Linux

В OpenBSD устранена локальная уявзимость, позволяющая получить права root Проект OpenBSD опубликовал исправление уязвимости (CVE-2026-57589) в ядре, затрагивающей реализацию системных вызовов для работы с семафорами System V (sem). Проблема вызвана обращением к уже освобождённой памяти в функции sys_semget() и может эксплуатироваться для получения прав root непривилегированным локальным пользователем в конфигурации по умолчанию. 👉@sysadminoff https://www.opennet.ru/opennews/art.shtml?num=65867

agentOS: Run AI Agents in a Secure Linux Sandbox The post agentOS: Run AI Agents in a Secure Linux Sandbox first appeared on Tecmint: Linux Howtos, Tutorials & Guides .I burned through half of my E2B sandbox budget in just two weeks by spinning up containers for AI agents The post agentOS: Run AI Agents in a Secure Linux Sandbox first appeared on Tecmint: Linux Howtos, Tutorials & Guides. 👉@sysadminoff https://www.tecmint.com/agentos-run-ai-coding-agents-secure-linux-sandbox/

Обновление NTFS-3G 2026.7.7 с устранением 9 уязвимостей Опубликован релиз пакета NTFS-3G 2026.7.7, включающего свободный драйвер, работающий в пространстве пользователя с использованием механизма FUSE, и комплект утилит ntfsprogs для манипуляций с разделами NTFS. Код проекта распространяется под лицензией GPLv2. В новой версии устранено 9 уязвимостей, которые могут привести к выполнению кода с правами root при обработке специально оформленных разделов или дисковых образов с ФС NTFS. 👉@sysadminoff https://www.opennet.ru/opennews/art.shtml?num=65911

📰 Linus Torvalds Reaffirms That Linux Is Not "Anti-AI" & Not A "Social Warrior" Project Overnight Linux creator Linus Torvalds wrote another well crafted message that reaffirms the Linux kernel position of not being against AI and lashing back against some kernel developers that are against AI/LLM usage within the kernel project... 🔗 Source: #kernel #linux 👉@sysadminoff https://www.phoronix.com/news/Linux-Is-Not-Anti-AI

OpenAI outperforms Anthropic in terms of intelligence-per-cost Anthropic currently holds a slight lead in raw model intelligence with Claude 3.5 Sonnet, yet a critical decision to limit infrastructure investment two years ago has created significant service hurdles. While Anthropic struggles with capacity constraints that limit user quotas, OpenAI is leveraging its superior compute reserves to offer more generous access and frequent usage resets. This disparity is forcing developers to choose between the highest benchmark performance and the practical reliability of more accessible platforms. Source 👉@sysadminoff https://4sysops.com/archives/openai-outperforms-anthropic-in-terms-of-intelligence-per-cost/

📰 Vocalinux 0.14 Beta Released For Offline Voice Dictation / Speech-To-Text On Linux Ubuntu 26.10 is notably working on laying the foundation for a context-aware desktop and their initial deliverable being worked on is Myna as a speech-to-text solution for the Linux desktop. Interestingly there is already a promising voice dictation / speech-to-text solution for the Linux desktop called Vocalinux that continues advancing and is usable right now for those looking at their own speech-to-text desktop solution... 🔗 Source: #linux #ubuntu 👉@sysadminoff https://www.phoronix.com/news/Vocalinux-0.14-Beta

Windows Server vNext preview: ReFS boot, NVMe-oF initiator, and Quick Machine Recovery Microsoft's Windows Server vNext preview program has introduced three significant features since build 29531, released on February 12, 2026: boot support for the Resilient File System (ReFS), a built-in NVMe-over-Fabrics (NVMe-oF) initiator for high-performance network storage, and Quick Machine Recovery (QMR) for automated recovery from boot failures. Build 29621, released July 13, 2026, adds an early implementation of Trusted Launch for Hyper-V virtual machines. This article covers what each feature does, its requirements, and its current limitations. Source 👉@sysadminoff https://4sysops.com/archives/windows-server-vnext-preview-refs-boot-nvme-of-initiator-and-quick-machine-recovery/

Обновление nginx 1.31.3 с устранением RCE-уязвимости Сформирован выпуск основной ветки nginx 1.31.3, в рамках которой продолжается развитие новых возможностей, а также выпуск параллельно поддерживаемой стабильной ветки nginx 1.30.4, в которую вносятся только изменения, связанные с устранением серьёзных ошибок и уязвимостей. В обновлениях устранено 3 уязвимости. 👉@sysadminoff https://www.opennet.ru/opennews/art.shtml?num=65910

📰 Firefox to Test a Two-Week Release Cycle Starting in September Mozilla plans to release Firefox Desktop and Android every two weeks in an experiment beginning with Firefox 155 on September 1. 🔗 Source: #android #firefox 👉@sysadminoff https://linuxiac.com/firefox-to-test-a-two-week-release-cycle-starting-in-september/

PrismML releases Bonsai 27B to run large language models on smartphones PrismML has launched Bonsai 27B, a 27-billion-parameter model based on Alibaba’s Qwen 3.6 that utilizes extreme quantization to run locally on consumer hardware. The release includes a 3.9 GB 1-bit binary variant designed for mobile devices and a 5.9 GB ternary variant optimized for laptops. By reducing weights to just one or two bits, these models bypass the high memory and bandwidth requirements that typically restrict large-scale AI to cloud infrastructure. Source 👉@sysadminoff https://4sysops.com/archives/prismml-releases-bonsai-27b-to-run-large-language-models-on-smartphones/

Firefox is getting updates twice as fast from September Mozilla is changing Firefox’s release schedule from September, with new stable versions planned every two weeks instead of once a month. The company says the new release cadence is designed to make releases more predictable and allow improvements to reach users more often. But in a message to developers, Mozilla described the changes as an experiment and said it will review the impact before deciding whether to continue. Firefox 154, due for release on 18 August, will be the final stable version released under the current schedule. Firefox 155 is then planned for 1 September, after which new stable releases […] You're reading Firefox is getting updates twice as fast from September, a blog post from OMG! Ubuntu. Do not reproduce elsewhere without permission. 👉@sysadminoff https://www.omgubuntu.co.uk/2026/07/firefox-two-week-release-cycle

📰 MKVToolNix 100 MKV Manipulation Tool Brings New Features and Enhancements MKVToolNix 100 open-source Matroska (MKV) manipulation software is now available for download with new MKVToolNix GUI features, mkvmerge improvements, and various other chnages. 🔗 Source: https://9to5linux.com/mkvtoolnix-100-mkv-manipulation-tool-brings-new-features-and-enhancements #opensource 👉@sysadminoff

📰 OpenMandriva's accused repo wrecker says it wasn't sabotage – it was a message Ex-contributor claims he wasn't a rogue admin, hadn't left the project, and never intended to harm the distro. 🔗 Source: #distro 👉@sysadminoff https://www.theregister.com/os-platforms/2026/07/15/openmandrivas-accused-repo-wrecker-says-it-wasnt-sabotage-it-was-a-message/5270511

OpenAI’s new AI companion device OpenAI is developing a screenless, mobile smart speaker designed to function as a human-like AI companion for the home. The device features mechanical elements that allow it to move autonomously, creating a more lifelike presence than traditional smart speakers. It utilizes integrated cameras and sensors to maintain contextual awareness of its surroundings and the user's activities. Source 👉@sysadminoff https://4sysops.com/archives/openais-new-ai-companion-device/

📰 FreeRDP 3.29 Is Out With Extensive Security Fixes The open-source Remote Desktop Protocol implementation FreeRDP rolls out 22 security advisories, stronger runtime protections, and multiple client fixes. 🔗 Source: #opensource 👉@sysadminoff https://linuxiac.com/freerdp-3-29-is-out-with-extensive-security-fixes/

📰 Valve adds a "Great on Machine" section to the Steam Machine library Valve continue to roll out changes to improve the user experience on the new Steam Machine, which now has a "Great on Machine" section in the Steam Library.Read the full article on GamingOnLinux. 🔗 Source: #steam 👉@sysadminoff https://www.gamingonlinux.com/2026/07/valve-adds-a-great-on-machine-section-to-the-steam-machine-library/

ZXC 0.13.0 и 0.13.1 14 и 15 июля, после почти месяца разработки, состоялись выпуски 0.13.0 и 0.13.1 SIMD-оптимизированной библиотеки и кроссплатформенной консольной утилиты ZXC (github.com), реализующих высокопроизводительное многопоточное асимметричное сжатие без потерь и оптимизированное для игровых ресурсов, прошивок и пакетов приложений. Формат разработан по принципу «один раз записать, многократно читать» (WORM). В отличие от таких кодеков, как LZ4, ZXC жертвует скоростью сжатия ради максимальной пропускной способности при распаковке. Декларируется скорость распаковки на 9-48% выше, чем у LZ4 с уровнем компрессии по умолчанию, с равным или более высоким коэффициентом сжатия. Главное изменение этих версий — новый уровень сжатия 7 (ультра) с использованием сравнительно недавно опубликованным SIMD-оптимизированным алгоритмом PivCo-Huffman («pivot-coded Huffman») (репозиторий), который позволяет превысить коэффициент сжатия zstd (-1), при этом скорость декодирования остаётся примерно в 2 раза выше, чем у zstd. Энтропийное декодирование было переработано, что позволило увеличить её скорость на уровне 6 на 5–13% на всех эталонных процессорах. Формат контейнера обновлён до версии 7 и это последнее запланированное изменение формата. Версия 0.13 является последним этапом перед выпуском версии 1.0, в которой формат контейнера ZXC будет зафиксирован, что обеспечит его стабильность и обратную совместимость. Проект написан на языке C и распространяется по лицензии BSD 3. ( читать дальше... )  c, zxc, компрессия, консоль, утилита 👉@sysadminoff https://www.linux.org.ru/news/opensource/18339089

Apple evaluates PrismML technology to run 27B parameter AI models on iPhone Apple is exploring a partnership with PrismML, a Caltech spinout, to integrate extreme model-compression technology into future iPhone hardware. The startup utilizes ultra-low-bit weight architectures, such as 1-bit and ternary representations, to drastically reduce the memory footprint of large language models. This mathematical approach allows a 27-billion-parameter model like Alibaba’s Qwen 3.6 to shrink from 54 GB to under 4 GB, making it compatible with the 8 GB of RAM found in current Pro-tier iPhones. Source 👉@sysadminoff https://4sysops.com/archives/apple-evaluates-prismml-technology-to-run-27b-parameter-ai-models-on-iphone/

📰 After almost two years, KDE Plasma’s animations finally look good again on Nvidia cards If you've used KDE Plasma on an Nvidia card lately and felt the animations weren't as smooth as they could be, it's not just you. Ever since October 2024, triple buffering has been disabled for Nvidia hardware due to a bug. Fortunately, the fix has just been released as part of Plasma 6.7.3, and you can grab it now. 🔗 Source: #kde #plasma 👉@sysadminoff https://www.xda-developers.com/after-almost-two-years-kde-plasmas-animations-finally-look-good-again-on-nvidia-cards/

ClickFix malware kits industrialize social engineering through fake browser errors ClickFix has evolved from a simple social engineering trick into a sophisticated Malware-as-a-Service ecosystem that bypasses traditional antivirus defenses. The attack convinces users to manually execute malicious code by presenting fake CAPTCHAs, browser crashes, or software update errors. Victims are instructed to copy a hidden command and paste it directly into the Windows Run dialog or a terminal window, which triggers a payload via trusted system tools. Source 👉@sysadminoff https://4sysops.com/archives/clickfix-malware-kits-industrialize-social-engineering-through-fake-browser-errors/