Skip to content
Gateways
ASD ISMISM-0634

ASD ISM 0634 Central logging for gateway security events

What it says

Security-relevant events for gateways are centrally logged, including:

  • data packets and data flows permitted through gateways
  • data packets and data flows attempting to leave gateways
  • real-time alerts for attempted intrusions.

Quoted as published. Everything else on this page is written by Control Stack.

In plain English

Log gateway security events centrally, including traffic permitted through the gateway, traffic trying to leave it whether allowed or blocked, and real-time intrusion alerts.

Security-relevant events for gateways must be centrally logged, meaning sent to the organisation's central event logging facility rather than kept only on each device. A gateway manages data flows between networks in different security domains. The logged events must include:

  • data packets and data flows permitted through the gateway
  • data packets and data flows attempting to leave the gateway, whether allowed out or blocked
  • real-time alerts for attempted intrusions.

Other security-relevant events, such as configuration changes, are covered too.

Why it matters

Gateways carry every data flow between your network and other security domains, so their records can reveal intrusion attempts and data leaving. Logged centrally, they can be analysed together and stay available for investigating an incident, even if a gateway device is wiped or compromised.

How to implement

  1. List every gateway component that records security events.

    Include firewalls, proxies, network intrusion detection or prevention sensors (NIDS or NIPS), mail relays, remote access concentrators and cloud-delivered gateway services. Check the list against the network diagrams and the gateway's system security plan, so nothing is missed.

  2. Log permitted traffic and every outbound attempt.

    Turn on logging for every allow rule, and for rules that block traffic leaving the gateway. If the default deny rule does not log, add an explicit final deny rule that does. Where devices can export flow records, send those too.

  3. Forward logs and intrusion alerts to the central facility.

    Send events to the centralised event logging facility (ISM-1405) as soon as possible after they occur (ISM-1983), encrypted in transit with ASD-approved cryptography (ISM-1984). Include alerts from the NIDS or NIPS that ISM-1030 places inside the outermost firewall, plus configuration changes and administrator actions.

  4. Get logs out of gateway services you buy.

    For a cloud-delivered or managed gateway, feed the provider's log export or API into your central facility, and check it includes permitted traffic, outbound attempts and intrusion alerts. If the service cannot supply one of these, the control is not met for it until that changes.

Keep it running

Alert when a gateway log source goes quiet, since a stopped feed looks the same as no activity. Recheck logging after any rule change, software upgrade or new gateway component, because a new rule or device may not log by default.

Evidence and audit

  1. Check 1 of 3What an auditor asks for

    Ask for

    The network diagrams, the list of gateway components and the log sources configured in the central facility.

    Look at

    Whether every gateway component in the diagrams, including cloud-delivered services, is a log source in the central facility.

    Good evidence

    Every gateway component in the diagrams sends its events to the central facility, with events present for the period sampled, and none logs only to its own console.

  2. Check 2 of 3What an auditor asks for

    Ask for

    Two test connections run during the audit: one the gateway permits, and one outbound attempt its rules block.

    Look at

    Whether both appear in the central facility and what each entry records.

    Good evidence

    Both the permitted connection and the blocked outbound attempt appear in the central facility, with time, source, destination and the device that handled each.

  3. Check 3 of 3What an auditor asks for

    Ask for

    A harmless test that triggers an intrusion alert, such as a known test signature sent past the NIDS or NIPS.

    Look at

    Whether the alert reaches the central facility, and how long it takes.

    Good evidence

    The alert is recorded in the central facility within moments of being raised, not only on the sensor's console or in a later batch.

Mappings

Closest match in ISO/IEC 27001: A.8.15, Logging, which this control partially meets.

ISO 27001

ISO/IEC 27001: 2 mappings

  • A.8.15LoggingPartially meets
  • A.8.16Monitoring activitiesSupports
Essential Eight

ASD Essential Eight: 2 mappings

Mappings show how controls relate across frameworks. They do not imply full equivalence or certification.