Executive summary
Ransomware is no longer only a malware problem—it’s an identity + data + business continuity problem. Winning means designing for resilience, not perfection:
-
Prevent common entry points.
-
Detect abnormal behaviors early.
-
Contain fast to stop lateral spread.
-
Recover from immutable backups with tested playbooks.
This guide is a practical, India-first blueprint you can adopt today—whether you’re an MSME or an enterprise SOC.
Threat model (how modern ransomware wins)
Initial access: phishing, exposed RDP/VPN, vulnerable edge apps, malicious ads, supply chain.
Post-compromise: scriptable tools (PSExec/WMI), living-off-the-land, credential theft, GPO abuse.
Objectives: data theft → encryption → extortion (sometimes encryption-less extortion only).
Key MITRE ATT&CK: T1566 (Phishing), T1078 (Valid Accounts), T1021 (Remote Services), T1486 (Encrypt), T1047 (WMI), T1059 (Command & Scripting), T1110 (Brute Force), T1489 (Service Stop), T1490 (VSS deletion), T1071 (Exfil), T1484.001 (GPO).
Phase-by-phase defense (what to do)
1) Identity & Access
-
FIDO2/WebAuthn MFA for admins and finance/HR. Disable legacy/basic auth.
-
Just-in-Time privileged roles; vault and rotate break-glass accounts.
-
Conditional Access (device health, location, risk).
-
Service principal hygiene in cloud (expire secrets; least privilege).
2) Email & Web
-
DMARC (p=reject), SPF, DKIM, MTA-STS, TLS-RPT.
-
URL detonation / sandbox; block macros from the Internet; banner warn external senders.
-
DNS layer security (block newly registered domains, malware C2s).
3) Endpoints (Windows/Linux/macOS)
-
EDR everywhere (tamper protection on).
-
Windows hardening:
-
ASR rules on (block credential theft, Office child-processes, script obfuscation).
-
Controlled Folder Access for business data paths.
-
LSA protection, Credential Guard, LSASS as PPL.
-
AppLocker/WDAC allow-listing for high-risk roles.
-
Disable SMBv1; restrict RDP (NLA + VPN/jump host only).
-
-
Linux:
-
Mount shares with
noexec,nodev,nosuid; auditd for chmod/chattr on backups; -
restrict SSH (keys + MFA), fail2ban; EDR/AV for Linux enabled.
-
4) Network & Segmentation
-
Tiered admin model; isolate domain controllers, backup servers, hypervisors.
-
Block RDP/SMB from the Internet; default-deny east-west where possible; egress allow-lists.
-
LAPS or equivalent for local admin password rotation.
5) Data Protection & Backup
-
Follow 3-2-1-1-0: 3 copies, 2 media, 1 offsite, 1 immutable/air-gapped, 0 errors after restore test.
-
Object-lock/immutability for cloud backups; separate credentials & IAM boundaries.
-
Back up SaaS (M365/Google) and config (GPOs, device policies), not just files.
-
Quarterly restore drills with measured RTO/RPO.
6) Cloud & SaaS
-
Disable legacy auth; Conditional Access; session controls (sign-in risk, device compliance).
-
Safe Links/Safe Attachments (or vendor equivalent); OneDrive/SharePoint versioning & retention.
-
Protect service accounts; alert on mass download/sharing.
7) OT/ICS (if applicable)
-
Strict segmentation (conduits/zones), vendor access via jump hosts + session recording.
-
Controller backups and gold images; allow-listing on HMI/engineering workstations.
Early detection & hunting (practical)
High-signal events to alert on (Windows):
-
Shadow copy deletion:
vssadmin delete shadows,wmic shadowcopy delete,diskshadow /s. -
Backup tamper:
wbadmin delete catalog, stopping backup/AV/EDR services. -
System sabotage:
bcdeditsafe-mode or recovery config changes; Event ID 1102 (audit log cleared). -
Mass file changes with unusual extensions; rapid spikes in failed file opens on shares.
-
LOLBins for staging/exfil:
rclone,7z/WinRARwith passwords,bitsadmin,certutil.
Elastic KQL (examples):
-
Process creates shadows deletion:
-
Suspicious archiving to network share:
Microsoft Sentinel (KQL):
Splunk SPL:
Rapid response playbook (when alarms fire)
0–15 min – Confirm & contain
-
Open a case; snapshot volatile data; isolate endpoints (EDR network containment).
-
Disable suspected accounts; revoke tokens/sessions; block C2 domains/IPs.
-
Stop propagation: temporarily block RDP/SMB between segments if needed.
15–60 min – Scope & kill
-
Hunt for GPO changes, new scheduled tasks, Run keys, WMI subscriptions.
-
Remove web shells / persistence; rotate creds (admins first), reset service principals.
-
If encryption started: keep encrypted hosts isolated; do not wipe yet—collect artifacts.
1–24 h – Eradicate & recover
-
Patch/close the entry vector (phish user, VPN, edge CVE).
-
Restore only from clean, immutable backups; re-introduce in rings; monitor for re-infection.
-
Report per Indian regulations: CERT-In (time-bound notification) and DPDP obligations; coordinate legal/PR.
After – Lessons & hardening
-
Root-cause report; new detections; tabletop the same scenario next week.
Hardening snippets (copy-paste starters)
Windows PowerShell – Attack Surface Reduction (common high-value rules):
Block macros from the Internet (GPO):
User Config → Admin Templates → Microsoft Office → Block macros from running in Office files from the Internet = Enabled
Linux – protect backup indexes from tamper:
Samba share – reduce blast radius:
Governance for India (pragmatic notes)
-
CERT-In: prepare contact list and reporting template; keep logs for 180 days by default.
-
DPDP Act: limit data collection; document lawful purpose; have breach notification playbooks.
-
Maintain vendor SLAs for IR support and forensics (time-to-engage ≤2h).
MSME “starter stack” (cost-conscious)
-
EDR (business-grade) + DNS filter + email security + backup with immutability (object lock / WORM).
-
Cloud SSO with MFA; disable legacy auth; device compliance before access.
-
Monthly restore test, quarterly tabletop exercise.
-
Run autopatching with rings (pilot → IT → all).
Metrics that matter
-
MTTD/MTTR for ransomware behaviors (VSS delete, mass encryption).
-
Patch latency for internet-facing CVEs.
-
% endpoints with ASR enabled & EDR healthy.
-
Backup success rate and restore success time.
-
Phishing click-through and report rate.
0–7 / 30 / 60 / 90-day roadmap
0–7 days: Enforce MFA, block macros-from-Internet, disable legacy auth, close exposed RDP, EDR everywhere, snapshot/validate backups, add VSS-deletion alerts.
30 days: Segment critical servers; LAPS; ringed patching; centralize logs; tune detection rules.
60 days: AppLocker/WDAC for finance/ops; immutable backup & offsite copy; SaaS backup; privileged access workstation for admins.
90 days: IR tabletop with executives; purple-team tests of encryption & exfil; automate restore runbooks.
Final word
Ransomware defense is a race against time. If you can detect early, contain within minutes, and restore with confidence, you win—no matter how the payload looks.
Need this as a one-page checklist for your team? It’s included below.
One-page Checklist (print & stick on the wall)
-
MFA (FIDO2) for admins & finance/HR
-
Legacy/basic auth off
-
EDR on 100% endpoints (tamper protection on)
-
ASR rules + Controlled Folder Access
-
RDP/SMB not exposed to Internet; VPN/jump hosts only
-
Segmented DC/backup/hypervisor networks
-
3-2-1-1-0 backups; immutability + monthly restore test
-
Alerts: VSS delete, backup tamper, log clear, mass encryption, archiver to shares
-
Cloud: Safe Links/Attachments, versioning & retention, OAuth app review
-
Incident kit: isolation steps, contacts, CERT-In template, PR/legal counsel
-
Quarterly tabletop; update runbooks after every exercise/incident
