Cybercriminals Attack VPS to Access Business Email Systems
Briefly

Malicious actors exploit anonymous VPS hosting to hijack live email sessions and gain persistent access to business mailboxes. Attackers obtain clean ASNs and fresh IPs to appear trusted and bypass traditional security, riding active sessions rather than relying solely on harvested passwords. Compromise tactics include changing inbox rules, stealing tokens, resetting passwords, and removing traces. Detection and prevention require freezing sessions (not users), making inbox rules visible and attested, alerting on rule churn, scoring infrastructure by volatility and provenance, and blocking remote tools by context. Autonomous containment and stronger controls around initial login events reduce adversary dwell time.
Attackers now rent trust. Five dollar VPS nodes buy entry to your allow list and they accomplish this by getting a clean ASN and fresh IP making traffic feel like a trusted source, not a criminal. In this case, the adversary is riding live sessions and no longer just harvesting passwords. The mailbox becomes the control plane. Vague rules act like a kind of stealth policy.
Concurrency, sequence, and locality must line up. If they do not, you must have a way to freeze the session, not the user. Make inbox rules visible, named, and attested. Alert on rule churn the way you alert on privilege churn. Score infrastructure by volatility and provenance, not brand. Expect remote tools to appear where they never should and block by context. Autonomous containment is a governance choice that decides outcomes. In this campaign, the absence of it gave the intruders time, which is the adversary's most important currency.
The playbook isn't new - it's the same old tricks as you would see on a desktop: changing inbox rules, stealing tokens, resetting passwords, and cleaning up tracks. The only twist is that it's happening on a rented cloud desktop, which makes the activity blend in with normal traffic a slightly differently. The real issue is the first break-in - usually stolen logins, hijacked sessions, weak MFA, or a maliciou
Read at Securitymagazine
[
|
]