Guide 6 of 730 days
The server has been breached. Who do you tell, and by when?
Discovery starts several clocks at once, and they do not all belong to the same regulator or the same set of businesses. This page puts them in order from the moment you become aware, and names who each one binds.
General information, not legal or security advice.
Four clocks from one moment of discovery
The first clock is ASD's own, and it has no fixed number. At Maturity Level Two of the Essential Eight, the control is that cyber security incidents are reported to ASD as soon as possible after they occur or are discovered. ASD's small-business guidance gives the same advice: all cyber security incidents should be reported through the ReportCyber portal or by calling 1300 CYBER1 (1300 292 371).
The second clock is the Privacy Act's, and it runs for 30 days. Where a business covered by the Act becomes aware of grounds to suspect an eligible data breach, the entity must carry out a reasonable and expeditious assessment of whether there are reasonable grounds to believe the circumstances amount to an eligible data breach, and take all reasonable steps to ensure the assessment is completed within 30 days after the entity becomes aware. If the assessment confirms an eligible breach, the duties follow: businesses covered by the Privacy Act must report eligible data breaches to the Office of the Australian Information Commissioner, and, subject to some exceptions, notify the individuals who are likely to suffer serious harm as a result. ASD's plainer advice: businesses must understand and meet their reporting obligations if a breach involves customers' personal information, and affected customers or users should also be informed if their personal data may have been compromised.
The third clock is 72 hours, and it applies once a ransomware or cyber extortion payment has been made, by the business or on its behalf. Under the ransomware payment reporting rules, a reporting business entity has an obligation to report to the Government if it has made a ransomware or cyber extortion payment, or is aware that another entity has made one on its behalf, within 72 hours. The obligation attaches to an entity that is a reporting business entity, is impacted by a cyber security incident, and has provided, or is aware another has provided on its behalf, a ransomware payment to an entity seeking to benefit from the impact or the incident. The 72 hours also bounds what is expected of the report itself: information is only required to the extent the entity knows or is able, by reasonable search or enquiry, to find out within the 72 hour period. ASD's own advice on the payment itself is blunt: do not pay the ransom, as there is no guarantee your files will be recovered.
The fourth clock belongs to critical infrastructure, and it is the fastest. In the Cyber and Infrastructure Security Centre's guidance, notification of cyber security incident reporting is the obligation to report critical or other cyber security incidents that have an impact on critical infrastructure assets to ASD's Australian Cyber Security Centre. A critical cyber security incident, meaning an incident that has occurred or is occurring and has had, or is having, a significant impact on the availability of the asset, must be reported within 12 hours of becoming aware, and if the report is made verbally, a written record within 84 hours of the verbal report. An "other" cyber security incident, meaning one that has occurred, is occurring or is imminent and has had, is having, or is likely to have a relevant impact, must be reported within 72 hours, with a written record within 48 hours of a verbal report. Two working definitions keep the clock honest. Unauthorised does not mean intentional: an accident that results in unauthorised access, modification or impairment is still a cyber security incident. And awareness is not an alert: a responsible entity does not become aware merely because an automated system detected an incident, but when there has been confirmation that it took place, or when the entity should have become aware had it followed its monitoring and escalation processes. Late is still better than never, in the guidance's own words: if the timeframe has been missed, report anyway, because late reports are considered less egregious than no report at all.
Which of these clocks are yours depends on which entity you are in each law's terms. The Privacy Act's clock runs for businesses the Act covers; the report within 72 hours runs for a reporting business entity; and the 12-hour and 72-hour clocks run only for a responsible entity for a critical infrastructure asset. One related duty sits on the other side of the counter: entities must notify external data service providers if they are storing or processing business critical data for a critical infrastructure asset, so the provider knows it may carry obligations too.
Who has to report
The ransomware payment rules and the reporting form disagree about one word, and it is the word that decides who is caught. The rules say that, generally, an entity will only be a reporting business entity if it is either a responsible entity for a critical infrastructure asset to which Part 2B of the Security of Critical Infrastructure Act 2018 applies, or is carrying on a business in Australia with an annual turnover for the previous financial year that exceeds the turnover threshold, and the threshold is set elsewhere in the same instrument: for the purposes of the Act, the amount of turnover threshold for a business for the previous financial year is $3 million.
The reporting form puts the same category in slightly different words: an entity carrying on business in Australia with an annual turnover for the previous financial year that is equal to or exceeds $3 million. The form also carries the other category the rules name, in their words: a responsible entity for a critical infrastructure asset to which Part 2B of the Security of Critical Infrastructure Act 2018 applies.
The rules' word is "exceeds". The form's is "equal to or exceeds". If your turnover for the previous financial year is at or near $3 million, keep the two side by side and ask your adviser which one applies to you.
You can only report what you recorded
Every clock above depends on the same raw material, which is the record the machine kept. Event logging keeps a record of important system and user activities to help identify problems and investigate security incidents, and it is what lets a small business detect threats, respond quickly and understand what has happened across its systems. ASD's advice for keeping it useful is to focus on key security events rather than collecting large volumes of data, and to ensure the logs are accurate, secure and reviewed regularly.
The retention question is the one that decides how far back an investigation can see. Organisations should ensure they retain logs long enough to support cyber security incident investigations, and default log retention periods are often insufficient for that job. The reason is arithmetic: the longer logs can be kept, the higher the chances of determining the extent of a cyber security incident, including the potential intrusion vectors that require remediation. On a single server the practical form of that is a short policy with a date in it: what is logged, where it is kept, for how long, and who reads it.
Two protections are worth wiring in at the same time. Logs should live somewhere they can be trusted after the fact: the storage of logs should be in a separate or segmented network with additional security controls, to reduce the risk of logs being tampered with in the event of a network or system compromise, and at Maturity Level Two of the Essential Eight, event logs are protected from unauthorised modification and deletion. And they should be read: centralise, secure and retain event logs to support monitoring and investigations, prioritise collection of high-value logs from critical systems, cloud services and network infrastructure, and analyse cyber security events in a timely manner to identify incidents. One last seam to leave visible: the logging policy should take into account any shared responsibilities between service providers and the organisation, because on a rented machine some of that logging may fall on the provider's side of the split, so it is worth asking about before the day it is needed.
Where to report
This page deliberately stops at the doors, because the steps themselves belong to the agencies that set them. A written report goes to cyber.gov.au/report, and a verbal report to 1300 CYBER1 (1300 292 371). ASD is clear about the one thing that outranks the paperwork: if there is a threat to life or risk of harm, call 000 immediately. One practical warning from the same guidance, for anyone filling in a report while an incident is live: do not complete the form on any network you believe has been compromised, and use a separate system and contact details instead.
Keep the doors distinct in your head. The ReportCyber portal takes every cyber security incident, while the Privacy Act's assessment runs on its own 30-day clock, and under the NDB scheme an eligible data breach brings, subject to some exceptions, a duty to notify the individuals likely to suffer serious harm, and the critical infrastructure reports run to ASD's ACSC on the hours above. A single breach can open more than one of these doors, so check each one.
The whole stack, bottom to top (dedicated-servers.com.au home)