BackupRansomwareSecurity

Backup Strategy: The 3-2-1 Rule and Protection Against Ransomware

May 10, 2025 · 9 min read · System Networks

94% of companies that lost data for more than 10 days went bankrupt within a year. 57% of ransomware attacks specifically target and destroy backup copies before encrypting primary data. Having a backup is not enough — you need the right backup strategy.

The Uncomfortable Reality

57%of ransomware attacks target backup systems before encrypting primary data
93%of companies that lost their datacenter for 10+ days filed for bankruptcy within a year
34%of backup restores fail the first attempt due to untested backup integrity
₽12Maverage cost of a ransomware incident for a mid-size Russian company (ransom + downtime + recovery)

The 3-2-1 Rule Explained

3

Three copies

1 primary + 2 backups. If you have one backup and it fails during restoration, you have nothing.

2

Two different media

Store backups on at least two different types of storage: e.g., local disk + object storage. Protects against media failure.

1

One offsite copy

At least one copy must be geographically separate. Protects against fire, flood, theft, and physical server seizure.

3-2-1 Is Not Enough Against Modern Ransomware

Modern ransomware (like LockBit, BlackCat, Cl0p) is designed by sophisticated criminal organisations. They have learned that victims restore from backup rather than paying. So they target backups first. The 3-2-1 rule needs to be extended:

3-2-1-1: Add one immutable copy

Immutable backup storage cannot be modified or deleted for a set period — even by an admin with credentials. Ransomware actors who compromise your backup server cannot delete immutable copies. S3 Object Lock, Wasabi, Backblaze B2 Object Lock all provide this.

3-2-1-1-0: Zero trust for backup infrastructure

Backup servers should not be on the same network segment as production. Use separate credentials for backup systems. Consider air-gapped backup (physically disconnected) for the most critical data.

Recovery Testing: The Most Neglected Part

⚠️ A backup you have never tested is not a backup

34% of backup restores fail the first attempt. Bit rot, software version mismatches, missing dependencies, incomplete backups — these only surface during actual restoration. By then it is too late.

Monthly: restore a random non-critical server from backup and verify it functions
Quarterly: full DR exercise — restore critical systems and test application functionality
Annually: full simulation of ransomware scenario — isolate, restore, verify recovery time
Document RTO (Recovery Time Objective) and RPO (Recovery Point Objective) for each system

Storage Options for Offsite Backup

OptionCostImmutabilityBest for
Colocation in EU DC₽2,000–₽5,000/TB/moPhysicalLarge volumes, 152-FZ sensitive data
Backblaze B2$6/TB/moObject LockCost-efficient offsite, SMB
Wasabi$7/TB/moObject LockNo egress fees, good for recovery testing
AWS S3 Glacier$4/TB/moObject LockArchival, infrequent access
Yandex Object Storage (Russia)₽2,000/TB/moPartial152-FZ localisation requirement
Tape (LTO-9)High CAPEX, low/moAir-gappedMaximum protection, large DCs

Backup infrastructure in EU and Russia

Offsite backup storage in Frankfurt and Prague →

Immutable storage · Separate from production · 152-FZ compliant · From €89/month

Начнём работу

Защитите инфраструктуру до того, как это потребуется

Бесплатная предварительная консультация. Оценка текущего уровня защиты за 1 день. Первые результаты — в течение недели.