CVE-2026-109618
This CVE is being exploited. Apply the vendor fix, or take the service off the internet until you can, then check your logs for attempts.
What it is
PHP uses longjmp() for out-of-memory (OOM) errors to exit the current context and abort handling of the request. Some contexts may not properly account for this scenario and leave request memory in an uninitialized state before performing an allocation. The subsequent request cleanup procedure may read or branch on this memory. Multiple OOM strategies can consistently trigger this condition and lead to remote code execution (RCE).
Live exploitation on the huntback network (30 days)
Detection
Evidence
Confirmed exploited. Our decoys saw 22 exploitation attempts in the last 14 days; CISA added it to KEV 17 days ago.
Last attempt. 22 attempts on our decoys, 22 in the last 14 days
First exploitation attempt on our decoys
Reported exploited in the wild: CISA ADP SSVC decision: Exploitation active
CISA Confirms Active Exploitation of Two PHP2Shell Flaws Enabling Remote Code Execution
Added to CISA Known Exploited Vulnerabilities
Published
Vendor advisory: Vulnerabilities in PHP Affecting Cisco Products: October 2026
366 public proof-of-concept repos
406 public detection scripts (scanner or checker), no exploit code
Exploit code published (ExploitDB)
Nuclei detection template available
Observed: our own decoys saw it. Reported: CISA, the vendor or another named source says so. Weaponised: public exploit code or a scanner template exists. Methodology
Questions
Is CVE-2026-109618 being exploited in the wild?
Yes. The huntback decoy network captured 22 exploitation attempts from 3 distinct source IPs; the most recent was on 2026-11-11. It is also in the CISA KEV catalogue.
Can CVE-2026-109618 be detected on the network?
Yes. huntback carries a network signature for CVE-2026-109618, so attempts against a decoy are tagged in real time.
How huntback helps
Deploy a decoy that emulates Amis Ts and you will see every attempt at this CVE the moment it lands, with the full payload, the attacker's infrastructure, and any stage-2 loader. Then huntback can scan the attacker back.
huntback research
Inside SOÛL SOCIETY: automated exploitation, AI agent orders and a Telegram drug platformAn operator used the pentestcode AI agent against 25 mid-size and large European companies and reports breaching 10 of themPHP2Shell one week on: inside two operators' exploit labsPHP2Shell in the wild: confirmed RCE, probing, and exposed attacker infrastructureRelated
NVD record → Browse all CVEs → Live malware and C2 finds → Novel-attack discovery → Deploy a decoy →API
This record, with its evidence timeline, as JSON:
curl -H "X-API-Key: $HUNTBACK_KEY" https://app.huntback.io/api/v1/evidence/CVE-2026-109618