Event logging is the backbone of incident response, forensic analysis, and compliance reporting — ECC Control 2-12-1 makes that explicit by requiring documented retention schedules, machine-readable formats, and an approval trail for logging controls; this post shows how to build an audit-ready logging program for the Compliance Framework with practical steps, small-business examples, and technical configuration pointers.
Why retention, format, and approvals matter for Compliance Framework
Retention ensures that you can investigate incidents and meet legal or regulatory obligations; standardized machine-readable formats ensure logs are parseable and searchable by SIEMs and forensic tools; documented approvals demonstrate governance and that retention choices are risk-informed. Auditors look for: a policy that states retention durations and formats, technical evidence that logs were collected and retained as stated, and signed approvals for exceptions or deviations.
Implementation: inventory, classification, and policy
Inventory and classify log sources
Start with an inventory: list servers, endpoints, network devices, cloud services, containers, and business apps. For a small SaaS company (50 employees), typical items include web app logs, authentication logs (IdP), cloud audit logs (AWS CloudTrail), database audit logs, and endpoint telemetry. Classify each source by sensitivity and purpose (security, access, billing, debug). Use a spreadsheet or lightweight CMDB with fields: source name, owner, event types, retention requirement, format, and onboarding status.
Define retention and format policy
Create a retention schedule mapped to classification. Typical minimums to start with (adjust to legal/regulatory needs): authentication and security events — 365 days; network flow metadata — 90 days (retain longer if needed for investigations); application error and transaction logs — 180–365 days depending on business need. Prefer structured, machine-readable formats (JSON with well-defined schemas, RFC5424 syslog with structured data, or CEF/LEEF where SIEM integrations expect them). Required fields: ISO8601 timestamp with timezone, event_id, event_type, source_host, user_id, severity, source_ip, destination_ip, and raw_message. Ensure NTP/chrony is enforced across infrastructure so timestamps are reliable.
Technical configuration: collection, storage, and integrity
Implement collectors and forwarding: use rsyslog/journald + Filebeat/Winlogbeat for endpoint forwarding; example rsyslog forwarding for RFC5424 over TLS: *.* action(type="omfwd" target="siem.example.local" protocol="tcp" StreamDriver="gtls" StreamDriverMode="1" StreamDriverAuthMode="x509/name"). In cloud, enable native audit logs (AWS CloudTrail, Azure Monitor) and pipe them to a central store. For long-term retention, use object storage with lifecycle rules (e.g., S3 + Glacier) and consider S3 Object Lock or vendor WORM feature for immutable retention. Protect log integrity by creating periodic manifests and hashes (e.g., sha256sum logs and store manifest files separately or in a signed artifact). For example: sha256sum /var/log/auth.log >> /var/log/manifests/auth.log.sha256; store the manifest in an immutable repository and keep access logs for that repository.
Approval, onboarding, and change control
Control 2-12-1 requires that retention, format, and any exceptions be approved and documented. Implement a lightweight "Log Onboarding" form that must be approved by the system/application owner and the security owner before a source is trusted for retention. Form fields: source name, data types, retention justification, storage location, format/schema, expected daily volume, access restrictions, and expiration/archival plan. Route approvals through your change management system (Jira/ServiceNow) so there is an auditable ticket trail. For exceptions (shorter retention, non-standard format), require business risk justification and an explicit approval signature with review cadence (e.g., quarterly). Keep a signed policy document that maps each log class to retention and the approver's identity and date.
Small-business scenarios and actionable examples
Scenario 1 — 20-person e-commerce shop: enable application and payment gateway logs, forward to a central VM running an open-source SIEM (Elastic + Filebeat). Retain 365 days for auth and payment audit trails, 90 days for web access logs. Use S3 with Object Lock for monthly log archives; configure lifecycle to move to Glacier after 30 days and enforce deletion only via retention expiry. Scenario 2 — 50-person SaaS: enable CloudTrail, VPC Flow Logs, and exfiltration detection. Forward logs to a managed SIEM (Splunk Cloud); use index lifecycle policies to freeze/archival older indexes and produce a signed export (checksum manifest) quarterly for auditors. Both scenarios should have a single "log owner" per source and an onboarding ticket with approvals recorded.
Risks of not implementing these controls
Without defined retention and format policies you risk losing evidence for investigations, failing legal holds, and inability to produce required logs during an audit — this can lead to regulatory fines, breach notification costs, and extended incident resolution time. Non-standard or poorly timestamped logs hamper correlation and increase mean time to containment. Lack of documented approvals or change history will make auditors question governance, potentially escalating findings to senior management or regulators.
Compliance tips and best practices
Operationalize compliance with these actions: (1) automate health checks that validate expected log rates and alert on dropped sources, (2) run periodic retention audits that validate lifecycle rules and attempt test restorations, (3) maintain a versioned schema registry for log formats, (4) sign log manifests (SHA256) and log access audit trails, (5) enforce RBAC for log stores and separate duties (admins who manage collectors should not be sole approvers), and (6) include retention and log-format requirements in procurement contracts for managed services. Provide auditors: policy, retention schedule, sample configuration snippets (rsyslog/Filebeat/lifecycle rules), a recent manifest and hash verification, and the onboarding/approval tickets.
Summary: meeting ECC 2-12-1 is a mix of governance and engineering — inventory and classify your sources, codify retention and format rules, implement secure collection and immutable storage, require documented approvals for onboarding and exceptions, and maintain artifacts (manifests, policies, tickets) auditors can review; for small businesses this can be achieved incrementally using built-in cloud logging, open-source collectors, lifecycle rules for archives, and a simple approval workflow that together deliver an audit-ready logging program.
