Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

What Are the Ivanti Connect Secure Vulnerabilities?

A sequence of critical flaws in Ivanti Connect Secure VPN appliances, exploited from December 2023 onwards, gave unauthenticated attackers persistent root access to the network edge.

Illustration for the cybersecurity term Ivanti Connect Secure Vulnerabilities
Home / Blog / Cybersecurity Terms / What Are the Ivanti Connect Secure Vulnerabilities?
17 November 2026

What Are the Ivanti Connect Secure Vulnerabilities?

What Are the Ivanti Connect Secure Vulnerabilities?

Ivanti Connect Secure, formerly Pulse Connect Secure, is an SSL VPN appliance used by thousands of organisations to publish internal applications to remote staff. Between January 2024 and January 2025 it was the subject of several critical, actively exploited vulnerabilities that gave unauthenticated attackers code execution as root on a device sitting directly on the internet. The sequence is worth studying not for any single bug but for what it shows about the security of the network edge.

The January 2024 Chain

Volexity identified the first intrusions in December 2023 while investigating anomalous traffic at a customer. Ivanti published advisories on 10 January 2024 covering two flaws:

Chained, the two produced unauthenticated remote code execution with no user interaction. Volexity attributed the earliest activity to UNC5221, a suspected China-nexus espionage group, which deployed web shells it called GIFTEDVISITOR and THINSPOOL, trojanised legitimate appliance components, and harvested credentials as users authenticated.

Two more followed within weeks: CVE-2024-21893, a server-side request forgery in the SAML component that was itself exploited in the wild, and CVE-2024-21888, a privilege escalation in the web interface. CISA issued Emergency Directive 24-01 on 19 January 2024, instructing US federal agencies to disconnect affected appliances, and the NCSC published parallel guidance urging UK organisations to assume compromise rather than wait for confirmation.

Why the Integrity Checker Did Not Save Anyone

Ivanti shipped an Integrity Checker Tool intended to tell administrators whether an appliance had been tampered with. Attackers worked out how to defeat it, in some cases by replacing the checker’s own comparison data, so a clean report proved nothing. Several organisations ran the tool, saw no findings, and remained compromised.

The appliance design made this worse. Connect Secure is a sealed Linux system with no shell access for the customer, no endpoint agent, and limited logging that is lost on factory reset—the exact action responders were told to take. You cannot run EDR on a device the vendor does not let you into, which is precisely why these appliances are attractive targets.

The 2025 Round

On 8 January 2025 Ivanti disclosed CVE-2025-0282, a stack-based buffer overflow allowing unauthenticated remote code execution, alongside CVE-2025-0283, a local privilege escalation. Mandiant reported exploitation from mid-December 2024, with the SPAWN malware family deployed by UNC5337. A further overflow, CVE-2025-22457, followed in April 2025. Earlier history matters too: Pulse Connect Secure had suffered CVE-2019-11510 and CVE-2021-22893, both exploited at scale.

Responding to an Edge Appliance Compromise

The standard patch-and-move-on process does not fit. A credible response looks like this:

  1. Treat exposure as compromise. If the appliance was internet-facing and unpatched during a known exploitation window, work from that assumption.
  2. Rebuild rather than clean. Factory reset and reimage from vendor media, then apply the current patch before reconnecting.
  3. Rotate everything the device touched. Local accounts, the appliance certificate and private key, any service account used for LDAP or RADIUS integration, and the credentials of every user who authenticated during the exposure window.
  4. Hunt inside the network. A VPN concentrator is an entry point, not a destination. Look for lateral movement from the appliance’s address in the days before patching.
  5. Preserve evidence first. Export configuration, logs, and any available support bundle before the reset destroys them.

The Structural Lesson

Security appliances now account for a large share of the initial access entries in the CISA Known Exploited Vulnerabilities catalogue. They are internet-facing by design, run complex web stacks written in older languages, cannot be instrumented by the customer, and hold the credentials of the whole remote workforce. Reducing that risk means limiting administrative interface exposure to management networks, monitoring the appliance’s outbound connections as closely as its inbound ones, subscribing to vendor advisories as an operational feed rather than a newsletter, and keeping a tested rebuild procedure for every edge device.

At Privia Security, our Infrastructure and Architecture Review Services examine exactly these trust assumptions at the perimeter.

FAQ – Frequently Asked Questions

Is Patching Enough After an Ivanti Appliance Was Exposed?

No. Patching closes the vulnerability but leaves any web shell, trojanised binary, or stolen credential in place. Attackers routinely maintained access through the update process, and the Integrity Checker Tool was bypassed. The accepted guidance from both CISA and Ivanti was to factory reset, reimage, apply the fix, and rotate every secret the appliance held before returning it to service.

Which Credentials Need Rotating After a VPN Appliance Compromise?

The appliance’s local administrator accounts, its TLS certificate and private key, any API tokens, and the service accounts used for directory, RADIUS, or MFA integration. Beyond that, every user credential submitted through the login page during the exposure window should be treated as captured, because the deployed malware was specifically designed to log plaintext authentication as it passed through.

Why Are VPN Appliances Targeted So Often?

They are reachable from the internet by definition, they terminate authentication for the entire remote workforce, and they sit at a trust boundary with routes into internal networks. They also run large closed-source web applications that customers cannot inspect, patch independently, or monitor with endpoint tooling. That combination of high value and low visibility makes them a priority for both espionage groups and ransomware affiliates.

Does Zero Trust Remove This Risk?

It reduces the blast radius rather than the vulnerability. A compromised appliance still gives an attacker a foothold, but under a zero trust model that foothold grants no implicit network access, because every internal service independently verifies identity and device posture. Network segmentation and per-application authorisation turn an edge compromise into a contained incident instead of a domain-wide one.

Related Terms

You May Be Interested In These

What Is Ransomware as a Service?

Cybersecurity Terms

17 November 2026

What Is Ransomware as a Service?

Read More
What Is an IMDS Attack?

Cybersecurity Terms

17 November 2026

What Is an IMDS Attack?

Read More