Malware · Vulnerabilities
Hackers use Check Point flaws to backdoor gateways and wipe ESXi servers
Attackers exploited vulnerabilities in Check Point Security Gateways to install a persistent backdoor and wipe files on VMware ESXi servers. The campaign targeted Israeli organisations first, then spread globally.
Ross Grant3 min read
The short version
Hackers broke into Check Point Security Gateways by exploiting known and newly discovered vulnerabilities. They installed a backdoor to keep access and, in some cases, wiped data on VMware ESXi servers. The attacks started in Israel and then spread to other countries. If you use affected Check Point products, make sure you have installed the latest security updates.
By the numbers
- 2number of attack waves reported (30 October and 2 November 2026)The Hacker News
- 6affected Check Point product versions (R82.20, R82.10, R82, R81.20, R81.10, R81)The Hacker News
What happened?
Attackers exploited vulnerabilities in Check Point Security Gateways starting on 30 October 2026. They used these flaws to install a persistent backdoor and, in some cases, wiped files on VMware ESXi servers. The first wave targeted Israeli organisations, followed by a broader campaign worldwide. Only The Hacker News has reported this so far.
Who is affected
According to The Hacker News (opens in a new tab), the attacks began on 30 October 2026 and initially targeted Israeli organisations using Check Point Security Gateways. A second wave on 2 November 2026 hit organisations in other countries and sectors. The affected products include Check Point Multi-Domain Security Management and Multi-Domain Log Server, specifically versions R82.20, R82.10, R82, R81.20, R81.10, and R81.
How the attack worked
Attackers first exploited unpatched PHP2Shell vulnerabilities in Check Point Security Gateways to gain initial access. They then used a local privilege escalation flaw in Gaia OS (CVE-2026-123039, CVSS 7.8) to gain root access. This flaw allows any authenticated local user to run commands as root by sending a single request to a socket in the DynamicCLI process. The attackers deployed PHP web shells, which let them execute commands and upload files, and then installed a persistent backdoor called TSGA with a networking component called TSProxy.
Once inside, the attackers scanned internal networks and moved laterally to VMware ESXi hosts using SSH and the vpxuser service account. On these ESXi servers, they deployed a Linux wiper called destroy_v6, which deleted all files and attempted to overwrite them with zeros. They also left a politically themed note named "DREAM_ANNEXATION.txt" in each directory.
What was taken
The TSGA backdoor enabled attackers to steal configurations, maintain access, inject code into processes, and tamper with firewall logs. The web shells allowed further command execution and file uploads. The wiper malware destroyed data on ESXi servers, but there is no mention of specific personal or customer data being stolen.
Who is behind it
Check Point and The Hacker News report that the toolkit used (PhiliKit, including TSGA and TSProxy) was previously linked by ESET to a China-based group known as Armory Castle (formerly UNC5221). However, Check Point says there is not enough evidence to confirm the same group is behind this campaign. The attackers' focus on Israeli targets and references to the Israeli-Palestinian conflict in the wiper note complicate attribution. The report suggests several scenarios, including a false-flag operation or the use of access brokers.
The response
Check Point released a security fix for the privilege escalation flaw about a week before the attacks began. The company observed active exploitation against "a couple of dozen customers" and has published details of the attack chain and indicators of compromise. There is no mention of a wider advisory or patch for the PHP2Shell vulnerabilities in this report.
Is there a fix?
A security update for CVE-2026-123039 was available before the attacks started. Organisations running affected Check Point products should ensure all relevant patches are applied and check for signs of compromise, such as the presence of web shells, the TSGA backdoor, or wiper activity on ESXi hosts.
Claimed vs confirmed
- Unconfirmed
The campaign is linked to the China-based Armory Castle group.
ESET previously linked the toolkit to Armory Castle, but Check Point says there is not enough evidence to confirm this group is behind the current attacks.
Why it matters
This attack chain shows how quickly attackers can exploit newly patched flaws, especially when combined with older vulnerabilities. It highlights the risk of delayed patching and the potential for destructive attacks on virtual infrastructure.
Key points
- Attackers exploited both old and new flaws in Check Point gateways.
- A persistent backdoor (TSGA) was installed on compromised devices.
- Some attacks wiped files on VMware ESXi servers using custom malware.
- The campaign started in Israel and then spread globally.
- A patch for the main privilege escalation flaw was released before the attacks.
Sources
The Hacker News
Reference




