---
id: "CVE-2026-124760"
page: "https://infrastack.ai/live/CVE-2026-124760"
kind: "vulnerability"
aliases: ["CVE-2026-124760"]
title: "P-Net: failed alarm parsing leaves later alarms unacknowledged until relationship release"
severity: "low"
cvss: 2.3
epss: null
exploited: false
status: "open"
category: "package"
products: []
ecosystems: []
tags: ["embedded", "industrial-control", "p-net", "package", "profinet"]
not_affected_if: ["no-application-relationship", "p-net-1.0.4-or-later"]
published_at: "2026-11-13T14:17:50+00:00"
first_seen_at: "2026-11-13T15:19:49+00:00"
updated_at: "2026-11-13T15:21:36+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-124760", "first_seen_at": "2026-11-13T15:19:49+00:00"}]
exploit_sources: []
---

# P-Net: failed alarm parsing leaves later alarms unacknowledged until relationship release

Link for people: https://infrastack.ai/live/CVE-2026-124760 (this .md is for agents)

P-Net is an industrial PROFINET device stack; before 1.0.4, a failed parse of an incoming alarm leaves the alarm receiver stuck waiting for confirmation. Later alarms of the same priority are then silently dropped without acknowledgement until the Application Relationship is released.

_Summary, check and fix written by a model from the sources; labelled as such._

## Fix

- P-Net: upgrade to `1.0.4` or later

## Timeline

- 2026-11-13T15:19:49+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-124760

## From the sources

P-Net before 1.0.4 sets the alarm receiver state to waiting for confirmation before the data of a received alarm notification is parsed. If the parsing fails the state is not restored, and later alarm notifications of the same priority are discarded without acknowledge until the Application Relationship is released. The Application Relationship and the cyclic data exchange keep working. The attacker must be able to send an alarm notification in the context of an established Application Relationship and requires ADJACENT network access.

---

_Free to read, with fair use of about 60 pages a minute. For higher limits or API access: sales@infrastack.ai_