Cyber Resilience, Incident Response & Crisis Management, 10 min read

Ready reckoner

Why cyber is different

Adversaries target recovery, spread (replication, shared identity), wait (dwell time), return (reinfection) and extort (data leaks). Fast failover to an identical copy can make things worse.

NIST SP 800-160 Vol. 2 Rev. 1

Goal Objectives
Anticipate prevent/avoid · prepare · understand
Withstand continue · constrain
Recover reconstitute · understand
Adapt transform · re-architect

14 techniques. Start with: redundancy · diversity · segmentation · privilege restriction · non-persistence · substantiated integrity.

SEBI CSCRF goals: anticipate · withstand · contain · recover · evolve; functions aligned with CSF (governance, identify, protect, detect, respond, recover).

Assume-breach questions: What can an admin not destroy? How are customers served if prod is encrypted? Which data is trustworthy? Can prod reach DR? How do we rebuild without reintroducing the attacker?

Ransomware

Chain: access → escalate/spread → kill backups → exfiltrate → encrypt (often weeks).

Patterns: immutable isolated backups · tiered admin + PAM · segmentation · EDR with tamper protection · golden images / IaC · out-of-band comms · offline plans · business workarounds at MBCO.

Recovery sequence: contain (preserve evidence) → identity (rebuild, reset privileged, rotate secrets) → core infra + security tools → known-good data (clean room, validate) → Tier 1 at MBCO → rest. Estimated times in the plan.

Ransom: policy and authority decided in advance; payment does not guarantee decryption or deletion; legal, sanctions and law-enforcement input. Plan for data-leak extortion.

Detection and SOC

Regulator Expectation
RBI C-SOC for banks; proportionate for NBFCs/UCBs
SEBI CSCRF SOC by category; Market SOC or third-party SOC for smaller REs
CERT-In 180-day logs in India · NTP sync

Priority use cases: backup deletion · security tools disabled · logging changed · mass data change · privilege escalation · cloud IAM changes · payment anomalies. Logs and monitoring outside what they monitor. MSSP contract: SEV-1 escalation time (e.g. 15 min). Threat intel: CERT-In, CSIRT-Fin, IB-CART, regulator alerts, NCIIPC → record actions. Metrics: MTTD · MTTR · coverage · use-case coverage · true-positive rate.

Incident response (SP 800-61 Rev. 3, April 2025)

Organised around CSF 2.0: Govern + Identify + Protect = preparation · Detect · Respond (manage, analyse, report, mitigate) · Recover · lessons learned throughout.

One severity scheme (SEV-1 to SEV-4) across ops and security → escalation · reporting clock · BC invocation. Easy to escalate; downgrade later.

Playbooks: ransomware · DDoS · data breach · privileged account compromise · supplier compromise · payment fraud. Each: triggers · first 15/60 min · decisions · contacts · reporting · when to invoke BC.

Three streams in parallel: technical IR · BC plans · crisis team. Shared log. Ask "is DR clean?" before failover.

CCMP

Governance (declare, decide) · scenarios/playbooks · reporting · communication · coordination (law enforcement, CERT-In, forensics, legal, insurer, suppliers) · recovery + learning. Board-approved; linked to BCP.

Reporting matrix (check current instruments)

To Deadline
CERT-In 6 h from noticing
RBI (DAKSH) 6 h from detection
SEBI 6 h from detection
IRDAI copy of CERT-In report within 6 h
DPDP Board + individuals without delay; detailed to Board within 72 h
Customers of a TSP per contract, often 2–4 h
NPCI / payment operators · NCIIPC · law enforcement · insurer per rules / policy

Clock starts at detection. Report early, mark preliminary, update. One fact base. Portals pre-registered; credentials and templates offline; named submitters + deputies.