The 31 October batch carries four records. Two of them belong to one attack chain on a single appliance, and CISA listed both on 30 October, with the escalation to root due 2 November 1. That pair goes first. The Chrome records are two 9.6 fixes, neither in the KEV catalog and neither with exploitation reported in NVD.

Three Ivanti bugs, one appliance

Ivanti’s advisory covers three CVEs on Connect Secure, and the company states it saw exploitation in the wild against a limited number of customers at disclosure, with the three chained 2. Two of the three entered CISA’s catalog on 30 October 1.

CVE-2026-104935 opens the chain. The flaw is an authentication bypass by spoofing that lets a remote unauthenticated attacker run commands. NVD scores it 8.6, and CISA added it to the catalog on 13 October with a 16 October deadline 3 1.

CVE-2026-111357 escalates to root. NVD’s record describes the dsinvoked broker trusting DSCache configuration data and states that exploitation happened in the wild in October 2026 4. The two scores differ on attack complexity: NVD says 7.0 with high complexity, Ivanti says 8.8 with low complexity, and both assume a local authenticated attacker 4 2. Volexity describes the chain reaching code execution as the cgi-server user, then a separate privilege escalation in which the attacker edits that process’s in-memory copy of one library so dsinvoked runs a Perl script as root 5.

CVE-2026-118527 is a predictable-filename bug that lets a remote authenticated attacker upload files with attacker-controlled content to a known location on disk 1. Ivanti scores it 4.3, limits the impact to integrity, and lists it as the third link 2.

The escalation is due 2 November and carries a forensics triage requirement; the upload bug is due 13 November 1.

Fixed builds for the pair are 25.1.1.3 and 22.8R2.6, covering 25.1.1.2 and prior plus 22.8R2.5 and prior 2. Ivanti calls 22.8 a bridge to 25.1 and sets its end of life at 31 January 2027 2. Moving the bridge to 25.1.3.0 on the ISA4500, ISA6500 or ISA8500 needs new hardware or a rebuilt virtual device first, so that migration does not fit in a two-day window 2.

In my environment the check starts with the build number on every appliance that terminates remote access, then whether any of them still run the 22.8 bridge, then whether the Integrity Checker Tool has run since 13 October. Ivanti describes indicators of compromise as limited and points customers at Volexity’s write-up and the checker 2. Exploitation started before the fix existed, so a clean build number does not settle the history of the box on its own.

Two Chrome bugs at 9.6

Chrome 156.0.8078.79 fixes CVE-2026-119261, an input validation failure in the Skia graphics library, and CVE-2026-119263, memory corruption in garbage collection 6 7. NVD published both records on 30 October at 9.6, with the same stated outcome: arbitrary code execution outside the sandbox from a crafted HTML page 6 7.

The vector carries that score: network attack, no privileges, a user who opens the page, and a scope change that pushes the impact past the browser boundary 6. Neither CVE sits in the catalog, and NVD’s records carry no exploitation note 6 7.

Browsers update on their own schedule, so my work here is a version report rather than a deployment. I want the build number per machine and the list of machines that stopped checking in, since a browser pinned by policy or held by an application dependency stays on the old build past every release note. An endpoint that has not reported for a month tells me nothing about its Chrome version.

Order for the weekend

The 2 November date for the Ivanti escalation falls on Monday. Patching and deployment in my environment run through automation and the teams that own the systems, so what I add is the order: internet-facing VPN appliances first, the 22.8 bridge machines second, and the platform migration on its own track because it needs hardware 2. The 13 November date for the upload bug leaves room, and one upgrade closes both legs 1 2.

Chrome runs alongside that work. The install needs no change window and the version check needs no approval, so a report on Monday morning says whether the browser side is done.

The appliance work needs a person to read a version string and a triage result, and that is the part I would not leave to Monday morning.


  1. Cybersecurity and Infrastructure Security Agency. (2026, October 30). Known exploited vulnerabilities catalog (catalog version 2026.10.30). CISA. https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎

  2. Ivanti. (2026, October 13). Security advisory: Ivanti Connect Secure (CVE-2026-104935) (Article 000109571, updated October 29). Ivanti Innovators Hub. https://hub.ivanti.com/s/article/Security-Advisory-Ivanti-Connect-Secure-CVE-2026-104935 ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎

  3. National Vulnerability Database. (2026, October 13). CVE-2026-104935 detail. National Institute of Standards and Technology. https://nvd.nist.gov/vuln/detail/CVE-2026-104935 ↩︎

  4. National Vulnerability Database. (2026, October 15). CVE-2026-111357 detail. National Institute of Standards and Technology. https://nvd.nist.gov/vuln/detail/CVE-2026-111357 ↩︎ ↩︎

  5. Volexity. (2026, October 13). A series of unauthenticated events: Chained exploitation of Ivanti Connect Secure zero-days. Volexity. https://www.volexity.com/blog/2026/10/13/a-series-of-unauthenticated-events-chained-exploitation-of-ivanti-connect-secure-zero-days/ ↩︎

  6. National Vulnerability Database. (2026, October 30). CVE-2026-119261 detail. National Institute of Standards and Technology. https://nvd.nist.gov/vuln/detail/CVE-2026-119261 ↩︎ ↩︎ ↩︎ ↩︎

  7. National Vulnerability Database. (2026, October 30). CVE-2026-119263 detail. National Institute of Standards and Technology. https://nvd.nist.gov/vuln/detail/CVE-2026-119263 ↩︎ ↩︎ ↩︎