Qilin VPN intrusions start with CVE-2026-0257, an authentication bypass in Palo Alto Networks PAN-OS GlobalProtect that lets attackers forge a login cookie and open a VPN session without a password. Then Qilin affiliates dump credentials, spread with PsExec, steal data and encrypt the domain. Seven fixes, applied in order, close that path.
How Does a Qilin VPN Attack Start?
The attack begins at the front door. Palo Alto Networks published the CVE-2026-0257 advisory on May 13, 2026, and described an authentication bypass in the GlobalProtect portal and gateway. Specifically, the flaw (CWE-565) lets an attacker skip the login and establish an unauthorized VPN connection. As a result, the attacker never needs to phish anyone or buy a stolen password.
However, the bug only works under one configuration. First, the portal or gateway must accept authentication override cookies. Second, the device must use the same certificate for those cookies and for its HTTPS service. Because anyone who visits the portal can read that certificate’s public key, an attacker can forge a cookie, and the device then trusts the contents without checking a signature. In short, the session looks exactly like normal external remote services traffic from an employee.
Which PAN-OS Versions Need the Patch?
Palo Alto Networks lists fixed releases for four PAN-OS branches and two Prisma Access branches. Meanwhile, the vendor confirms that Panorama and Cloud NGFW are not exposed. Check your version against this table, which reflects the advisory as last updated on June 3, 2026.
| Product | Upgrade to at least one of these releases |
|---|---|
| PAN-OS 12.1 | 12.1.4-h6 or 12.1.7 |
| PAN-OS 11.2 | 11.2.4-h17, 11.2.7-h14, 11.2.10-h7 or 11.2.12 |
| PAN-OS 11.1 | 11.1.4-h33, 11.1.6-h32, 11.1.7-h6, 11.1.10-h25, 11.1.13-h5 or 11.1.15 |
| PAN-OS 10.2 | 10.2.7-h34, 10.2.10-h36, 10.2.13-h21, 10.2.16-h7 or 10.2.18-h6 |
| Prisma Access 11.2 | 11.2.7-h13 (vendor upgrades on its schedule) |
| Prisma Access 10.2 | 10.2.10-h36 (vendor upgrades on its schedule) |
| Panorama, Cloud NGFW | Not affected |
What Happens After Attackers Get Inside?
During June 2026, incident responders investigated multiple Qilin VPN intrusions that began with CVE-2026-0257 and ended in Qilin ransomware. Although each case differed, the chain followed the same broad steps.
- Access: a forged cookie opens a VPN session, often from VPS infrastructure, and some attacker machines identified themselves as kali
- Credential theft: attackers dumped LSASS memory with rundll32 and comsvcs.dll MiniDump, then copied the Active Directory database with ntdsutil IFM
- Discovery: they scanned the network with SoftPerfect Network Scanner and NetExec
- Lateral movement: PsExec over C$ admin shares served as the main tool, with RDP as a backup
- Persistence: they installed AnyDesk, Ngrok, LogMeIn and, in one case, MeshAgent
- Exfiltration: in double extortion cases, Rclone, ProtonDrive and FileZilla moved data, mostly to MEGA, and attackers also went after Veeam backups
- Impact: they staged win.exe in C:\PerfLogs, turned off Defender in some cases, cleared event logs, and encrypted the domain
Notably, the payload demanded a 32 character password at runtime, which blocks sandbox analysis. In addition, a PowerShell routine walked every event log and wiped each one, a clear case of clearing Windows event logs to destroy the timeline. Some intrusions reached encryption with very little dwell time, while others involved long reconnaissance first.
How Fast Did the Threat Move?
The gap between disclosure and ransomware measured weeks, not months. For that reason, the dates below matter for your log review window.
| Date (2026) | Event |
|---|---|
| May 13 | Palo Alto Networks publishes the advisory |
| May 17 | Earliest reported exploitation in the wild |
| May 29 | Researchers publish technical analysis and a public test script; CISA adds the flaw to KEV |
| June 3 | Vendor updates the advisory and marks exploit maturity as attacked |
| Early June | Responders report a rise in exploitation activity |
| June | Multiple intrusions end in Qilin ransomware |
| July 20 | Research linking the chain to Qilin goes public |
How Strong Is the Evidence?
The core facts behind the Qilin VPN campaign rest on solid ground, yet some popular claims stretch them. Therefore, this table separates the confirmed points from the reported and the open ones.
| Claim | Status |
|---|---|
| Authentication bypass in GlobalProtect, fixed versions, cookie and certificate conditions | Confirmed by Palo Alto Networks |
| Exploitation in the wild | Confirmed by the vendor and by CISA KEV on May 29 |
| Severity | 7.8 High under CVSS v4.0, not Critical, with the vendor’s highest urgency |
| Qilin deployment after CVE-2026-0257 access | Reported by one incident response firm from multiple June cases |
| Tools, paths and source IPs listed in this post | Reported from those same investigations |
| One actor or several Qilin affiliates | Open; the research allows either a single actor or shared infrastructure |
| Number of victims and time to encryption | Not published |
What Are the 7 Critical Qilin VPN Fixes?
Order matters for every Qilin VPN fix, because a patch without the follow up steps can leave an intruder in place. Work through these seven fixes from top to bottom.
Fix 1: Patch Every Affected GlobalProtect Device
First, inventory every GlobalProtect portal and gateway, then upgrade each one to a fixed release from the table above. Do not wait for the next change window, because CISA already lists the flaw as exploited. Likewise, confirm with Palo Alto Networks that your Prisma Access tenant has received its upgrade.
Fix 2: Terminate Every Active VPN Session
Next, end all active GlobalProtect sessions right after patching and force full re-authentication. A patch stops new forged cookies; however, it does not close a session that an attacker opened last week.
Fix 3: Remove the Cookie Risk
Then, turn off authentication override on the portal and the gateway if your users do not need it. If they do, generate a new certificate used only for those cookies, and never share the portal or gateway HTTPS certificate with this feature.
Fix 4: Review VPN Logs Back to May 17
After that, pull GlobalProtect logs from May 17, 2026, through the day you terminated sessions. Look for sessions from hosting or VPS address ranges, hostnames such as kali or default DESKTOP names, and sequential internal address assignments. Also, correlate each suspicious session with firewall, identity and endpoint data.
Fix 5: Hunt Credential Theft and Rotate Accounts
Meanwhile, search for LSASS access through comsvcs.dll MiniDump, output files such as C:\Windows\Temp\output.odt, and ntdsutil IFM runs. If you find them, assume domain compromise and rotate domain admin accounts, the KRBTGT account twice, service accounts and cloud storage credentials. In addition, restrict who holds IFM or DCSync rights, and protect LSASS with Credential Guard.
Fix 6: Block Lateral Movement, Staging and Exfiltration
Likewise, alert on PSEXESVC.exe, writes to admin shares, and bursts of Logon Type 3 events across many hosts. Put file integrity monitoring and application control on C:\PerfLogs. Finally, block or alert on unsanctioned AnyDesk, Ngrok, LogMeIn, Rclone and MEGA traffic, and isolate backup servers so attackers cannot reach Veeam first.
Fix 7: Forward Logs Before Attackers Wipe Them
Above all, stream Security, System, PowerShell, Sysmon, VPN and firewall logs to a remote SIEM in real time. Then, if attackers clear local logs, your SIEM keeps the copy, and Event ID 1102 becomes an alarm instead of a dead end.
Which Indicators Should You Load Now?
Load these reported source addresses, hostnames and file paths into your SIEM and EDR. Because Qilin VPN operators rotate infrastructure, treat the IP addresses as leads for your May to June review rather than a complete blocklist.
Why Do VPN Gateways Make Such Good Targets?
Ultimately, Qilin VPN attacks succeed because gateways carry trust that endpoints do not. Once a session looks authenticated, internal controls often treat it as an employee, and stolen domain credentials then make every later step look like routine administration. Our 2026 Mid Year Threat Pattern Report tracks the same shift toward edge devices as the first foothold, and our analysis of an AI PowerShell Active Directory attack shows how quickly attackers turn domain access into control.
For that reason, edge devices need an owner, a patch deadline and a session revocation plan. Our team runs vulnerability assessments that find exposed and misconfigured VPN gateways, penetration testing that checks whether your SIEM would catch a chain like this one, and a PTaaS platform that keeps that testing continuous.
Key Takeaways
- CVE-2026-0257 lets attackers forge a GlobalProtect authentication override cookie when the device reuses its HTTPS certificate for those cookies
- Palo Alto Networks rates it 7.8 High with its highest urgency, and CISA added it to KEV on May 29, 2026
- Incident responders tied multiple June 2026 Qilin VPN intrusions to ransomware deployment
- Patch first, then terminate sessions, remove the cookie risk and review logs back to May 17
- Hunt LSASS and NTDS theft, PsExec, C:\PerfLogs staging, Rclone to MEGA and Event ID 1102, and stream logs to a remote SIEM
Frequently Asked Questions About Qilin VPN Attacks
What Are Qilin VPN Attacks?
They are intrusions where Qilin ransomware affiliates exploit CVE-2026-0257 in Palo Alto Networks GlobalProtect to open a VPN session without valid credentials. From there, the attackers steal domain credentials, move laterally, take data and encrypt the network.
What Is CVE-2026-0257?
CVE-2026-0257 is an authentication bypass in the GlobalProtect portal and gateway in PAN-OS. Palo Alto Networks rates it 7.8 High under CVSS v4.0 with its highest urgency, and CISA added it to the Known Exploited Vulnerabilities catalog on May 29, 2026.
Which GlobalProtect Setups Can Attackers Exploit?
The risk exists when a portal or gateway accepts authentication override cookies and uses the same certificate for those cookies and for its HTTPS service. In that case, anyone can pull the public certificate and forge a cookie the device trusts.
Does Patching Alone Remove an Intruder?
No. A patch blocks new forged cookies, but it does not end VPN sessions that attackers opened earlier. Therefore, teams should terminate every active GlobalProtect session after patching and then hunt for activity inside the network.
What Should Teams Hunt for After Exposure?
Look for VPN sessions from hosting or VPS addresses, hosts named kali, LSASS dumps through comsvcs.dll, ntdsutil IFM, PsExec and admin share writes, files in C:\PerfLogs, Rclone or MEGA traffic, and Event ID 1102.
Which Credentials Should Teams Rotate?
If exploitation looks likely, assume domain compromise. Rotate domain admin accounts, the KRBTGT account twice, service accounts, VPN users and any cloud storage credentials that attackers could have reached.
Is CVE-2026-0257 Rated Critical?
Palo Alto Networks scores it High rather than Critical, at 7.8 under CVSS v4.0. However, the vendor still assigns its highest urgency, because attackers exploit it in the wild and the flaw sits on internet facing VPN gateways.

