful content – No malware, doxxing, hate speech, or otherwise illegal material.
- Respect throttles – Clients must honour announce windows, PoW gates, and upload quotas.
- Identity hygiene – Rotate session keys (
--key-rotation), protect control tokens, and avoid sharing secrets publicly. - Telemetry transparency – Operators may collect aggregate metrics (announce rate, failure counters) solely for capacity planning and abuse mitigation; payload contents remain off-limits.
Prohibited behaviour
- Flooding control endpoints beyond published quotas.
- Circumventing proof-of-work or replaying manifests to exhaust storage.
- Attempting to deanonymise peers or intercept encrypted traffic.
- Using EphemeralNet infrastructure for spam, phishing, or coordinated abuse.
Reporting & response workflow
- Intake: Route abuse reports to a dedicated mailbox or ticket queue. Capture timestamps, manifests, peer IDs, and evidence.
- Triage: Within 24 hours, validate the report, replay manifests when safe, and inspect daemon logs/metrics for corroboration.
- Containment: Apply countermeasures such as raising PoW difficulty, tightening announce burst/window, revoking tokens, or blacklisting peer IDs.
- Eradication: Force TTL expiry on offending manifests or issue targeted cleanup commands.
- Recovery: Restore baseline throttles when traffic normalises, rotate relay credentials if required, and notify affected operators.
- Post-incident review: Document root cause, controls applied, and follow-up tasks in the runbook.
Enforcement toolkit
- Reputation system: Use built-in scoring to demote abusive peers automatically.
- Control tokens: Issue per-tenant secrets and rotate them when compromised.
- Announce throttles & PoW: Adjust interval/burst/window and difficulty bits based on telemetry trends.
- Blocklists: Maintain encrypted lists of banned peer IDs/IPs, distribute via secure channels.
Data retention & privacy
- Retain control-plane logs no longer than 30 days unless required for active investigations.
- Anonymise IPs when exporting metrics outside the immediate ops team.
- Never share raw manifests or chunk data without explicit consent or legal mandate.
Review cadence
- Revisit this policy quarterly or after major incidents.
- Align updates with changes to performance tuning and security documentation to keep guidance consistent.
Follow this governance chapter when onboarding new infrastructure partners, responding to abuse, or designing runbooks; it ensures the swarm stays open while respecting legal and ethical boundaries.