Overview
CISA republished Hitachi Energy PSIRT advisory 8DBD000258 as ICSA-26-302-03 on 29 October 2026 (Hitachi’s initial release was 27 October). It covers two flaws in the CoreTec product line: an OS command injection in the configuration daemon (CVE-2026-13162) and a Linux kernel privilege escalation known as “Dirty Frag” (CVE-2026-43284). CISA lists the affected sector as Energy, deployed worldwide, with Hitachi Energy headquartered in Switzerland. The headline score on the CISA page is CVSS v3 8.8.
Sources opened (checked 2026-11-11): the CISA advisory (C), the Hitachi Energy advisory 8DBD000258 rev 1 of 2026-10-27 (H), the CVE records for both CVEs (R), NVD for both CVEs, and the CISA KEV catalog (K). Hitachi’s advisory link returns only a JavaScript page shell in a browser; the text was read through a text-extraction proxy.
The two vulnerabilities
| CVE | CWE | Summary (per CISA) | Scores |
|---|---|---|---|
| CVE-2026-13162 | CWE-78 OS Command Injection | Authenticated users with certain roles can run arbitrary system commands with full control of the device | CISA page: v3.1 7.5 (AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H), v4.0 7.7 |
| CVE-2026-43284 | CWE-123 Write-what-where Condition | Local unprivileged user escalates to root through the Linux IPsec ESP subsystem (esp4) | v3.1 8.8 (AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H) |
CVE-2026-13162: timezone handler command injection
Per CISA, the flaw sits in the configuration daemon’s time synchronization handler, which sets the system timezone. When processing timezone updates, the daemon builds a shell command by concatenating user-supplied input with no sanitization or validation. A role-holding authenticated user can therefore inject commands and run them with full control over the device. The v4.0 vector is CVSS:4.0/AV:N/AC:H/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N: network reachable, high attack complexity, low privileges.
Scoring source: the CVE record is assigned by Hitachi Energy as CNA and carries only the CVSS v4.0 score of 7.7. The v3.1 score of 7.5 comes from Hitachi’s own advisory, which CISA republished verbatim; it is not in the CVE record or NVD. NVD was checked: its status for CVE-2026-13162 is Deferred, with no NVD primary score and only the Hitachi CNA’s 7.7 listed as Secondary.
CVE-2026-43284: Dirty Frag in the esp4 module
Per CISA, a crafted ESP packet can make the kernel decrypt data directly into memory pages it does not own, including cached copies of privileged binaries. When such a binary next executes, the attacker’s injected code runs as root. On CoreTec5 the vulnerable ESP component is compiled into the kernel as a built-in, so it is present from boot; in our reading, that means it cannot be avoided by blacklisting a module.
The CVE is a Linux kernel CVE (CNA: Linux), and the 8.8 score comes from that CNA record. Its scoring rationale says triggering needs local access to splice and UDP sockets and a local ESP-in-UDP xfrm path; a remote packet alone is not sufficient. The same rationale notes that the XFRM setup needs CAP_NET_ADMIN, which an unprivileged user can obtain inside a new user and network namespace where user namespaces are allowed. Scope is Changed because exploitation from a namespace can reach host root. Other scorers differ: CISA-ADP gives 7.8 (AV:L/AC:H/PR:L/UI:N/S:C) and Red Hat gives 7.8 (AC:L, S:U). NVD was checked: its status is Modified, it has no NVD primary score, and it lists all three as Secondary.
Affected and fixed versions
| Product | Affected (per CISA) | CVEs | Vendor fix / action |
|---|---|---|---|
| CoreTec5 | 3.4.0 and prior | CVE-2026-13162, CVE-2026-43284 | Update to CoreTec5 5.0.0 |
| CoreTec4 | 3.4.0 and prior | CVE-2026-13162 | Upgrade to CoreTec5 5.0.0; follow general mitigation factors |
Version-range discrepancy: Hitachi’s advisory and CISA’s CSAF data both list CoreTec5 3.4.0 and prior as affected and 5.0.0 as the fixed, recommended version. CISA’s HTML summary line “CoreTec5/<=3.4.0, 5.0.0” is, in our reading, built from that product tree and is not an affected list; CISA does not say so. The CVE-2026-13162 record (Hitachi as CNA) disagrees: it lists CoreTec 5 from 3.0.0 up to and including 5.0.0, and CoreTec 4 from 3.0.0 up to 3.4.0. Follow the vendor advisory: the fix is 5.0.0. Hitachi PSIRT could confirm which range is intended; we think the record is in error, but Hitachi has not said so. Check the record again for updates (last updated 2026-11-04).
Hitachi describes CoreTec4 as an end-of-life product. For CVE-2026-13162 its CoreTec4 guidance is to upgrade to CoreTec5 5.0.0 or follow the general mitigation factors. The vendor table lists CVE-2026-43284 for CoreTec5 only. CISA’s HTML page shows a CoreTec4 entry under CVE-2026-43284 with no remediation; CISA’s CSAF data has no CoreTec4 for that CVE, so this looks like a conversion artifact.
Exploitation status
Hitachi’s FAQ, dated to the advisory’s publication on 2026-10-27, says it “is not aware of these vulnerabilities being exploited in the wild to affect CoreTec products at the time of this advisory publication”. That is a vendor statement for CoreTec products, not a general one. CISA’s page records an SSVC decision point of E:N (Exploitation: none) timestamped 2026-10-26, and the CISA-ADP entry on the CVE-2026-13162 record shows “Exploitation: none”, “Automatable: no”, “Technical Impact: total”. CVE-2026-13162 was reported through coordinated disclosure. We found no evidence of exploitation as of 2026-11-11 in H, C, R or K; neither CVE is in the KEV catalog (version 2026.11.10). Absence from those sources is not proof of safety.
CVE-2026-43284 is different on public detail: Hitachi states the open-source vulnerability is publicly disclosed, the CISA-ADP entry on its record shows SSVC Exploitation “poc” (dated 2026-05-08), and the record’s references include oss-security posts, vendor detection scripts and a public GitHub repository. Public PoC code exists. Microsoft (8 May 2026) reported limited in-the-wild activity that “may be indicative of” Dirty Frag or Copy Fail on Linux in general, and SecurityWeek (11 May 2026) reported possible exploitation. These reports are not CoreTec-specific, Microsoft could not separate Dirty Frag from Copy Fail, and the CVE is not in CISA KEV (v2026.11.10). Hitachi’s CoreTec-scoped statement above still stands as its position at publication. We did not open the oss-security posts, the GitHub repository, or any VulnCheck or exploit-intel listing.
Operational impact
CISA says exploitation of either flaw affects confidentiality, integrity and availability of the product. Both need an existing foothold: valid credentials with a suitable role for CVE-2026-13162, or local code execution for CVE-2026-43284. The advisory says command execution through CVE-2026-13162 gives “full control over the device”, so in our assessment a chain with the kernel flaw may not be needed on CoreTec5.
Recommendations
Vendor and CISA actions: move to CoreTec5 5.0.0; apply Hitachi’s general mitigation factors (no direct internet exposure, firewalls with minimal open ports, physical protection, removable media scanning); and follow CISA’s recommended practices of minimizing exposure and isolating control networks.
Our assessment:
- Inventory CoreTec units and record build versions now, given the range discrepancy above.
- Review which accounts hold roles able to change time and timezone settings, and remove the permission from accounts that do not need it.
- Restrict management-plane access to a dedicated jump host or management VLAN so that the network-reachable path to the configuration daemon is not available from general corporate or vendor networks.
- Plan CoreTec4 migration: the product is end-of-life per Hitachi, and the only vendor fix path is CoreTec5 (the alternative is the general mitigation factors). That no patch will arrive in the CoreTec4 branch is our inference from the end-of-life status.
Unverified
Hitachi has not stated whether the CVE-2026-13162 record’s version range is an error. The oss-security posts and the public GitHub repository for CVE-2026-43284 were not opened. CoreTec4 exposure to CVE-2026-43284 is not listed by the vendor.