Beget Wiper Attack Erases Cloud VPS, S3 and Backups at Russian Host
Attackers reached the management layer of Russian hosting giant Beget and ran data-zeroing scripts that erased its VPS, S3 and Kubernetes cloud plus all VPS binary backups. Beget is rebuilding from scratch and says some data is gone for good.
Threat card
- Threat type
- Malware
- Disclosed
- Nov 2, 2026
- Affected
- Beget cloud platform: VPS, S3 storage, Kubernetes, CDN settings, IP configuration and all VPS binary backups; shared hosting and dedicated servers disrupted
The short version
- On the night of 1 November 2026 an attacker reached Beget's infrastructure management systems and ran scripts that zeroed server data.
- Beget's VPS, S3 and Kubernetes cloud, CDN settings and all VPS binary backups were destroyed; VPS file backups survived for roughly half of users.
- Beget found no sign of data download or modification and says the attack aimed only at destruction; no group has claimed it.
- More than 1,000 servers must be checked by hand for vulnerabilities and hidden scripts before going back online.
The Beget wiper attack is one of the most destructive incidents ever disclosed by a hosting provider. On the night of 1 November 2026 an attacker reached the management systems behind Beget’s whole architecture, one of Russia’s largest registrars and hosts, and ran data-zeroing scripts that erased its VPS, S3 and Kubernetes cloud along with every VPS binary backup. More than a week later Beget is still rebuilding from scratch. Any organisation whose backups share a control plane with production should treat this as a live-fire test case.
What happened
Attackers gained access to the systems Beget uses to manage its infrastructure and used them to execute scripts that overwrote server data across the platform. The Beget cyberattack destroyed cloud workloads, CDN settings, IP configurations and all VPS binary backups. Beget disconnected every server to stop the spread and says the goal was destruction, not theft.
Beget, founded in 2007, serves more than 200,000 customers, from online stores and media outlets to educational services, according to BleepingComputer. Citing Kommersant, The Record reports it was the second-largest .RU registrar in October, with about 540,700 domains and a 14.7% share. Hundreds of websites went down, including the Naryan-Mar city administration site and Russia’s Chestny Znak product labelling system.
Timeline
| Date (2026) | Event |
|---|---|
| 1 November (night, Sunday) | Attacker reaches Beget’s management layer and deploys data-wiping scripts; Beget detects the activity and disconnects all servers from the network. |
| 2 November | Russian media begin reporting widespread outages of sites hosted by Beget. |
| 3 November | Beget publishes a detailed Telegram statement confirming management-layer access and data-zeroing scripts. |
| 4 November (Wednesday) | Beget starts bringing some servers back online; customer-managed dedicated servers restored, shared hosting recovery under way. |
| 6 November | Beget publishes its assessment of VPS backups: file backups only for about half of VPS users. |
| 9 November | Beget reports 60% of shared hosting and 40% of dedicated servers restored; recovery continues. |
How the attack works
Beget has withheld the initial access method, citing security concerns and the ongoing investigation, so the first stage of the kill chain is unknown. The later stages are clear from the company’s statements and point to a control-plane compromise rather than a server-by-server intrusion.
Initial access and privilege
Not disclosed. What is confirmed is the end state: the attacker held access to the management systems spanning Beget’s entire architecture. Whether that came from stolen administrator credentials, a vulnerable orchestration component or an insider path has not been said, and no CVE has been linked to the incident.
Execution through the management plane
From that position the attacker ran scripts designed to erase data. Using a central management system to push one action to every host is what turned a single intrusion into a platform-wide outage. It is the same structural weakness we described in the N-able N-central Able2Shell case: whoever owns the management tool owns every managed machine.
Impact: data destruction and inhibited recovery
The scripts zeroed data across VPS instances, S3 object storage and Kubernetes services, and wiped CDN settings and IP configurations. Critically, VPS binary backups were destroyed as well. Only file-level backups partly survived, which Beget estimates will cover roughly half of VPS users, and it calls matching those backups to users “technically complex”. Beget’s verdict on VPS data is blunt: “Unfortunately, it is impossible to recover all data.”
No theft observed
Beget says it found no evidence that the attacker downloaded or modified customer data. That pattern, with no ransom note and no claim, fits a purely destructive operation. Beget is still urging customers to change credentials stored on affected systems.
MITRE ATT&CK mapping
| Tactic | Technique (ID) | How it was used here |
|---|---|---|
| Execution, Lateral Movement | Software Deployment Tools (T1072) | Infrastructure management systems used to run scripts across the platform. |
| Execution | Command and Scripting Interpreter (T1059) | Data-zeroing scripts executed on servers. |
| Impact | Data Destruction (T1485) | VPS, S3 and Kubernetes data, CDN settings and IP configurations erased. |
| Impact | Inhibit System Recovery (T1490) | All VPS binary backups destroyed, leaving only partial file backups. |
Who is behind it
Not yet attributed. No hacking group has publicly claimed the attack, and Beget has not named a suspect. It has notified police and government agencies and is working with cybersecurity firms. For context, The Record notes that the EU sanctioned Beget and a co-owner earlier this year for providing digital infrastructure to Russian occupation authorities in Ukraine and defence-related entities, and that another Russian host, FirstVDS, was breached in October. Neither fact establishes who carried out this attack.
Affected services and impact
| Service | Status reported | Recovery outlook |
|---|---|---|
| Cloud VPS | Data wiped; binary backups lost | File backups for about half of VPS users; Beget says not all data can be recovered |
| S3 storage and Kubernetes | Data wiped | Platform being rebuilt from scratch |
| CDN and IP configuration | Settings wiped | Being rebuilt |
| Shared hosting | Disrupted | Nearly all sites likely recoverable; 60% restored as of 9 November |
| Dedicated servers | Disrupted | 40% restored as of 9 November |
About 1,000 servers must be inspected manually for vulnerabilities and hidden scripts before they return. Beget says it now treats all infrastructure as potentially infected and is reworking every weak point, including ones it had rated low risk. It has also said the incident may qualify as force majeure under Russian law.
Indicators of compromise
No IOCs, script hashes or infrastructure have been published by Beget or the security firms assisting it. Any release is most likely to come through Beget’s official Telegram channel. In the meantime, defenders running similar hosting or private cloud platforms should hunt for these behaviours:
- Scripts or jobs pushed from orchestration, configuration management or hypervisor consoles that call disk or volume overwrite utilities (for example
ddreading from/dev/zero) or bulk delete storage objects. - Management-plane sessions from new source IPs, at unusual hours, or by accounts that rarely run fleet-wide jobs.
- Deletion or truncation of backup repositories, snapshot schedules or retention policies shortly before or during a mass job.
- New or modified scheduled tasks and startup scripts on hypervisors and storage nodes, which is the category of hidden script Beget is now checking for.
Detection and hunting
Instrument the management plane as a crown-jewel system. Forward orchestration and automation audit logs (job submissions, target host counts, command bodies) to a SIEM outside the platform they manage, and alert when a single identity issues destructive commands to many hosts within minutes. On Linux hypervisors and nodes, auditd rules on execution of dd, shred, wipefs and blkdiscard against block devices give early warning; on Kubernetes, watch the API audit log for bulk delete verbs on namespaces, persistent volumes and persistent volume claims.
For object storage, alert on spikes in delete operations and on changes to versioning or object lock settings. Backup platforms should log and alert on job deletions and retention changes, ideally to a separate tenant. Our coverage of the Qubole AWS breach shows the same lesson from the theft side: privileged platform identities need their own monitoring.
Mitigation and response
- Beget customers: change passwords, access keys and any credentials stored on affected systems, as Beget advises, and check restored sites for unexpected files.
- Restore from your own off-provider copies where you have them instead of waiting on provider recovery.
- Keep at least one backup copy immutable and outside the reach of production management credentials, with separate identity and separate approval for deletion.
- Lock down orchestration and hypervisor management: phishing-resistant MFA, just-in-time privilege, and two-person approval for fleet-wide or destructive jobs.
- Rate-limit or require confirmation for commands that target large numbers of hosts or volumes at once.
- Rehearse a full-environment rebuild, including backup-to-customer mapping, which Beget has called technically complex.
Analyst take
Beget is a reminder that the management plane is the single point of failure in any cloud. The attacker did not need to break into 200,000 customer workloads; one foothold in the systems that administer them was enough, and losing binary backups alongside production turned an outage into permanent data loss for some customers. Backups that can be deleted with the same access that runs production are copies, not backups.
Whoever is behind it, a destructive attack with no ransom demand and no data theft points to disruption as the goal. Hosting providers, MSPs and private cloud operators should worry most. If an intruder reached your orchestration layer tonight, ask how many hosts one job could touch and whether your last good backup would survive it.
What to do now
0 of 6 done
Questions readers ask
What happened to Beget?
An attacker gained access to Beget's infrastructure management systems on 1 November 2026 and ran scripts that wiped its cloud platform, including VPS, S3, Kubernetes and VPS binary backups.
Can Beget customers recover their VPS data?
Beget says not all data can be recovered; VPS file backups should be available for about half of VPS users, while binary backups were lost entirely.
Who was behind the Beget attack?
No group has claimed responsibility and Beget has not disclosed the initial access method, citing the ongoing investigation.
Sources
One short email when something serious breaks.
Threatspedia covers major cyber threats only, verified against primary sources. Read our editorial policy, or report a correction.


