P-Net industrial network stack can crash or be hijacked by crafted requests
A flaw in P-Net, a networking library used in industrial devices, lets a specially crafted connection request write data past the memory it should stay within. This can crash the device or, in the worst case, let an attacker run their own c
| Severity | CriticalCVSS 9.3 of 10 |
|---|---|
| Being exploited? | UnclearNo reports either way yet |
| Fix available? | YesP-Net 1.0.3 or later. · since 29 Oct 2026 |
| How sure is this? | Official sourceBacked by an official source |
How the attack works simplified
- 1Attacker sends requestCrafted Connect request to the device
- 2Request reaches deviceParsed by the ARVendorBlock handler
- 3Unchecked length usedLength field not validated against buffer size
- 4Out-of-bounds writeMemory beyond buffer is overwritten
- 5Crash or code executionDevice stops or runs attacker code
- ResultDevice crashes or attacker runs code
What happened
In P-Net before 1.0.3, built with PNET_OPTION_AR_VENDOR_BLOCKS enabled, the ARVendorBlock handling copies a length-specified number of bytes from a Connect request into a fixed-size buffer without checking bounds, causing an out-of-bounds write.
Are you affected?
You are affected if you run P-Net before 1.0.3 built with the PNET_OPTION_AR_VENDOR_BLOCKS option enabled (disabled by default).
P-Net versions before 1.0.3, only when built with PNET_OPTION_AR_VENDOR_BLOCKS enabled.
What to do
- Update P-Net to version 1.0.3 or later
- If you cannot update immediately, rebuild without PNET_OPTION_AR_VENDOR_BLOCKS enabled, since the option is off by default
- Limit network access to affected industrial devices to trusted hosts
Timeline
- 29 Oct 2026, 16:17 UTCPublicly disclosed
- 29 Oct 2026, 16:17 UTCReported by NVD · NVD
- 29 Oct 2026, 16:20 UTCPicked up by Zero Radar
Sources 1 report, oldest first
Official vendor, government CERT or advisory database
- CVE-2026-111732NVD · 29 Oct 2026, 16:17 UTC
Technical details
- ID
- CVE-2026-111732
- CVSS
- 9.3 / 10
- Disclosed
- 29 Oct 2026, 16:17 UTC
- On Zero Radar
- 29 Oct 2026, 16:20 UTC (2 min after disclosure)
- Last change
- 29 Oct 2026, 16:20 UTC
The plain-language parts and the attack diagram were written by an AI model (anthropic/claude-haiku-5.5) from the sources above. Facts like versions, exploitation and dates come from the sources themselves. How Zero Radar works