Skip to content
User application hardening
Essential EightE8-AH-ML3.4

Essential Eight AH-ML3.4 Analyse event logs from non-internet-facing servers for cyber threats

What it says

Event logs from non-internet-facing servers are analysed in a timely manner to detect cybersecurity events.

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

In plain English

Analyse event logs from servers that cannot be reached from the internet, such as domain controllers and file servers, in a timely manner to detect cyber security events.

At Maturity Level Three, event logs from non-internet-facing servers must be analysed in a timely manner to detect cyber security events, meaning signs that something may be wrong. These are servers that cannot be reached directly from the internet, such as domain controllers, file servers and internal application servers. Under user application hardening, the logs include the PowerShell and command line process creation events that E8-AH-ML2.11 and E8-AH-ML2.12 require to be centrally logged. E8-AH-ML2.14 covers internet-facing servers. The statement sets no time limit, so the organisation decides what timely means.

Why it matters

Intruders who get inside often move to internal servers such as domain controllers, using PowerShell and command line tools there to take wider control. Timely analysis of these servers' logs gives a chance to notice that activity before it reaches the whole network.

How to implement

  1. List every non-internet-facing server and connect its logs.

    Compare the directory, virtualisation and cloud consoles and a network discovery scan with the log sources in your central logging platform. Connect any server that is missing; where a vendor-managed server cannot run your agent, use built-in Windows event forwarding or have the vendor send its logs.

  2. Confirm the PowerShell and command line events arrive.

    Check each server sends script block logging (event 4104), module logging (event 4103) and process creation events with command lines (event 4688), plus transcripts, as E8-AH-ML2.11 and E8-AH-ML2.12 require.

  3. Write detection rules for misuse on these servers.

    Look for encoded or obfuscated PowerShell, commands that download files, and administration tools run by accounts that do not normally use that server. Test that each rule fires.

  4. Set and meet a time to analysis.

    The statement sets no limit, so decide how quickly alerts from these servers are reviewed, sooner for domain controllers and other servers that control access, and measure the delay from the event itself, including log arrival. One process can serve the same requirement under other strategies (for example E8-AC-ML3.4).

Keep it running

Watch for servers that stop sending logs or send them late, and keep detection rules current. Recheck coverage whenever servers are built, migrated or retired, and after any incident on these servers that was found some other way.

Evidence and audit

  1. Check 1 of 3What an auditor asks for

    Ask for

    An independent list of non-internet-facing servers, such as from the directory, virtualisation consoles or a discovery scan, and the log sources in the central logging platform.

    Look at

    Whether every listed server sends PowerShell and command line process creation events, and whether any stopped sending during the period reviewed.

    Good evidence

    Every non-internet-facing server on the independent list, domain controllers included, is sending these events, and any server that stopped sending during the period was noticed and reconnected.

  2. Check 2 of 3What an auditor asks for

    Ask for

    A test on a sample of servers, including a domain controller, such as a harmless encoded PowerShell command run without warning the analysts.

    Look at

    Whether each test was detected and reviewed by an analyst, and how long that took from when it ran.

    Good evidence

    Every test was detected and reviewed by an analyst within the organisation's set time, measured from when it ran.

  3. Check 3 of 3What an auditor asks for

    Ask for

    Alert and review records for these servers over a recent period, and the incident register.

    Look at

    How long alerts waited, whether any were closed unread, and whether any incident on these servers was first found another way.

    Good evidence

    Alerts were reviewed within the set time with none closed unread, and no incident on these servers was first noticed by another route while its events sat unreviewed in the logs.

Mappings

Closest match in ASD Information Security Manual: ISM-1907, Timely analysis of non-internet-server logs, which asks for the same thing.

ISO 27001

ISO/IEC 27001: 2 mappings

  • A.8.15LoggingPartially meets
  • A.8.16Monitoring activitiesPartially meets
ASD ISM

ASD Information Security Manual: 9 mappings

ASD Essentials

ASD Essentials for enterprise IT: 2 mappings

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