bastion

Incident Response Fundamentals

Build this skill

Sequence the phases of incident response and decide the right containment, triage, eradication and recovery steps for a live ransomware incident.

Saturday, 4:40 pm: the counter laptop starts renaming files

At 4:40 on a busy Saturday, Arjun Pawar notices that the files on the Nashik Road counter laptop are changing names one after another. Each one now ends in .lotus, and none of them will open. What Tulsi Health does over the following hour decides whether this stays one damaged laptop or becomes fourteen closed stores. An incident is any event that harms, or seriously threatens, an organisation's information or systems: keeping data secret, keeping it correct, or keeping it available when people need it. Incident response is the planned, practised way an organisation handles one, so that people act in a sensible order instead of panicking.

Not every odd thing is an incident. A single failed login, a jammed printer or a slow website is an event: something that happened and may be worth a note in a log, which is a computer's automatic diary of what happened and when. Most events are harmless. An event becomes an incident when it actually threatens data or systems, and files being scrambled on a billing laptop clearly does.

IF YOU SEE AN INCIDENT 1. Unplug the cable 2. Do not switch off 3. Call IT: ext. 200 4. Write down the time 4:40 pm Arjun Neha, the files are changing names! Files being renamed… stock_feb.xlsx.lotus bill_4471.pdf.lotus orders.db.lotus renaming 2,140 files… ✕ Cable pulled power left on Tulsi Health · Nashik Road branch

Arjun did not have to invent a response. The card on the wall, written months earlier, told him the first three things to do.

The fire brigade way of working

A good fire brigade does not buy its hoses after the phone rings. It prepares in advance, receives the alarm, finds out how big the fire is, stops it spreading, puts it out, helps people rebuild, and then studies what went wrong. Incident response follows the same shape. A widely used version, published by NIST, the US national standards body, groups the work into four phases: preparation; detection and analysis; containment, eradication and recovery; and post-incident activity, where lessons are learned. The phases form a loop, because every incident should leave the next response better prepared.

Prepareplan, contacts, drillsDetect, analyseis it real? how big?Contain, eradicatestop spread, remove causeRecoverback to normal, carefullyLearnreview, then improveIncident response

The third phase is split into two steps here to show its order: first stop the damage, then remove its cause, then rebuild. Learning feeds straight back into preparation.

Before reading the details, try putting the phases in order yourself for the Nashik Road incident. Each card shows what that phase produced on the day, so ask what each step needs from the one before it.

Preparation happens before the alarm

Incident preparation is everything done in calm times so that the bad day goes better. For Tulsi Health it means a short written plan that says who leads the response (Neha), who can make business decisions such as closing a store (Vikram), and who must be told. It means a contact list kept somewhere other than the computers that might be locked, an offline backup copy that an attacker on the network cannot reach, and logs that record logins and file changes. It also means the card at every counter, telling staff what to do in the first two minutes. Arjun did not have to think at 4:40 pm; he followed the card.

Preparation also means practice. A one-hour drill, where the team sits around a table and talks through an imaginary attack step by step, shows the gaps while they are still cheap to fix. Does Sameer know whom to call? Does anyone remember where the backup is kept? A drill that finds those answers missing is worth far more than discovering it during a real attack.

Detection and analysis: is it real, and how big?

Detection is noticing that something is wrong. Signs arrive from many places: an alert from security software, a strange entry in a log, a customer complaint, or a staff member like Arjun reporting what is in front of them. Many signs turn out to be false alarms, so analysis follows. It confirms whether this is really an incident, what kind it is, which systems are touched and whether it is still spreading. Neha checks the renamed files and the note left on the screen and confirms ransomware: malware, meaning harmful software, that scrambles files and demands payment to unscramble them.

Next comes triage, a word borrowed from hospital casualty wards. When three patients arrive together, the nurse at the door does not treat them in order of arrival. She asks who is in most danger and how quickly things will get worse. Incident triage does the same, judging each incident by its impact (how much harm to data, money, customers or safety) and its urgency (how fast it spreads or worsens). Together these set a severity level, and the severity decides who is called and how fast.

HighA billing laptop stolen from a lockedstore overnightCriticalRansomware spreading toward the shareddriveLowOne failed login from a known storelaptopMediumPhishing email reported by five staff,nobody clickedUrgency: how fast it worsens →Impact: how much harm →

The Nashik Road ransomware lands top right: high harm and spreading by the minute. That rating is what makes Neha call Vikram at once instead of on Monday.

Containment and eradication: stop the spread, then remove the cause

Containment means stopping the damage from spreading, even before you fully understand it, like firefighters soaking the house next door. On the laptop, short-term containment means pulling its network cable and turning off its Wi-Fi, so it can no longer reach the shared drive or the other laptops. Leaving it switched on matters. The computer's memory, its short-term working space that empties when the power goes off, may hold clues about how the attacker got in. Wider containment might mean pausing the Nashik Road accounts' access to the shared drive, or blocking a suspicious internet address at the firewall, the gatekeeper device between the store network and the internet.

Eradication removes the cause so that the incident cannot simply restart. That means finding how the attacker got in and closing that door: removing the malware (usually by wiping and reinstalling the laptop rather than trying to clean it), changing any password the attacker stole, and checking other machines for the same signs. Skipping eradication is a classic failure. If Neha restores the laptop but the attacker still holds Arjun's stolen password, they can walk straight back in the following night.

ContainPull the network cableLeave the power onPause shared-drive accessBlock the attacker's addressEradicateFind how they got inWipe and reinstallReset stolen passwordsCheck the other laptopsRecoverRestore a clean backupTest before reconnectingBring back in stagesWatch closely for weeks

Each column depends on the one before. Recovering before eradicating simply rebuilds the house with the fire still smouldering inside.

Recovery: back to normal, carefully

Recovery returns systems to normal working. Neha restores the laptop's files from a backup taken before the attacker first got in, not merely the latest one, because a recent backup may already contain the malware. She tests the restored laptop before reconnecting it, brings systems back one at a time, and watches the logs closely for a few weeks in case the attacker tries again. Only when everything is stable is the incident closed, followed by a review meeting where the team asks what happened, what worked and what to change. That review feeds straight back into preparation, which closes the loop.

Now play Neha for the evening. The incident unfolds over five decisions, and a wrong first pick in any round counts as an incident in Tulsi Health's year, so before choosing, ask what each option protects and what it destroys.

Follow your curiosity.

↑ ↓ to explore · Enter to open · Escape to close