
Quick Answer
Today’s briefing covers 13 security issues reported in the last 24 hours, grouped by type: 2 actively exploited, 1 breaches and data exposure, 3 ransomware and attacks, 5 patches and advisories, 2 scams and phishing. Each entry says who is affected and what to do, with links to the original reports.
Actively Exploited
BO Team Exploits Log4Shell to Compromise Russian Code-Signing Server
Who is affected: An unnamed Russian software developer with a customer base in the Russian Federation had its development infrastructure, code-signing server and IT infrastructure compromised over several months.
Kaspersky’s ICS CERT team attributed the intrusion to BO Team after the affected developer contacted it in June 2025. According to Kaspersky, the intrusion resulted in the theft of confidential information and the compromise of all components of the organization’s IT infrastructure, including Windows and Linux servers, workstations, hypervisors, backup servers, security solution management servers and network equipment. The attackers maintained command channels and planted backdoors rather than destroying data or claiming responsibility publicly.
What to do: Review VMware vCenter for Log4Shell exposure and investigate the organisation’s servers, workstations, hypervisors, backups, security management systems and network equipment for compromise.
Sources: thehackernews.com
Hackers Targeted Flaws in Copeland XWEB EVO Refrigeration Controllers
Who is affected: Organizations using Copeland XWEB EVO controllers are affected. The product drew VulnCheck’s attention after the October 11-12 attack that triggered defrost cycles at roughly 700 French supermarkets.
VulnCheck disclosed 14 vulnerabilities in the XWEB EVO supervisory system, and exploitation attempts began within two hours of the public disclosure. Two flaws were targeted, including unauthenticated configuration-file writing and unauthenticated root command execution, with both affecting older firmware. Nine of the flaws disclosed by VulnCheck only affect firmware builds that predate a 2021 rewrite. The other five vulnerabilities also impact the latest firmware. According to SecurityWeek, these five issues will not be fixed.
What to do: Copeland advises customers using XWEB EVO controllers to move to XWEB Pro models.
Sources: securityweek.com
Breaches and Data Exposure
Uber Data Breach Exposed Data Belonging to 149,346 People
Who is affected: Uber is notifying roughly 150,000 drivers whose personal information was stolen. Uber reportedly notified the Texas Attorney General’s Office that a total of 149,346 people were affected.
According to Uber’s investigation, the hackers used compromised service provider credentials on October 7 to access data Uber had shared with the provider. A company spokesperson confirmed to TechCrunch that the incident involved Fireworks AI. The accessed information included images of driver’s licenses, names, addresses, dates of birth, driver’s license numbers, and government-issued ID numbers. Uber learned of the incident on October 9, severed its data connections with the provider, and launched an investigation.
What to do: Potentially affected individuals should use the free credit monitoring and fraud assistance services Uber is providing.
Sources: securityweek.com
Ransomware and Attacks
Tor 0.4.9.17 Patches DoS Protection Issues During Relay Attacks
Who is affected: Most relay operators have been affected by the ongoing DDoS attack on the Tor network.
Tor 0.4.9.17 patches several security issues, including how built-in DoS protections handle IPv6. Reports indicate that the disruption continued after the update, and one operator said two relays had lost HSDir status before the update.
What to do: Update Tor relay software without delay.
Sources: securityweek.com
SANSAR Hides GPU Mining Activity With a Modified NVIDIA Library
Who is affected: Linux systems compromised through an exposed Ray job agent are affected by the REF3364 cryptocurrency mining operation.
The actor uses SANSAR, a Rust-based orchestrator that receives commands over Tor to manage CPU and GPU miners, along with a kernel rootkit, a modified NVML wrapper, and a netns backdoor. The NVML wrapper filters process information and alters GPU telemetry to conceal mining activity, while netns retrieves its command-and-control address from an Ethereum smart contract.
What to do: Secure exposed Ray job agents and review Linux systems for unauthorized miners, modified NVIDIA monitoring output, kernel rootkits, and added SSH keys.
Sources: thehackernews.com
Unpatched GeoVision Camera Software Let Cryptojackers Compromise 13 Organizations
Who is affected: The article is aimed at security teams managing GeoVision devices, especially organizations in GeoVision’s 110-country customer base. Huntress reports that attackers exploited internet-exposed GeoVision camera management software to compromise Windows machines at 13 organizations.
Attackers exploited internet-exposed GeoVision camera management software to compromise Windows machines at 13 organizations, mostly between November 8 and 10, and install the XMRig cryptocurrency miner. The exploited software was an April 2024 build vulnerable to three unauthenticated remote-code-execution flaws, but it was not determined which flaw was used. The attackers also disabled security processes through IFEO debugger hijacks and deployed a resilient botnet-like loader.
What to do: Review affected systems for compromise and take steps to reduce exposure of GeoVision camera management software.
Sources: helpnetsecurity.com
Patches and Advisories
AI Finds 99 Flaws in TCP/IP Stacks Used Across Connected Devices
Who is affected: The affected TCP/IP stack software is used in industrial equipment, medical devices, network equipment and other connected products, including Siemens APOGEE, Desigo and TALON building automation controllers.
Forescout researchers used AI models and existing security tools to identify 99 vulnerabilities across 10 TCP/IP stacks, including 66 previously unknown vulnerabilities. Fifty-two percent of the newly discovered flaws were rated high or critical in severity. The analysis took about 10 hours, followed by approximately 80 hours of validation.
What to do: Security teams should identify, prioritise and mitigate cyber risks in affected legacy and difficult-to-patch devices.
Sources: helpnetsecurity.com
Google Project Zero Shortens Vulnerability Disclosure Deadlines
Who is affected: The policy applies to software vendors, with most grace periods and variants remaining and exceptions covering actively exploited bugs, publicly available details and some hardware flaws.
Google Project Zero will reduce its disclosure deadline from 90 days to 60 days on February 2, 2027, and to 30 days on January 3, 2028, regardless of whether a vendor has fixed the bug. It said large language models help attackers find and exploit vulnerabilities faster while its weekly discovery volume has risen.
What to do: Software vendors should plan for faster rollouts and other user protections.
Sources: helpnetsecurity.com
Microsoft Fixes Windows Activation Failures Caused by November 2026 Updates
Who is affected: Organizations that automate Windows activation with the SLMGR command-line tool are affected across specified Windows and Windows Server editions.
Microsoft released emergency out-of-band updates for activation and licensing failures introduced by its November 2026 security updates. The fixes cover Windows 10 21H2 and 22H2, Windows 11 23H2, Windows Enterprise LTSC, and Windows Server 2019, 2022, and 2025.
What to do: Apply the applicable emergency update to affected Windows systems and organizations.
Sources: bleepingcomputer.com
Rspamd 4.3.0 Improves DMARC Rejection and Authentication Header Handling
Who is affected: Rspamd administrators should test local scoring and scripts before upgrading because scoring, plugin symbol output, and DKIM d= value handling have changed.
Rspamd 4.3.0 prevents forwarded mail from softening DMARC rejects through improperly trusted ARC sets and stops unescaped sender text from injecting fake authentication results into Authentication-Results headers. It also ensures failed messages produce exactly one TEMPFAIL and lowercases DKIM d= values.
What to do: Test local scores and scripts before upgrading.
Sources: helpnetsecurity.com
NoisyTCP Flaw Lets Unauthenticated Attackers Crash Linux Servers
Who is affected: Almost all Linux distributions are affected, with servers running Nginx, Apache, or SSH with default configurations at risk.
According to Hyunwoo Kim, the security researcher who discovered it, CVE-2026-99288 is a pre-authenticated remote corruption flaw that stems from a NULL pointer dereference weakness in the Linux kernel’s TCP networking subsystem. Kim said the exploits cause a kernel panic and take down the host, but don’t allow threat actors to alter data or gain arbitrary code execution. He added that the race window is narrow, so triggering it against a quiet bare-metal server can take a long time. The researcher said it can be triggered against servers that have two or more RX queues, or where RPS distributes packets across multiple CPUs.
What to do: Apply the available patch or upgrade, or disable SYN cookies and use a single RX queue if an immediate update is not possible.
Sources: bleepingcomputer.com
Scams and Phishing
AI Agents Target Exposed AI Apps for API Keys
Who is affected: According to Amazon’s threat intelligence team, AI agents using the open-source ARTEX penetration-testing tool hunted for weaknesses in internet-exposed AI apps such as LiteLLM, One API, Dify, FastGPT, Langflow, RAGFlow, Open WebUI, LobeChat and Ollama to steal provider API keys, though the team says it could not verify the agents’ reported outcomes and has moderate confidence the operators were separate China-linked groups.
Amazon’s threat intelligence team said AI agents using the open-source ARTEX penetration-testing tool hunted for weaknesses in exposed AI applications. It could not verify the agents’ reported outcomes and has moderate confidence that the operators were separate China-linked groups.
What to do: Review exposed AI applications for provider API key exposure and rotate any keys that may have been exposed.
Sources: thehackernews.com
Microsoft 365 Phishing Emails Exploit Outage to Threaten Message Deletion
Who is affected: Similar phishing messages targeted organizations across several US sectors. Organizations and government agencies in several countries reported interruptions to email and other operations during the November 12 outage.
Microsoft said the Microsoft 365 outage was caused by a network capacity upgrade and a subsequent network change, not a cyberattack, and restored services in stages. Attackers sent phishing emails that referenced incident EX1550543 and falsely claimed held emails would be deleted within 24 hours unless recipients clicked a link. Microsoft identified three sending domains and recipient-specific SendGrid tracking links that redirected to another domain, but said credential collection and malware delivery were not confirmed.
What to do: Security teams should examine suspicious emails and links and investigate whether users clicked them.
Sources: securityweek.com
