Junos OS Evolved kernel flaw allows remote code execution, patch still pending
Juniper Networks issued an out-of-cycle bulletin for a Junos OS Evolved vulnerability, CVE-2026-98432, that allows remote code execution and denial of service. Every version is affected and no fix is available yet, so administrators of Juniper devices running this OS should apply the vendor's workaround.
What to do now
- Product
- Juniper Networks Junos OS Evolved
- Affected versions
- Junos OS Evolved, all versions
- CVE
- CVE-2026-98432
France's CERT-FR published advisory CERTFR-2026-AVI-1551 on 17 November 2026. It covers a vulnerability in Junos OS Evolved, the Linux-based operating system from Juniper Networks. According to the advisory, a remote attacker can exploit the flaw to run arbitrary code or cause a denial of service.
The vulnerability is tracked as CVE-2026-98432. Juniper describes it in security bulletin JSA119568, dated 16 November 2026 and released outside the vendor's regular schedule. The bulletin's title identifies it as a use-after-free bug in the IPv4 TCP handling of the underlying Linux kernel. CERT-FR does not explain the exploitation conditions or say whether the flaw is being exploited.
The scope is broad: CERT-FR lists all versions of Junos OS Evolved as affected. Juniper says a fix will be released soon but has not given a date. No fixed version has been named yet.
Juniper has published a workaround in the meantime. The CERT-FR advisory does not describe it and points readers to bulletin JSA119568 on Juniper's support portal for the details.
Devices running this OS often sit at the core of a network. Remote code execution on them could have wide consequences, and a denial of service alone could disrupt connectivity.
What to do now
- Identify every device in your estate running Junos OS Evolved, regardless of version.
- Read Juniper bulletin JSA119568 and apply the workaround it describes to each affected device without delay.
- Watch Juniper's announcements and install the fix as soon as it is released.
- As general good practice, restrict network exposure of these devices to required traffic only and investigate any unexplained crash or reboot.