[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-logicalclocks--hopsworks":3,"tool-logicalclocks--hopsworks":64},[4,17,27,35,43,56],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":16},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,3,"2026-04-05T11:01:52",[13,14,15],"开发框架","图像","Agent","ready",{"id":18,"name":19,"github_repo":20,"description_zh":21,"stars":22,"difficulty_score":23,"last_commit_at":24,"category_tags":25,"status":16},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",140436,2,"2026-04-05T23:32:43",[13,15,26],"语言模型",{"id":28,"name":29,"github_repo":30,"description_zh":31,"stars":32,"difficulty_score":23,"last_commit_at":33,"category_tags":34,"status":16},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",107662,"2026-04-03T11:11:01",[13,14,15],{"id":36,"name":37,"github_repo":38,"description_zh":39,"stars":40,"difficulty_score":23,"last_commit_at":41,"category_tags":42,"status":16},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[13,26],{"id":44,"name":45,"github_repo":46,"description_zh":47,"stars":48,"difficulty_score":23,"last_commit_at":49,"category_tags":50,"status":16},2268,"ML-For-Beginners","microsoft\u002FML-For-Beginners","ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程，旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周，包含 26 节精炼课程和 52 道配套测验，内容涵盖从基础概念到实际应用的完整流程，有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。\n\n无论是希望转型的开发者、需要补充算法背景的研究人员，还是对人工智能充满好奇的普通爱好者，都能从中受益。课程不仅提供了清晰的理论讲解，还强调动手实践，让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持，通过自动化机制提供了包括简体中文在内的 50 多种语言版本，极大地降低了全球不同背景用户的学习门槛。此外，项目采用开源协作模式，社区活跃且内容持续更新，确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路，ML-For-Beginners 将是理想的起点。",84991,"2026-04-05T10:45:23",[14,51,52,53,15,54,26,13,55],"数据工具","视频","插件","其他","音频",{"id":57,"name":58,"github_repo":59,"description_zh":60,"stars":61,"difficulty_score":10,"last_commit_at":62,"category_tags":63,"status":16},3128,"ragflow","infiniflow\u002Fragflow","RAGFlow 是一款领先的开源检索增强生成（RAG）引擎，旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体（Agent）能力相结合，不仅支持从各类文档中高效提取知识，还能让模型基于这些知识进行逻辑推理和任务执行。\n\n在大模型应用中，幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构（如表格、图表及混合排版），显著提升了信息检索的准确度，从而有效减少模型“胡编乱造”的现象，确保回答既有据可依又具备时效性。其内置的智能体机制更进一步，使系统不仅能回答问题，还能自主规划步骤解决复杂问题。\n\n这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统，还是致力于探索大模型在垂直领域落地的创新者，都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口，既降低了非算法背景用户的上手门槛，也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目，它正成为连接通用大模型与行业专有知识之间的重要桥梁。",77062,"2026-04-04T04:44:48",[15,14,13,26,54],{"id":65,"github_repo":66,"name":67,"description_en":68,"description_zh":69,"ai_summary_zh":70,"readme_en":71,"readme_zh":72,"quickstart_zh":73,"use_case_zh":74,"hero_image_url":75,"owner_login":76,"owner_name":77,"owner_avatar_url":78,"owner_bio":79,"owner_company":80,"owner_location":80,"owner_email":81,"owner_twitter":67,"owner_website":82,"owner_url":83,"languages":84,"stars":105,"forks":106,"last_commit_at":107,"license":108,"difficulty_score":109,"env_os":110,"env_gpu":111,"env_ram":112,"env_deps":113,"category_tags":118,"github_topics":119,"view_count":23,"oss_zip_url":80,"oss_zip_packed_at":80,"status":16,"created_at":136,"updated_at":137,"faqs":138,"releases":167},3449,"logicalclocks\u002Fhopsworks","hopsworks","Hopsworks - Data-Intensive AI platform with a Feature Store","Hopsworks 是一款专为数据密集型人工智能打造的实时 AI Lakehouse 平台，其核心亮点在于提供了以 Python 为中心的特征存储（Feature Store）和完整的 MLOps 能力。在机器学习开发中，团队常面临特征数据难以复用、模型管理混乱以及协作效率低下的挑战，Hopsworks 通过构建一个安全、受管控的统一平台有效解决了这些问题。它不仅支持独立作为特征存储使用，还能帮助用户高效管理、治理和部署模型，同时涵盖特征管道与训练管道的开发与运维全流程。\n\n该平台特别适合机器学习工程师、数据科学家以及需要协同工作的 AI 研发团队使用。无论是希望快速上手的个人开发者，还是对数据合规与安全有严格要求的企业用户，都能从中受益。Hopsworks 具备高度的模块化特性，既支持一键安装的本地化部署，让用户在自有基础设施上运行以获得最大控制权，也提供无缝集成 AWS、Azure、GCP 等主流云环境的托管服务，并能与 Databricks、SageMaker 等第三方平台灵活对接。此外，其推出的无服务器版本让新用户仅需注册即可立即体验教程与核心功能，极大降低了探索门槛。通过集中","Hopsworks 是一款专为数据密集型人工智能打造的实时 AI Lakehouse 平台，其核心亮点在于提供了以 Python 为中心的特征存储（Feature Store）和完整的 MLOps 能力。在机器学习开发中，团队常面临特征数据难以复用、模型管理混乱以及协作效率低下的挑战，Hopsworks 通过构建一个安全、受管控的统一平台有效解决了这些问题。它不仅支持独立作为特征存储使用，还能帮助用户高效管理、治理和部署模型，同时涵盖特征管道与训练管道的开发与运维全流程。\n\n该平台特别适合机器学习工程师、数据科学家以及需要协同工作的 AI 研发团队使用。无论是希望快速上手的个人开发者，还是对数据合规与安全有严格要求的企业用户，都能从中受益。Hopsworks 具备高度的模块化特性，既支持一键安装的本地化部署，让用户在自有基础设施上运行以获得最大控制权，也提供无缝集成 AWS、Azure、GCP 等主流云环境的托管服务，并能与 Databricks、SageMaker 等第三方平台灵活对接。此外，其推出的无服务器版本让新用户仅需注册即可立即体验教程与核心功能，极大降低了探索门槛。通过集中管理特征、模型、训练数据及日志等资产，Hopsworks 真正实现了机器学习全生命周期的标准化与协作化。","\u003Cp align=\"center\">\n    \u003Ca href=\"https:\u002F\u002Fhopsworks.ai\">\n        \u003Cimg src=\"https:\u002F\u002Fuploads-ssl.webflow.com\u002F5f6353590bb01cacbcecfbac\u002F6202a13e7cafec5553703f6b_logo.svg\" width=\"55%\" >\n    \u003C\u002Fa>\n\u003C\u002Fp>\n\u003Cbr \u002F>\n\n\u003Cp align=\"center\">\n    \u003Ca href=\"https:\u002F\u002Fhopsworks.ai\" alt=\"hopsworks.ai\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fhopsworks-ai-brightgreen\" \u002F>\u003C\u002Fa>\n     \u003Ca href=\"https:\u002F\u002Fapp.hopsworks.ai\" alt=\"app\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FHopsworks-app-green\" \u002F>\u003C\u002Fa>    \n    \u003Ca href=\"https:\u002F\u002Fdocs.hopsworks.ai\" alt=\"docs.hopsworks.ai\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fhopsworks-docs-orange\" \u002F>\u003C\u002Fa>\n    \u003Ca href=\"https:\u002F\u002Fcommunity.hopsworks.ai\" alt=\"community.hopsworks.ai\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fhopsworks-community-blueviolet\" \u002F>\u003C\u002Fa>\n    \u003Ca href=\"https:\u002F\u002Ftwitter.com\u002Fhopsworks\" alt=\"Hopsworks Twitter\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fhopsworks-twitter-blue\" \u002F>\u003C\u002Fa>\n    \u003Ca href=\"https:\u002F\u002Fbit.ly\u002Fpublichopsworks\" alt=\"Hopsworks Slack\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fstatic\u002Fv1?label=Hopsworks&message=Slack&color=36C5F0\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\n\n# Quick Install\nGet up and running with a single command:\n```bash\ncurl -O https:\u002F\u002Fraw.githubusercontent.com\u002Flogicalclocks\u002Fhopsworks-k8s-installer\u002Fmaster\u002Finstall-hopsworks.py\npython3 install-hopsworks.py\n```\n\n\n\u003Ca name=\"what\">\u003C\u002Fa>\n# What is Hopsworks?\n\nHopsworks is a **Real-Time AI Lakehouse** for ML with a **Python-centric Feature Store** and MLOps capabilities. Hopsworks is a modular platform. You can use it as a standalone Feature Store, you can use it to manage, govern, and serve your models, and you can even use it to develop and operate feature pipelines and training pipelines. Hopsworks brings collaboration for ML teams, providing a secure, governed platform for developing, managing, and sharing ML assets - features, models, training data, batch scoring data, logs, and more.\n\u003Cbr \u002F>\n\u003Cp align=\"center\" style=\"background-color:white; border-radius:4px;\">\n\u003Cimg src=\"https:\u002F\u002Fuploads-ssl.webflow.com\u002F5f6353590bb01cacbcecfbac\u002F62f21c38bc47b2d313fbf76d_Marchitecture%20-%20readme.svg\" width=\"90%\">\n\u003C\u002Fp>\n\u003Cbr \u002F>\n\n\u003Ca name=\"quick\">\u003C\u002Fa>\n# 🚀 Quickstart\n\n## **APP - Serverless (beta)**\n### → **Go to [app.hopsworks.ai](https:\u002F\u002Fapp.hopsworks.ai)**\nHopsworks is available as a serverless app, simply head to [app.hopsworks.ai](https:\u002F\u002Fapp.hopsworks.ai) and register with your **Gmail** or **Github** accounts. You will then be able to run a tutorial or access Hopsworks directly and try yourself. This is the preferred way to first experience the platform before diving into more advanced uses and installation requirements. \n\n## **Azure, AWS & GCP**\n[Managed Hopsworks](https:\u002F\u002Fmanaged.hopsworks.ai) is our platform for running Hopsworks and the Feature Store in the cloud and integrates directly with the customer AWS\u002FAzure\u002FGCP environment. It also integrates seamlessly with third party platforms such as Databricks, SageMaker and KubeFlow.\n\nIf you wish to run Hopsworks on your Azure, AWS or GCP environment, follow one of the following guides in our documentation:\n- [AWS Guide](https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fsetup_installation\u002Faws\u002Fgetting_started\u002F#step-1-connecting-your-aws-account)\n- [Azure Guide](https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fsetup_installation\u002Fazure\u002Fgetting_started\u002F#step-1-connecting-your-azure-account)\n- [GCP Guide](https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fsetup_installation\u002Fgcp\u002Fgetting_started\u002F#step-1-connecting-your-gcp-account)\n\n## **Installer - On-premise**\nIt is possible to use Hopsworks on-premises, which means that companies can run their machine learning workloads on their own hardware and infrastructure, rather than relying on a cloud provider. This can provide greater flexibility, control, and cost savings, as well as enabling companies to meet specific compliance and security requirements.\n\nWorking on-premises with Hopsworks typically involves collaboration with the Hopsworks engineering teams, as each infrastructure is unique and requires a tailored approach to deployment and configuration. The process begins with an assessment of the company's existing infrastructure and requirements, including network topology, security policies, and hardware specifications.\n\nFor further details about on-premise installations: [contact us](https:\u002F\u002Fwww.hopsworks.ai\u002Fcontact).\n\n### **Requirements**\nYou need at least one server or virtual machine on which Hopsworks will be installed with at least the following specification:\n- Centos\u002FRHEL 8.x or Ubuntu 22.04;\n- at least 32GB RAM,\n- at least 8 CPUs,\n- 100 GB of free hard-disk space,\n- a UNIX user account with sudo privileges.\n\u003Cbr \u002F>\n\n\u003Ca name=\"docs\">\u003C\u002Fa>\n# 🎓 Documentation and API\n### **Documentation**\n[Hopsworks documentation](https:\u002F\u002Fdocs.hopsworks.ai) includes user guides, feature store documentation and an administration guide. We also include concepts to help user navigates the abstractions and logics of the feature stores and MLOps in general:\n- **Feature Store:** [https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fconcepts\u002Ffs\u002F](https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fconcepts\u002Ffs\u002F)\n- **Projects:** [https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fconcepts\u002Fprojects\u002Fgovernance\u002F](https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fconcepts\u002Fprojects\u002Fgovernance\u002F)\n- **MLOps:** [https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fconcepts\u002Fmlops\u002Fprediction_services\u002F](https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fconcepts\u002Fmlops\u002Fprediction_services\u002F)\n\n### **APIs**\nHopsworks API documentation is divided in 3 categories; Hopsworks API covers project level APIs, Feature Store API covers feature groups, feature views and connectors, and finally MLOps API covers Model Registry, serving and deployment. \n- **Hopsworks API** - [https:\u002F\u002Fdocs.hopsworks.ai\u002Fhopsworks-api\u002Flatest.1\u002Fgenerated\u002Fapi\u002Fconnection\u002F](https:\u002F\u002Fdocs.hopsworks.ai\u002Fhopsworks-api\u002Flatest.1\u002Fgenerated\u002Fapi\u002Fconnection\u002F)\n- **Feature Store API** - [https:\u002F\u002Fdocs.hopsworks.ai\u002Ffeature-store-api\u002Flatest\u002Fgenerated\u002Fapi\u002Fconnection_api\u002F](https:\u002F\u002Fdocs.hopsworks.ai\u002Ffeature-store-api\u002Flatest\u002Fgenerated\u002Fapi\u002Fconnection_api\u002F)\n- **MLOps API** - [https:\u002F\u002Fdocs.hopsworks.ai\u002Fmachine-learning-api\u002Flatest\u002Fgenerated\u002Fconnection_api\u002F](https:\u002F\u002Fdocs.hopsworks.ai\u002Fmachine-learning-api\u002Flatest\u002Fgenerated\u002Fconnection_api\u002F)\n\n### **Tutorials**\nMost of the tutorials require you to have at least an account on [app.hopsworks.ai](https:\u002F\u002Fapp.hopsworks.ai). You can explore the dedicated [https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials) repository containing our tutorials or jump directly in one of the existing use cases:\n- Fraud (batch): [https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials\u002Ftree\u002Fmaster\u002Ffraud_batch](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials\u002Ftree\u002Fmaster\u002Ffraud_batch)\n- Fraud (online): [https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials\u002Ftree\u002Fmaster\u002Ffraud_online](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials\u002Ftree\u002Fmaster\u002Ffraud_online)\n- Churn prediction [https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials\u002Ftree\u002Fmaster\u002Fchurn](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials\u002Ftree\u002Fmaster\u002Fchurn)\n\u003Cbr \u002F>\n\n\u003Ca name=\"features\">\u003C\u002Fa>\n# 📦 Main Features\n\n### **Project-based Multi-Tenancy and Team Collaboration**\nHopsworks provides projects as a secure sandbox in which teams can collaborate and share ML assets. Hopsworks' unique multi-tenant project model even enables sensitive data to be stored in a shared cluster, while still providing fine-grained sharing capabilities for ML assets across project boundaries. Projects can be used to structure teams so that they have end-to-end responsibility from raw data to managed features and models. Projects can also be used to create development, staging, and production environments for data teams. All ML assets support versioning, lineage, and provenance provide all Hopsworks users with a complete view of the MLOps life cycle, from feature engineering through model serving.\n\n### **Development and Operations**\nHopsworks provides development tools for Data Science, including conda environments for Python, Jupyter notebooks, jobs, or even notebooks as jobs. You can build production pipelines with the bundled Airflow, and even run ML training pipelines with GPUs in notebooks on Airflow. You can train models on as many GPUs as are installed in a Hopsworks cluster and easily share them among users. You can also run Spark, Spark Streaming, or Flink programs on Hopsworks, with support for elastic workers in the cloud (add\u002Fremove workers dynamically).\n\n### **Available on any Platform**\nHopsworks is available as a both managed platform in the cloud on AWS, Azure, and GCP, and can be installed on any Linux-based virtual machines (Ubuntu\u002FRedhat compatible), even in air-gapped data centers. Hopsworks is also available as a serverless platform that manages and serves both your features and models.\n\u003Cbr \u002F>\n\n\u003Ca name=\"community\">\u003C\u002Fa>\n# 🧑‍🤝‍🧑 Community\n\n### **Contribute**\nWe are building the most complete and modular ML platform available in the market, and we count on your support to continuously improve Hopsworks. Feel free to give us suggestions, [report bugs](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks\u002Fissues) and [add features to our library](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Ffeature-store-api) anytime.\n\n### **Join the community**\n- Ask questions and give us feedback in the [Hopsworks Community](https:\u002F\u002Fcommunity.hopsworks.ai\u002F)\n- Join our Public [Slack Channel](https:\u002F\u002Fjoin.slack.com\u002Ft\u002Fpublic-hopsworks\u002Fshared_invite\u002Fzt-24fc3hhyq-VBEiN8UZlKsDrrLvtU4NaA)\n- Follow us on [Twitter](https:\u002F\u002Ftwitter.com\u002Fhopsworks)\n- Check out all our latest [product releases](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks\u002Freleases)\n\n### **Open-Source**\nHopsworks is available under the **AGPL-V3 license**. In plain English this means that you are free to use Hopsworks and even build paid services on it, but if you modify the source code, you should also release back your changes and any systems built around it as AGPL-V3.\n","\u003Cp align=\"center\">\n    \u003Ca href=\"https:\u002F\u002Fhopsworks.ai\">\n        \u003Cimg src=\"https:\u002F\u002Fuploads-ssl.webflow.com\u002F5f6353590bb01cacbcecfbac\u002F6202a13e7cafec5553703f6b_logo.svg\" width=\"55%\" >\n    \u003C\u002Fa>\n\u003C\u002Fp>\n\u003Cbr \u002F>\n\n\u003Cp align=\"center\">\n    \u003Ca href=\"https:\u002F\u002Fhopsworks.ai\" alt=\"hopsworks.ai\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fhopsworks-ai-brightgreen\" \u002F>\u003C\u002Fa>\n     \u003Ca href=\"https:\u002F\u002Fapp.hopsworks.ai\" alt=\"app\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FHopsworks-app-green\" \u002F>\u003C\u002Fa>    \n    \u003Ca href=\"https:\u002F\u002Fdocs.hopsworks.ai\" alt=\"docs.hopsworks.ai\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fhopsworks-docs-orange\" \u002F>\u003C\u002Fa>\n    \u003Ca href=\"https:\u002F\u002Fcommunity.hopsworks.ai\" alt=\"community.hopsworks.ai\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fhopsworks-community-blueviolet\" \u002F>\u003C\u002Fa>\n    \u003Ca href=\"https:\u002F\u002Ftwitter.com\u002Fhopsworks\" alt=\"Hopsworks Twitter\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fhopsworks-twitter-blue\" \u002F>\u003C\u002Fa>\n    \u003Ca href=\"https:\u002F\u002Fbit.ly\u002Fpublichopsworks\" alt=\"Hopsworks Slack\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fstatic\u002Fv1?label=Hopsworks&message=Slack&color=36C5F0\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\n\n# 快速安装\n只需一条命令即可快速启动并运行：\n```bash\ncurl -O https:\u002F\u002Fraw.githubusercontent.com\u002Flogicalclocks\u002Fhopsworks-k8s-installer\u002Fmaster\u002Finstall-hopsworks.py\npython3 install-hopsworks.py\n```\n\n\n\u003Ca name=\"what\">\u003C\u002Fa>\n# 什么是 Hopsworks？\n\nHopsworks 是一个面向机器学习的 **实时 AI 数据湖仓**，配备以 Python 为中心的特征存储和 MLOps 能力。Hopsworks 是一个模块化平台。您可以将其用作独立的特征存储，也可以用于管理、治理和部署您的模型，甚至可以用来开发和运维特征管道与训练管道。Hopsworks 为机器学习团队提供协作支持，打造一个安全、受治理的平台，用于开发、管理和共享机器学习资产——包括特征、模型、训练数据、批量评分数据、日志等。\n\u003Cbr \u002F>\n\u003Cp align=\"center\" style=\"background-color:white; border-radius:4px;\">\n\u003Cimg src=\"https:\u002F\u002Fuploads-ssl.webflow.com\u002F5f6353590bb01cacbcecfbac\u002F62f21c38bc47b2d313fbf76d_Marchitecture%20-%20readme.svg\" width=\"90%\">\n\u003C\u002Fp>\n\u003Cbr \u002F>\n\n\u003Ca name=\"quick\">\u003C\u002Fa>\n# 🚀 快速入门\n\n## **APP - 无服务器（测试版）**\n### → **前往 [app.hopsworks.ai](https:\u002F\u002Fapp.hopsworks.ai)**\nHopsworks 现已推出无服务器应用版本，您只需访问 [app.hopsworks.ai](https:\u002F\u002Fapp.hopsworks.ai)，使用您的 **Gmail** 或 **Github** 账户注册即可。注册后，您就可以运行教程或直接体验 Hopsworks。在深入探索更高级的功能及安装要求之前，这是体验该平台的首选方式。\n\n## **Azure、AWS 和 GCP**\n[Managed Hopsworks](https:\u002F\u002Fmanaged.hopsworks.ai) 是我们在云端运行 Hopsworks 和特征存储的平台，可直接与客户的 AWS\u002FAzure\u002FGCP 环境集成。它还能无缝对接 Databricks、SageMaker 和 KubeFlow 等第三方平台。\n\n如果您希望在 Azure、AWS 或 GCP 环境中运行 Hopsworks，请参考我们文档中的以下指南：\n- [AWS 指南](https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fsetup_installation\u002Faws\u002Fgetting_started\u002F#step-1-connecting-your-aws-account)\n- [Azure 指南](https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fsetup_installation\u002Fazure\u002Fgetting_started\u002F#step-1-connecting-your-azure-account)\n- [GCP 指南](https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fsetup_installation\u002Fgcp\u002Fgetting_started\u002F#step-1-connecting-your-gcp-account)\n\n## **安装程序 - 本地部署**\nHopsworks 支持本地部署，这意味着企业可以在自有硬件和基础设施上运行机器学习工作负载，而不必依赖云服务提供商。这种方式能够带来更高的灵活性、更强的控制能力以及成本节约，同时帮助企业满足特定的合规性和安全性要求。\n\n在本地环境中使用 Hopsworks 通常需要与 Hopsworks 工程团队密切合作，因为每个企业的基础设施都独一无二，部署和配置也需要量身定制。这一过程始于对企业现有基础设施和需求的评估，包括网络拓扑、安全策略以及硬件规格。\n\n如需了解更多关于本地部署的信息，请 [联系我们](https:\u002F\u002Fwww.hopsworks.ai\u002Fcontact)。\n\n### **系统要求**\n您至少需要一台服务器或虚拟机来安装 Hopsworks，其最低配置如下：\n- CentOS\u002FRHEL 8.x 或 Ubuntu 22.04；\n- 至少 32GB 内存，\n- 至少 8 个 CPU 核心，\n- 100 GB 的可用磁盘空间，\n- 具有 sudo 权限的 UNIX 用户账户。\n\u003Cbr \u002F>\n\n\u003Ca name=\"docs\">\u003C\u002Fa>\n# 🎓 文档与 API\n### **文档**\n[Hopsworks 文档](https:\u002F\u002Fdocs.hopsworks.ai)包含用户指南、特征存储文档以及管理指南。此外，我们还提供了相关概念，帮助用户理解特征存储和 MLOps 的抽象逻辑：\n- **特征存储：** [https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fconcepts\u002Ffs\u002F](https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fconcepts\u002Ffs\u002F)\n- **项目：** [https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fconcepts\u002Fprojects\u002Fgovernance\u002F](https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fconcepts\u002Fprojects\u002Fgovernance\u002F)\n- **MLOps：** [https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fconcepts\u002Fmlops\u002Fprediction_services\u002F](https:\u002F\u002Fdocs.hopsworks.ai\u002Flatest\u002Fconcepts\u002Fmlops\u002Fprediction_services\u002F)\n\n### **API**\nHopsworks API 文档分为三大类：Hopsworks API 涵盖项目级别的接口，Feature Store API 涉及特征组、特征视图和连接器，而 MLOps API 则专注于模型注册表、模型服务和部署。\n- **Hopsworks API** - [https:\u002F\u002Fdocs.hopsworks.ai\u002Fhopsworks-api\u002Flatest.1\u002Fgenerated\u002Fapi\u002Fconnection\u002F](https:\u002F\u002Fdocs.hopsworks.ai\u002Fhopsworks-api\u002Flatest.1\u002Fgenerated\u002Fapi\u002Fconnection\u002F)\n- **Feature Store API** - [https:\u002F\u002Fdocs.hopsworks.ai\u002Ffeature-store-api\u002Flatest\u002Fgenerated\u002Fapi\u002Fconnection_api\u002F](https:\u002F\u002Fdocs.hopsworks.ai\u002Ffeature-store-api\u002Flatest\u002Fgenerated\u002Fapi\u002Fconnection_api\u002F)\n- **MLOps API** - [https:\u002F\u002Fdocs.hopsworks.ai\u002Fmachine-learning-api\u002Flatest\u002Fgenerated\u002Fconnection_api\u002F](https:\u002F\u002Fdocs.hopsworks.ai\u002Fmachine-learning-api\u002Flatest\u002Fgenerated\u002Fconnection_api\u002F)\n\n### **教程**\n大多数教程要求您至少在 [app.hopsworks.ai](https:\u002F\u002Fapp.hopsworks.ai) 上拥有一个账户。您可以浏览专门的 [https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials) 仓库，其中包含了我们的教程，或者直接进入现有的用例之一：\n- 欺诈检测（批处理）：[https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials\u002Ftree\u002Fmaster\u002Ffraud_batch](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials\u002Ftree\u002Fmaster\u002Ffraud_batch)\n- 欺诈检测（在线）：[https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials\u002Ftree\u002Fmaster\u002Ffraud_online](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials\u002Ftree\u002Fmaster\u002Ffraud_online)\n- 客户流失预测：[https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials\u002Ftree\u002Fmaster\u002Fchurn](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials\u002Ftree\u002Fmaster\u002Fchurn)\n\u003Cbr \u002F>\n\n\u003Ca name=\"features\">\u003C\u002Fa>\n# 📦 主要特性\n\n### **基于项目的企业级多租户与团队协作**\nHopsworks 提供项目作为安全的沙盒环境，团队可以在其中协作并共享机器学习资产。Hopsworks 独特的多租户项目模型甚至允许将敏感数据存储在共享集群中，同时仍能实现跨项目边界的细粒度 ML 资产共享能力。项目可用于组织团队，使他们从原始数据到管理特征和模型都拥有端到端的责任。此外，项目还可以用于为数据团队创建开发、预发布和生产环境。所有 ML 资产均支持版本控制、血缘关系追踪和溯源功能，从而为所有 Hopsworks 用户提供从特征工程到模型服务的完整 MLOps 生命周期视图。\n\n### **开发与运维**\nHopsworks 为数据科学提供了丰富的开发工具，包括 Python 的 conda 环境、Jupyter 笔记本、作业调度任务，甚至可以直接将笔记本作为作业运行。您可以使用内置的 Airflow 构建生产级流水线，甚至在 Airflow 上通过 GPU 运行机器学习训练任务。您可以在 Hopsworks 集群中利用所有已安装的 GPU 训练模型，并轻松地在用户之间共享这些模型。此外，您还可以在 Hopsworks 上运行 Spark、Spark Streaming 或 Flink 程序，并支持云端弹性计算节点（动态添加或移除工作节点）。\n\n### **跨平台部署**\nHopsworks 既可在 AWS、Azure 和 GCP 等公有云上以托管平台的形式使用，也可以安装在任何基于 Linux 的虚拟机上（兼容 Ubuntu\u002FRedhat），甚至适用于气隙数据中心。Hopsworks 还提供无服务器平台，可管理和部署您的特征及模型。\n\u003Cbr \u002F>\n\n\u003Ca name=\"community\">\u003C\u002Fa>\n# 🧑‍🤝‍🧑 社区\n\n### **贡献**\n我们正在构建市场上最全面、最模块化的机器学习平台，并期待您的支持，持续改进 Hopsworks。欢迎您随时提出建议、[报告问题](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks\u002Fissues) 或为我们的库[添加新功能](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Ffeature-store-api)。\n\n### **加入社区**\n- 在 [Hopsworks 社区](https:\u002F\u002Fcommunity.hopsworks.ai\u002F) 提问并反馈；\n- 加入我们的公共 [Slack 频道](https:\u002F\u002Fjoin.slack.com\u002Ft\u002Fpublic-hopsworks\u002Fshared_invite\u002Fzt-24fc3hhyq-VBEiN8UZlKsDrrLvtU4NaA)；\n- 关注我们的 [Twitter](https:\u002F\u002Ftwitter.com\u002Fhopsworks)；\n- 查看我们所有的最新 [产品发布](https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks\u002Freleases)。\n\n### **开源**\nHopsworks 采用 **AGPL-V3 许可证** 开源。简单来说，这意味着您可以自由使用 Hopsworks，甚至在其基础上构建付费服务；但如果您对源代码进行修改，则必须以 AGPL-V3 许可证公开您的修改内容以及基于此构建的所有系统。","# Hopsworks 快速上手指南\n\nHopsworks 是一个**实时 AI 数据湖仓（AI Lakehouse）**，核心提供以 Python 为中心的特征存储（Feature Store）和 MLOps 能力。它支持团队协作、模型治理、特征管道开发及模型服务化，适用于从独立特征存储到完整 ML 生命周期管理的多种场景。\n\n## 1. 环境准备\n\n在本地或服务器部署前，请确保满足以下硬件与系统要求：\n\n*   **操作系统**：CentOS\u002FRHEL 8.x 或 Ubuntu 22.04\n*   **内存**：至少 32GB RAM\n*   **CPU**：至少 8 核\n*   **磁盘空间**：至少 100GB 可用空间\n*   **用户权限**：拥有 `sudo` 权限的 UNIX 用户账户\n*   **依赖环境**：已安装 Python 3 和 `curl`\n\n> **提示**：如果您希望快速体验而无需配置基础设施，推荐直接使用 [Hopsworks Serverless App](https:\u002F\u002Fapp.hopsworks.ai)（支持 Gmail 或 GitHub 账号登录），这是最便捷的入门方式。\n\n## 2. 安装步骤\n\n对于希望在自有基础设施（On-premise）上运行的用户，可以通过单条命令脚本快速启动安装程序。\n\n执行以下命令下载并运行安装脚本：\n\n```bash\ncurl -O https:\u002F\u002Fraw.githubusercontent.com\u002Flogicalclocks\u002Fhopsworks-k8s-installer\u002Fmaster\u002Finstall-hopsworks.py\npython3 install-hopsworks.py\n```\n\n*注：该安装程序将自动处理 Kubernetes 集群的配置与 Hopsworks 组件的部署。生产环境部署通常建议联系官方工程团队进行定制化评估。*\n\n## 3. 基本使用\n\n安装完成后（或登录 Serverless App 后），您可以通过以下方式开始使用：\n\n### 访问平台\n打开浏览器访问您的 Hopsworks 实例地址（本地部署通常为 `https:\u002F\u002F\u003Cyour-ip>:443`，云端则直接访问 App 地址）。\n\n### 运行教程示例\nHopsworks 提供了丰富的开源教程，涵盖欺诈检测、客户流失预测等场景。您可以克隆官方教程仓库并直接在 Jupyter Notebook 中运行：\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks-tutorials.git\ncd hopsworks-tutorials\n```\n\n**典型使用流程：**\n1.  **创建项目**：在 UI 中创建一个新项目（Project），作为团队协作和安全隔离的沙箱。\n2.  **导入特征**：使用 Python API 连接特征存储，上传或注册特征组（Feature Groups）。\n   ```python\n   import hopsworks\n   \n   # 连接到 Hopsworks\n   project = hopsworks.login()\n   fs = project.get_feature_store()\n   \n   # 获取或创建特征视图并进行训练\n   feature_view = fs.get_feature_view(name=\"fraud_detection\", version=1)\n   X_train, X_test, y_train, y_test = feature_view.train_test_split(test_size=0.2)\n   ```\n3.  **开发与训练**：利用内置的 Jupyter Notebook 环境（支持 Conda 和 GPU）进行模型训练。\n4.  **模型注册与服务**：将训练好的模型注册到 Model Registry，并一键部署为在线预测服务。\n\n更多详细 API 文档请参考：\n*   [Hopsworks 文档中心](https:\u002F\u002Fdocs.hopsworks.ai)\n*   [Feature Store API 文档](https:\u002F\u002Fdocs.hopsworks.ai\u002Ffeature-store-api\u002Flatest\u002Fgenerated\u002Fapi\u002Fconnection_api\u002F)\n*   [MLOps API 文档](https:\u002F\u002Fdocs.hopsworks.ai\u002Fmachine-learning-api\u002Flatest\u002Fgenerated\u002Fconnection_api\u002F)","某电商平台的推荐算法团队正面临从离线批量训练向实时个性化推荐转型的挑战，急需解决特征数据在训练与推理阶段不一致的难题。\n\n### 没有 hopsworks 时\n- **特征重复开发**：数据工程师为离线训练和在线推理分别编写两套特征计算代码，导致逻辑不一致，经常出现“训练 - 推理偏差”。\n- **协作效率低下**：特征定义散落在不同的 Notebook 和脚本中，新成员难以查找复用现有特征，每次新项目都要重新造轮子。\n- **数据治理缺失**：缺乏统一的版本控制和血缘追踪，当模型效果下降时，无法快速定位是哪些特征数据变更导致了问题。\n- **部署周期漫长**：将特征从实验环境迁移到生产环境需要手动配置复杂的管道，上线一个新模型往往需要数周时间。\n\n### 使用 hopsworks 后\n- **统一特征存储**：利用 Hopsworks 的特征存储（Feature Store），团队只需定义一次特征逻辑，即可同时服务于离线训练和在线低延迟推理，彻底消除数据偏差。\n- **高效资产共享**：所有特征被注册为可搜索、可复用的资产，数据科学家能直接发现并组合现有特征，将新模型的开发周期从数周缩短至几天。\n- **全链路可追溯**：平台自动记录特征的版本、来源及转换逻辑，一旦模型异常，团队可立即回溯数据血缘，快速锁定根因。\n- **无缝 MLOps 集成**：通过内置的流水线管理，特征数据能自动同步至训练和推理服务，实现了从实验到生产的一键部署。\n\nHopsworks 通过构建统一的实时 AI 湖仓架构，让数据团队从繁琐的数据工程中解放出来，专注于提升模型的核心业务价值。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Flogicalclocks_hopsworks_0deae6b0.png","logicalclocks","Hopsworks","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Flogicalclocks_b7198232.png","Developers of Hopsworks",null,"info@hopsworks.ai","https:\u002F\u002Fwww.hopsworks.ai\u002F","https:\u002F\u002Fgithub.com\u002Flogicalclocks",[85,89,93,97,101],{"name":86,"color":87,"percentage":88},"Java","#b07219",80.6,{"name":90,"color":91,"percentage":92},"Ruby","#701516",18.2,{"name":94,"color":95,"percentage":96},"Jupyter Notebook","#DA5B0B",0.8,{"name":98,"color":99,"percentage":100},"Python","#3572A5",0.2,{"name":102,"color":103,"percentage":104},"Shell","#89e051",0.1,1290,156,"2026-04-02T02:29:51","AGPL-3.0",4,"Linux (CentOS\u002FRHEL 8.x, Ubuntu 22.04)","未说明 (文档提及支持在集群中使用 GPU 进行训练，但未列出具体安装时的显卡型号、显存或 CUDA 版本要求)","最低 32GB",{"notes":114,"python":115,"dependencies":116},"1. 本地部署（On-premise）至少需要一台服务器或虚拟机，配置要求：8 核 CPU、32GB 内存、100GB 可用硬盘空间。2. 需要一个具有 sudo 权限的 UNIX 用户账户。3. 支持在 AWS、Azure、GCP 云环境托管运行，也提供 Serverless 版本。4. 本地部署通常建议与 Hopsworks 工程团队协作以适配特定基础设施。","Python 3 (安装脚本使用 python3，具体小版本未说明)",[117],"未说明 (作为平台级工具，依赖通过内部容器或环境管理，README 未列出具体 Python 库依赖)",[13,54,51],[120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,67],"feature-store","aws","azure","data-science","feature-engineering","feature-management","gcp","governance","kserve","machine-learning","mlops","model-serving","pyspark","python","serverless","ml","2026-03-27T02:49:30.150509","2026-04-06T08:44:24.315510",[139,144,149,154,159,163],{"id":140,"question_zh":141,"answer_zh":142,"source_url":143},15837,"在 Windows Conda 环境中运行 Hopsworks 时遇到 'ImportError' 或 'ModuleNotFoundError'（特别是涉及 confluent_kafka 或 pyhive）怎么办？","这是 Windows Conda 环境中的一个已知问题。解决方法是在启动 Python 或 Jupyter 进程之前，设置环境变量 `CONDA_DLL_SEARCH_MODIFICATION_ENABLE=1`。您可以在命令行中执行 `set CONDA_DLL_SEARCH_MODIFICATION_ENABLE=1` (Windows CMD) 或在系统环境变量设置中添加该变量，然后重新启动您的开发环境。","https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks\u002Fissues\u002F1191",{"id":145,"question_zh":146,"answer_zh":147,"source_url":148},15838,"Hopsworks 的 Schema 推断是否支持 Pandas 的原生数据类型（如 Int8, Int64 等）？","在旧版本中，Schema 推断不支持 Pandas 原生类型（会抛出 'dtype not supported' 错误），用户需要暂时使用 NumPy 类型。该功能已在 Hopsworks 3.1 版本中得到扩展支持。如果您使用的是托管服务 (app.hopsworks.ai)，请等待平台升级到 3.1 版本后即可直接使用 Pandas 原生类型。","https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks\u002Fissues\u002F1290",{"id":150,"question_zh":151,"answer_zh":152,"source_url":153},15839,"在 Hopsworks 1.0 中创建项目时界面无响应，或者找不到 'Services' 标签页怎么办？","首先请注意，'Services' 标签页仅存在于旧版本中，在 Hopsworks 1.0 中请在 'Monitoring and Control' 部分查看服务状态。如果创建项目卡住或服务未启动，请以 root 用户身份登录服务器，依次执行以下命令重启所有本地服务：\n1. `\u002Fsrv\u002Fhops\u002Fkagent\u002Fkagent\u002Fbin\u002Fshutdown-all-local-services.sh`\n2. `\u002Fsrv\u002Fhops\u002Fkagent\u002Fkagent\u002Fbin\u002Fstart-all-local-services.sh`","https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks\u002Fissues\u002F372",{"id":155,"question_zh":156,"answer_zh":157,"source_url":158},15840,"在单机模式（Single Machine）或 VirtualBox 中安装 Hopsworks 后，无法通过 localhost 访问界面怎么办？","这通常是因为端口转发配置问题或内部服务（如 hopsfs）启动失败。请检查 VirtualBox 的网络设置，确保主机端口已正确映射到客户机端口（例如主机端口 20766 映射到客户机 8080）。同时，查看 `karamel-chef\u002Fnohup.out` 日志文件以确认是否有服务启动报错。如果日志显示 hopsfs 有问题，可能是主机资源不足或网络配置冲突导致。","https:\u002F\u002Fgithub.com\u002Flogicalclocks\u002Fhopsworks\u002Fissues\u002F451",{"id":160,"question_zh":161,"answer_zh":162,"source_url":153},15841,"如何在本地调试 Hopsworks 部署问题，特别是当项目创建失败且没有明显错误提示时？","当界面操作无反应时，建议直接登录后端服务器检查服务状态。对于基于镜像或虚拟机的部署，可以通过 SSH 进入实例，检查关键服务的日志。如果是服务未运行的问题，尝试手动重启核心组件（如使用 `start-all-local-services.sh` 脚本）。此外，确认管理员账户是否正确激活，因为未激活的用户可能导致项目创建请求挂起。",{"id":164,"question_zh":165,"answer_zh":166,"source_url":143},15842,"在使用 Poetry 或 Conda 安装 hsfs 时，遇到缺少 'python' extras 依赖的错误如何解决？","该错误表明安装时未包含完整的 Python 引擎依赖。虽然错误信息建议运行 `pip install hsfs[python]`，但在 Conda 或 Poetry 环境中，更有效的做法是检查环境兼容性。如果是 Windows Conda 用户，优先尝试设置 `CONDA_DLL_SEARCH_MODIFICATION_ENABLE=1` 环境变量。如果问题依旧，请确保显式安装了 `confluent-kafka` 和 `pyhive` 等底层依赖库，或者切换到标准的 pip 虚拟环境进行安装。",[168,173,178,183,188,193,198,203,208,213,218,223,228,233,238,243,248,253,258,263],{"id":169,"version":170,"summary_zh":171,"released_at":172},90557,"v3.7.0","## Hopsworks\n\n### 错误\n\n[HWORKS-224](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-224) Hopsworks Python SDK 中的 OpenSearch 链接已损坏\n\n[HWORKS-853](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-853) 在集成测试中模型导出失败\n\n[HWORKS-862](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-862) 非 DAS 节点上无法使用远程客户端\n\n[HWORKS-864](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-864) Istio 未使用注册表中的 Docker 镜像\n\n[HWORKS-865](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-865) 监控已删除作业的执行时出现空指针异常\n\n[HWORKS-869](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-869) 邮箱验证正则表达式不支持大写字母\n\n[HWORKS-872](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-872) 无法从集群定义中禁用 OAuth 组映射\n\n[HWORKS-878](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-878) Grafana 使用了错误的 Prometheus 名称\n\n[HWORKS-879](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-879) 如果在首次上传正在进行时再次上传同一文件，上传将会失败。\n\n[HWORKS-881](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-881) 在启动作业之前，应验证 SparkJob 和 Python 作业的应用程序文件是否存在\n\n[HWORKS-882](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-882) 禁止将模型导出到数据集的根路径\n\n[HWORKS-890](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-890) 恢复 RonDB 备份会失败，因为 restore-data 不允许在具有唯一索引的表中进行恢复\n\n[HWORKS-898](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-898) PyTorch 安装不支持 GPU\n\n[HWORKS-899](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-899) gh-pages 文档中的图片链接已损坏\n\n[HWORKS-901](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-901) Git 安装支持不应在 git+ 前面添加前缀\n\n[HWORKS-907](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-907) CloudManager 应将所有实例列入黑名单，以防止自动移除\n\n[HWORKS-908](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-908) ProjectQuotaController 在尝试获取未成功创建的项目时出现空指针异常\n\n[HWORKS-910](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-910) Kafka 的 get_default_config() 中缺少端点识别算法\n\n[HWORKS-950](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-950) 如果用户有多个姓氏，git pull 将会失败\n\n[HWORKS-954](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-954) cert-operator 的 api_key 被创建为空\n\n[HWORKS-961](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-961) 如果未启用全局复制，则不应开启 MySQL 二进制日志\n\n[HWORKS-974](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-974) ndb-chef 中用于提取 RonDB 的资源保护机制不起作用\n\n[HWORKS-975](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-975) hopsworks-api 中的 OpenSearchApi 不应依赖 ELASTIC_ENDPOINT\n\n[HWORKS-976](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORK","2024-03-02T16:18:26",{"id":174,"version":175,"summary_zh":176,"released_at":177},90558,"v3.5.0","### 错误\n\n[HWORKS-224](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-224) Hopsworks Python SDK 中的 OpenSearch 链接已损坏\n\n[HWORKS-267](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-267) 检出失败后，拉取时分支为空\n\n[HWORKS-269](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-269) 文件状态和 UI 表格中显示的内容不一致\n\n[HWORKS-309](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-309) 搜索功能存在错误\n\n[HWORKS-331](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-331) Payara 重启后，Hopsworks 的指标未显示\n\n[HWORKS-374](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-374) opensearch_api.get_default_py_config() 返回的是主机的公网 IP，应返回 Consul 的 FQDN\n\n[HWORKS-386](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-386) Spark 作业不接受参数中的空格\n\n[HWORKS-469](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-469) 在失败情况下，依赖的 Conda 命令未被正确处理\n\n[HWORKS-475](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-475) Hopsworks Python 库在 sklearn 旧版部署中无法连接\n\n[HWORKS-492](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-492) Serving 日志未在 Kibana 中显示\n\n[HWORKS-493](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-493) 部署创建表单中的请求批处理配置已损坏\n\n[HWORKS-496](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-496) 如果不是重复的部署条目，则不会抛出异常\n\n[HWORKS-498](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-498) 字典格式的 Transformer 资源未被正确反序列化\n\n[HWORKS-499](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-499) 支持将 Jupyter Notebook 用作预测器脚本\n\n[HWORKS-561](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-561) Hopsworks Kafka 授权器不允许 IDEMPOTENT_WRITE 操作\n\n[HWORKS-569](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-569) Filebeat 无法列出节点，而这是抓取容器日志所必需的\n\n[HWORKS-602](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-602) 高可用集群无法连接到实时日志、RM UI、Spark UI、监控和 onlineFS\n\n[HWORKS-630](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-630) 创建项目时，无效的项目名称不会导致创建失败\n\n[HWORKS-633](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-633) 当 Kubernetes 停止重新启动 Pod 时，失败的部署会显示为“正在启动”\n\n[HWORKS-636](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-636) 使用 .save() 更新部署时，推理日志记录的默认值会被覆盖\n\n[HWORKS-656](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-656) NP 尝试强制删除一个部分创建的项目\n\n[HWORKS-693](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-693) 当预测器类初始化失败时，服务器日志不会被收集\n\n[HWORKS-786](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-786) 对于禁用标志的情况，存储连接器规范验证失败\n\n[HWORKS-787](https:\u002F\u002Fhopsworks.atlassian.","2023-11-30T23:01:14",{"id":179,"version":180,"summary_zh":181,"released_at":182},90559,"v3.4.0","## Hopsworks 3.4.0\n\n### Bug\n\n[HWORKS-372](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-372) 社区版样式缺少显式证明信息\n\n[HWORKS-397](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-397) 模型服务 API 参考文档链接已失效\n\n[HWORKS-426](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-426) 托管 Kubernetes 的最新版本可能不支持第一方 JWT\n\n[HWORKS-538](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-538) 数据集共享 ACL 错误\n\n[HWORKS-610](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-610) JupyterJWTManager 上出现 HazelcastSerializationException 异常\n\n[HWORKS-632](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-632) hopsworks-api 中的 Command.from_response_json 应该基于 items 字段进行解析\n\n[HWORKS-638](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-638) Kafka 3 已移除 --zookeeper 选项\n\n[HWORKS-639](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-639) 在 onlinefs-kafka.properties 文件中模板化证书存储密码\n\n[HWORKS-640](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-640) 夜间工作流和负载测试不应以管理员用户身份运行\n\n[HWORKS-654](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-654) 额外约束迁移未遵循列的顺序\n\n[HWORKS-663](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-663) remote_file 不支持 not_if 条件\n\n[HWORKS-665](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-665) 将 Hopsworks 中的 Zookeeper 依赖升级至 3.8.2，以修复 CVE-2023-34462 漏洞\n\n[HWORKS-666](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-666) 在工作节点上导入证书时，配置 HA Glassfish 会失败\n\n[HWORKS-670](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-670) 按用户设置 Flink 作业的暂存目录和检查点目录\n\n[HWORKS-675](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-675) 在 Hopsworks 中排除 Zookeeper 库中的 logback 依赖\n\n[HWORKS-680](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-680) Karamel 在存在配方并行性限制的情况下，即使前一个任务失败，仍会调度任务\n\n[HWORKS-687](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-687) ZooKeeper 节点无法启动，原因是它们无法解析其他 ZK Consul 域名\n\n[HWORKS-688](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-688) Python 库安装错误信息未使用 latin1_general_cs 字符集\n\n[HWORKS-689](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-689) OAuth 登出 URI 被设置为重定向 URI\n\n[HWORKS-711](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-711) 删除用户时因外键约束而失败\n\n[HWORKS-731](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-731) 使用错误的预测器脚本路径部署模型时，会抛出不易理解的错误\n\n[HWORKS-735](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-735) Chef 属性用于配置 ldap_groups_search_filter\n\n[HWORKS-736](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-736) ndb-chef 下载备份时应跳过主机密钥验证\n\n[HWORKS-750](https:\u002F\u002Fhopsworks.atla","2023-10-07T21:07:57",{"id":184,"version":185,"summary_zh":186,"released_at":187},90560,"v3.3.0","## Hopsworks 3.3.0\n\n### Bug\n\n[HWORKS-224](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-224) Hopsworks Python SDK 中的 OpenSearch 链接已损坏\n\n[HWORKS-267](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-267) 检出失败后，拉取时分支为空\n\n[HWORKS-269](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-269) 文件状态与 UI 表格中显示的内容不一致\n\n[HWORKS-309](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-309) 搜索功能存在错误\n\n[HWORKS-319](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-319) 用户的活跃项目未更新\n\n[HWORKS-331](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-331) Payara 重启后，Hopsworks 的指标无法显示\n\n[HWORKS-374](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-374) opensearch_api.get_default_py_config() 返回的是主机的公网 IP，应返回 Consul 的 FQDN\n\n[HWORKS-379](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-379) LDAP 用户无法下载证书\n\n[HWORKS-386](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-386) Spark 作业不接受参数中的空格\n\n[HWORKS-397](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-397) 到 Model Serving API 参考文档的链接已损坏\n\n[HWORKS-459](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-459) 审计日志中的调用者信息应保持一致\n\n[HWORKS-469](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-469) 在 Conda 命令失败时，依赖项未被正确处理\n\n[HWORKS-475](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-475) Hopsworks Python 库在 sklearn 旧版部署中无法连接\n\n[HWORKS-492](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-492) Serving 服务的日志未在 Kibana 中显示\n\n[HWORKS-493](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-493) 部署创建表单中的请求批处理配置已损坏\n\n[HWORKS-496](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-496) 如果部署条目不是重复的，则不会抛出异常\n\n[HWORKS-498](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-498) 字典格式的 Transformer 资源未被正确反序列化\n\n[HWORKS-499](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-499) 支持将 Jupyter Notebook 用作预测器脚本\n\n[HWORKS-501](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-501) 在 Flink UI 中收集日志并分发自定义库\n\n[HWORKS-521](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-521) 解决 Ruby 测试失败问题\n\n[HWORKS-524](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-524) HSML 忽视 knative_domain 参数\n\n[HWORKS-525](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-525) K8s 的 Python 资源监控器会统计所有节点\n\n[HWORKS-528](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-528) 代理用户不应被阻止\n\n[HWORKS-533](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-533) 每晚的 Jenkins 流水线报告有 500 次失败，尽管我们只有 40 个测试用例\n\n[HWORKS-539](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHWORKS-539) 在 requests 2.29.0 发布后，存储初始化程序出现故障\n\n","2023-08-02T07:12:36",{"id":189,"version":190,"summary_zh":191,"released_at":192},90561,"v3.2.0","# 发布说明 - 特征存储 - 3.2.0\n\n### 壮举\n\n[FSTORE-615](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-615) HSFS JavaDoc\n\n### Bug\n\n[FSTORE-22](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-22) 无法将数据插入到流式特征组中\n\n[FSTORE-313](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-313) 在 Python 引擎中对不支持的外部特征组调用 save 时，抛出可理解的错误信息\n\n[FSTORE-456](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-456) 验证时间在未来\n\n[FSTORE-516](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-516) feature\\_view.json 中的“transformationFunction”字段为空\\(\\)\n\n[FSTORE-540](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-540) 修复共享文件系统集成测试\n\n[FSTORE-549](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-549) 向 onlinefs 上传大量行时，会因 TOPIC\\_AUTHORIZATION\\_FAILED 而失败\n\n[FSTORE-566](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-566) 特征存储在线指标页面返回“未找到仪表板”\n\n[FSTORE-602](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-602) get\\_or\\_create\\_feature\\_group 不会验证特征组名称\n\n[FSTORE-607](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-607) 调查 hsfs Java 资源泄漏  \n\n[FSTORE-636](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-636) ExpectationSuite SonarQube 智能更新事务\n\n[FSTORE-639](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-639) 锁定 SQLAlchemy 版本或升级到新的 2.0.0 版本\n\n[FSTORE-653](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-653) get\\_feature\\_vector 功能损坏 \n\n[FSTORE-660](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-660) 文档：创建外部特征组的语法问题 \n\n[FSTORE-661](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-661) 在 Jupyter Notebook 中创建外部特征组未能成功 \n\n[FSTORE-682](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-682) 修复 fg save\\_expectation\\_suite\n\n[FSTORE-689](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-689) 浅拷贝 DataFrame 隐藏了时区 bug\n\n[FSTORE-693](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-693) 使用验证套件插入数据时抛出日期格式异常\n\n[FSTORE-701](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-701) 修复负载测试中的数据验证问题\n\n[FSTORE-730](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-730) 修复在数据加载负载测试中启动多个回填作业的问题\n\n[FSTORE-760](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-760) 修复负载测试作业堆栈跟踪中日志打印的问题\n\n[FSTORE-763](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-763) 在追加操作后，onlinefs 处理 Avro 异常\n\n[FSTORE-776](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-776) 训练与测试分割比例中的浮点数问题\n\n[FSTORE-780](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-780) 文档未提及 event\\_time 的类型限制\n\n[FSTORE-781](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-781) 外部特征的文档","2023-05-04T20:22:13",{"id":194,"version":195,"summary_zh":196,"released_at":197},90562,"v3.1.0","# 发布说明 - 特征存储 - 3.1.0\n\n### 巨型任务\n\n[FSTORE-472](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-472) 元数据管理\n\n[FSTORE-474](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-474) 功能标志\n\n[FSTORE-475](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-475) 存储连接器\n\n### 任务\n\n[FSTORE-33](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-33) 改进 HSFS 参数描述\n\n[FSTORE-35](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-35) 权限矩阵\n\n[FSTORE-42](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-42) 文档中“你可以通过 UI 完成”这句话不够有帮助（针对外部功能标志）\n\n[FSTORE-43](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-43) 文档中的示例数据存在语法问题（缺少逗号）\n\n[FSTORE-52](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-52) 修复 GE 教程中的拼写错误\n\n[FSTORE-309](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-309) 在查询过滤器中添加对 LIKE 操作符的支持\n\n[FSTORE-311](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-311) 支持 datetime64[ns, TZ] 类型的列\n\n[FSTORE-315](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-315) 将 GE 版本升级到 0.14.12，以支持 Jinja2 3.0.0 版本\n\n[FSTORE-316](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-316) 添加时间戳\u002F时区指南\n\n[FSTORE-322](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-322) 在 Python 客户端中启用特征视图元数据的更新\n\n[FSTORE-325](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-325) 使 Redshift 连接器配置中的驱动程序字段可选\n\n[FSTORE-328](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-328) 如果用户设置的主键和分区键不存在，则抛出清晰的错误信息\n\n[FSTORE-330](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-330) 使 Python 客户端适应为单期望 API 引入的变更\n\n[FSTORE-332](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-332) 实现 get_or_create_feature_view 方法\n\n[FSTORE-334](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-334) 明确的溯源信息——特征组、特征视图、训练数据集\n\n[FSTORE-336](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-336) 改进 get_batch_data 中的错误信息\n\n[FSTORE-337](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-337) 单期望 API 和期望套件智能更新\n\n[FSTORE-339](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-339) 添加 Great Expectations 验证活动\n\n[FSTORE-340](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-340) 为特征名称添加 Great Expectations 关键字参数验证\n\n[FSTORE-341](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-341) 期望的历史记录验证\n\n[FSTORE-342](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-342) 将 Java 测试添加到 GitHub 流水线中\n\n[FSTORE-345](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-345) 更新文档，以反映 hsfs 引擎中支持的方法\n\n[FSTORE-346](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FFSTORE-346) 创建特征失败","2023-04-07T21:22:35",{"id":199,"version":200,"summary_zh":201,"released_at":202},90563,"v3.0.0","# 发行说明 - Hopsworks - 版本 3.0.0\n\n### Bug\n\n[HOPSWORKS-3255](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3255) Elasticsearch - 高亮片段错误\n\n[HOPSWORKS-3253](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3253) 功能组类型枚举在升级后与流式功能组不兼容\n\n[HOPSWORKS-3252](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3252) 禁用功能组摄取主题的 Kafka 主题删除功能\n\n[HOPSWORKS-3246](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3246) 更新部署后，新创建的 Kafka 主题名称未包含在响应中\n\n[HOPSWORKS-3236](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3236) 改进模型推理错误响应\n\n[HOPSWORKS-3235](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3235) 当用户未在查询参数中指定 as_of 时，使用 Hudi 快照查询\n\n[HOPSWORKS-3233](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3233) Spark\u002FHive\u002FHudi 之间的时间戳不兼容\n\n[HOPSWORKS-3232](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3232) 不应允许删除已被现有部署使用的模型\n\n[HOPSWORKS-3228](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3228) 将 scikit-learn 升级至 v1.1.1\n\n[HOPSWORKS-3227](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3227) 仅返回有效的推理端点\n\n[HOPSWORKS-3221](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3221) Spark 分发文件未遵守 HopsFS 的副本因子\n\n[HOPSWORKS-3220](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3220) HopsKafkaAuthorizer 在无法找到任何 ACL 时会重试并刷新 ACL 缓存\n\n[HOPSWORKS-3212](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3212) 移除存储初始化器中的 pickle 文件验证\n\n[HOPSWORKS-3209](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3209) expectation_suite 更新失败\n\n[HOPSWORKS-3207](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3207) 未导出的服务日志查看器 Chef 属性\n\n[HOPSWORKS-3204](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3204) 当其中一个功能组被删除时，功能视图无法获取查询\n\n[HOPSWORKS-3197](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3197) 功能存储配额强制执行在非缓存功能组上出现 NPE 错误\n\n[HOPSWORKS-3193](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3193) 从 UI 创建没有 Expectation Suite 的功能组失败\n\n[HOPSWORKS-3191](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3191) 如果 OAuth 映射为空，则不会移除该组\n\n[HOPSWORKS-3187](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3187) 使用 Istio 进行推理时，如果 transformers 中未启用 HSML，则无法正常工作\n\n[HOPSWORKS-3185](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3185) API 密钥作用域在删除时不会级联\n\n[HOPSWORKS-3183](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-3183) 在线特征向量未按用户输入排序\n\n[HOPSWORKS-3171](https:\u002F\u002Fhopsworks.atlassian.net\u002Fbrowse","2022-07-23T23:11:55",{"id":204,"version":205,"summary_zh":206,"released_at":207},90564,"v2.5.0","\u003Cdiv class=\"csg-wrapper\" style=\"font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', 'Roboto', 'Oxygen', 'Ubuntu', 'Fira Sans', 'Droid Sans', 'Helvetica Neue', sans-serif; font-size: 14px; font-weight: 400; line-height: 24px; vertical-align: baseline;\">\u003Ch1 class=\"csg-h1\" style=\"font-style: inherit; color: #172B4D; font-weight: 600; margin-bottom: 0; font-size: 23px; line-height: 1.1034; margin-top: 40px; letter-spacing: -0.01em;\">发布说明 - Hopsworks - 版本 2.5.0\u003C\u002Fh1>\u003Ch3 class=\"csg-h3\" style=\"font-style: inherit; color: #172B4D; font-weight: 600; margin-bottom: 0; font-size: 16px; line-height: 1.2; margin-top: 32px; font-weight: 500; letter-spacing: -0.008em;\">错误\u003C\u002Fh3>\u003Cp class=\"csg-p\" style=\"margin: 0; padding: 0px; margin-bottom: 7px; padding-top: 7px; mso-line-height-rule: exactly; line-height: 24px; font-size: 14px;\">\u003Ca href=\"https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2983\" class=\"csg-mark-link\" style=\"border: none; background: transparent; color: #0052cc; text-decoration: none;\">HOPSWORKS-2983\u003C\u002Fa> 在 Kubernetes 中，如果没有 hops-system 命名空间，就无法创建项目\u003C\u002Fp>\u003Cp class=\"csg-p\" style=\"margin: 0; padding: 0px; margin-bottom: 7px; padding-top: 7px; mso-line-height-rule: exactly; line-height: 24px; font-size: 14px;\">\u003Ca href=\"https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2980\" class=\"csg-mark-link\" style=\"border: none; background: transparent; color: #0052cc; text-decoration: none;\">HOPSWORKS-2980\u003C\u002Fa> 允许模型制品中包含多个模型文件\u003C\u002Fp>\u003Cp class=\"csg-p\" style=\"margin: 0; padding: 0px; margin-bottom: 7px; padding-top: 7px; mso-line-height-rule: exactly; line-height: 24px; font-size: 14px;\">\u003Ca href=\"https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2973\" class=\"csg-mark-link\" style=\"border: none; background: transparent; color: #0052cc; text-decoration: none;\">HOPSWORKS-2973\u003C\u002Fa> zip\u002Funzip 的 URI 编码调用了系统 Python，应改用 hops-system Python\u003C\u002Fp>\u003Cp class=\"csg-p\" style=\"margin: 0; padding: 0px; margin-bottom: 7px; padding-top: 7px; mso-line-height-rule: exactly; line-height: 24px; font-size: 14px;\">\u003Ca href=\"https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2965\" class=\"csg-mark-link\" style=\"border: none; background: transparent; color: #0052cc; text-decoration: none;\">HOPSWORKS-2965\u003C\u002Fa> 从 hops-util-py 的依赖项中移除 pathlib、confluent-kafka、hops-petastorm 和 opencv-python\u003C\u002Fp>\u003Cp class=\"csg-p\" style=\"margin: 0; padding: 0px; margin-bottom: 7px; padding-top: 7px; mso-line-height-rule: exactly; line-height: 24px; font-size: 14px;\">\u003Ca href=\"https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2963\" class=\"csg-mark-link\" style=\"border: none; background: transparent; color: #0052cc; text-decoration: none;\">HOPSWORKS-2963\u003C\u002Fa> 列出模型标签时出现空指针异常\u003C\u002Fp>\u003Cp class=\"csg-p\" style=\"margin: 0; padding: 0px; margin-bottom: 7px; padding-top: 7px; mso-line-height-rule: exactly; line-height: 24px; font-size: 14px;\">\u003Ca href=\"https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse","2022-02-13T16:56:23",{"id":209,"version":210,"summary_zh":211,"released_at":212},90565,"v2.4.0","\u003Cdiv class=\"csg-wrapper\" style=\"font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', 'Roboto', 'Oxygen', 'Ubuntu', 'Fira Sans', 'Droid Sans', 'Helvetica Neue', sans-serif; font-size: 14px; font-weight: 400; line-height: 24px; vertical-align: baseline;\">\u003Ch1 class=\"csg-h1\" style=\"font-style: inherit; color: #172B4D; font-weight: 600; margin-bottom: 0; font-size: 23px; line-height: 1.1034; margin-top: 40px; letter-spacing: -0.01em;\">版本说明 - Hopsworks - 2.4.0 版\u003C\u002Fh1>\u003Ch3 class=\"csg-h3\" style=\"font-style: inherit; color: #172B4D; font-weight: 600; margin-bottom: 0; font-size: 16px; line-height: 1.2; margin-top: 32px; font-weight: 500; letter-spacing: -0.008em;\">缺陷\u003C\u002Fh3>\u003Cp class=\"csg-p\" style=\"margin: 0; padding: 0px; margin-bottom: 7px; padding-top: 7px; mso-line-height-rule: exactly; line-height: 24px; font-size: 14px;\">\u003Ca href=\"https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2770\" class=\"csg-mark-link\" style=\"border: none; background: transparent; color: #0052cc; text-decoration: none;\">HOPSWORKS-2770\u003C\u002Fa> 功能组清空内容端点重新设计\u003C\u002Fp>\u003Cp class=\"csg-p\" style=\"margin: 0; padding: 0px; margin-bottom: 7px; padding-top: 7px; mso-line-height-rule: exactly; line-height: 24px; font-size: 14px;\">\u003Ca href=\"https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2748\" class=\"csg-mark-link\" style=\"border: none; background: transparent; color: #0052cc; text-decoration: none;\">HOPSWORKS-2748\u003C\u002Fa> Kibana 和 Grafana 的 Consul 健康检查结果不正确\u003C\u002Fp>\u003Cp class=\"csg-p\" style=\"margin: 0; padding: 0px; margin-bottom: 7px; padding-top: 7px; mso-line-height-rule: exactly; line-height: 24px; font-size: 14px;\">\u003Ca href=\"https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2738\" class=\"csg-mark-link\" style=\"border: none; background: transparent; color: #0052cc; text-decoration: none;\">HOPSWORKS-2738\u003C\u002Fa> 工件文件的所有者不应为 GlassFish\u003C\u002Fp>\u003Cp class=\"csg-p\" style=\"margin: 0; padding: 0px; margin-bottom: 7px; padding-top: 7px; mso-line-height-rule: exactly; line-height: 24px; font-size: 14px;\">\u003Ca href=\"https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2730\" class=\"csg-mark-link\" style=\"border: none; background: transparent; color: #0052cc; text-decoration: none;\">HOPSWORKS-2730\u003C\u002Fa> 代理 Servlet 存在资源泄漏问题\u003C\u002Fp>\u003Cp class=\"csg-p\" style=\"margin: 0; padding: 0px; margin-bottom: 7px; padding-top: 7px; mso-line-height-rule: exactly; line-height: 24px; font-size: 14px;\">\u003Ca href=\"https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2717\" class=\"csg-mark-link\" style=\"border: none; background: transparent; color: #0052cc; text-decoration: none;\">HOPSWORKS-2717\u003C\u002Fa> Avro 在模式名称以数字开头时会失败\u003C\u002Fp>\u003Cp class=\"csg-p\" style=\"margin: 0; padding: 0px; margin-bottom: 7px; padding-top: 7px; mso-line-height-rule: exactly; line-height: 24px; font-size: 14px;\">\u003Ca href=\"https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2710\" class=\"csg-mark-link\" style=\"border: none; background: transparent; color: ","2021-10-26T13:59:08",{"id":214,"version":215,"summary_zh":216,"released_at":217},90566,"v2.3.0","# 发行说明 - Hopsworks - 版本 2.3.0\n\n### Bug\n\n[HOPSWORKS-2663](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2663) 修复在 Ubuntu 18.04 上的 Docker 安装问题\n\n[HOPSWORKS-2639](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2639) 在特征存储集成中，当没有 Databricks 集群时出现 NPE 异常\n\n[HOPSWORKS-2634](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2634) 训练数据集准备语句中未遵守特征顺序\n\n[HOPSWORKS-2631](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2631) SYNC\\_BASE\\_ENV 操作在失败时也应显示错误日志\n\n[HOPSWORKS-2628](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2628) 包含追加特征的特征组的 OnlineFS 模式缺少旧特征\n\n[HOPSWORKS-2626](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2626) MySQL 时区问题\n\n[HOPSWORKS-2625](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2625) 为 dataset 的 projectsSharedWith 添加状态\n\n[HOPSWORKS-2619](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2619) 为 Spark 3 的 Livy 添加缺失的 JVM 导入\n\n[HOPSWORKS-2618](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2618) 强制删除项目时出现 NPE 异常\n\n[HOPSWORKS-2614](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2614) 使用默认连接器从 Python 创建的训练数据集失败\n\n[HOPSWORKS-2609](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2609) 取消共享待处理的特征存储失败\n\n[HOPSWORKS-2594](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2594) 当使用 Git 后端运行带有 Spark 或 Python 内核的笔记本时，附加 Jupyter 配置失败\n\n[HOPSWORKS-2583](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2583) mnist 实验笔记本应在气隙环境中正常工作\n\n[HOPSWORKS-2582](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2582) Spark 驱动程序核心数被错误地设置为配置的执行器核心数\n\n[HOPSWORKS-2577](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2577) 带有额外写入参数的 Python 特征组摄取失败\n\n[HOPSWORKS-2574](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2574) 特征验证规则 Level.Warning 默认为 Error\n\n[HOPSWORKS-2569](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2569) 允许使用大写字母作为关键字\n\n[HOPSWORKS-2567](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2567) 在检索按需特征组的验证时出现空指针异常\n\n[HOPSWORKS-2556](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2556) Jupyter 应为实验和 Spark 使用默认的 DockerJobConfiguration\n\n[HOPSWORKS-2555](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2555) Python 内核中未设置 KAFKA\\_BROKERS\n\n[HOPSWORKS-2554](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2554) oauth2 获取用户时的 bug\n\n[HOPSWORKS-2553](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2553) 修复 kfserving 气隙部署问题\n\n[HOPSWO","2021-08-24T13:47:20",{"id":219,"version":220,"summary_zh":221,"released_at":222},90567,"v2.2.0","# Release notes - Hopsworks - Version 2.2.0\r\n\r\n### Bug\r\n\r\n[HOPSWORKS-2482](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2482) Read Hudi Feature Group from Hive Engine fails with unknown fg0\r\n\r\n[HOPSWORKS-2481](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2481) Only default deployments must be enabled when KFServing is not installed\r\n\r\n[HOPSWORKS-2473](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2473) dataset read acl list does not get cleared on dataset delete\r\n\r\n[HOPSWORKS-2471](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2471) fix prepared statement parameter for string PKs\r\n\r\n[HOPSWORKS-2447](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2447) Pip conflicts modal not closing when clicking confirm\r\n\r\n[HOPSWORKS-2415](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2415) reset elastic client in case it uses the wrong certificate\r\n\r\n[HOPSWORKS-2414](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2414) Verify that  API token used when starting Jupyter with AutoPushOnShutdown enabled has write access to git repo\r\n\r\n[HOPSWORKS-2352](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2352) Snowflake connector requires pytz \u003C= 2021.0\r\n\r\n[HOPSWORKS-2303](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2303) Change python pip search to use pypi simple index and JSON API\r\n\r\n[HOPSWORKS-2293](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2293) Fix test for kvm\r\n\r\n### Improvement\r\n\r\n[HOPSWORKS-2484](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2484) Fix SSL\u002FTLS settings for Glassfish network listeners\r\n\r\n[HOPSWORKS-2460](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2460) Create admin end-point for user management\r\n\r\n[HOPSWORKS-2433](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2433) Azure Root CA missing from glassfish trust store.\r\n\r\n[HOPSWORKS-2432](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2432) TLS for Resource Manager UI\r\n\r\n[HOPSWORKS-2423](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2423) Advertise Rondb managment node through consul\r\n\r\n[HOPSWORKS-2422](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2422) Advertise Kafka brokers through Consul\r\n\r\n[HOPSWORKS-2421](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2421) \\[ndb-chef\\] Fix config.ini for RonDB\r\n\r\n[HOPSWORKS-2408](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2408) Feature store activity should return validation results\r\n\r\n[HOPSWORKS-2405](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2405) Upgrade Python version we install in recipes\r\n\r\n[HOPSWORKS-2404](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2404) remove metadata designer code\r\n\r\n[HOPSWORKS-2396](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2396) Add AzureML Studio notebook\r\n\r\n[HOPSWORKS-2368](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2368) Do not require API token for working with public GitHub and Gitlab repos using Jupyter\r\n\r\n[HOPSWORKS-2367](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2367) Hopsworks Connector Docs\r\n\r\n[HOPSWORKS-2366](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2366) Replace service with systemctl in recipes\r\n\r\n[HOPSWORKS-2351](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2351) derived fg provenance tests - links\u002Fusage\r\n\r\n[HOPSWORKS-2345](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2345) Get Kafka Broker Endpoints through REST Api\r\n\r\n[HOPSWORKS-2336](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2336) It should not be mandatory to provide a version when installing a pip or conda library\r\n\r\n[HOPSWORKS-2314](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2314) Snowflake connector\r\n\r\n[HOPSWORKS-2247](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2247) Replace feature store job tracking with provenance\r\n\r\n[HOPSWORKS-2139](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2139) Use Hudi commit ID as statistics timestamp\r\n\r\n[HOPSWORKS-2071](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2071) Data validation v2\r\n\r\n[HOPSWORKS-2052](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2052) Save preferences for dataset and experiments filter\r\n\r\n### New Feature\r\n\r\n[HOPSWORKS-2474](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2474) Docker job\r\n\r\n[HOPSWORKS-2458](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2458) Online-FS Service Integration\r\n\r\n[HOPSWORKS-2399](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2399) Add new frontend deployment\r\n\r\n[HOPSWORKS-2384](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2384) Add support for coalesce in training dataset\r\n\r\n[HOPSWORKS-2195](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2195) Attach Jupyter configuration to notebook as Xattr\r\n\r\n[HOPSWORKS-1978](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1978) schematized tags\r\n\r\n### Sub-task\r\n\r\n[HOPSWORKS-2193](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2193) Add support for request batching\r\n\r\n[HOPSWORKS-2185](https:\u002F\u002Flogica","2021-04-09T17:05:09",{"id":224,"version":225,"summary_zh":226,"released_at":227},90568,"v2.1.0","# Release notes - Hopsworks - Version 2.1.0\r\n\r\n### Bug\r\n\r\n[HOPSWORKS-2359](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2359) Always show GPU configuration fields for Jobs and Jupyter\r\n[HOPSWORKS-2358](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2358) Remove ROCM GPU monitoring code\r\n[HOPSWORKS-2346](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2346) Identifying if a project is using the base image on ECR does not work\r\n[HOPSWORKS-2338](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2338) handle empty tfrecord files in TFDataEngine\r\n[HOPSWORKS-2333](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2333) create\\_training\\_dataset method throws Connector not found exception \r\n[HOPSWORKS-2311](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2311) Template required ec2-init parameters for the unmanaged version\r\n[HOPSWORKS-2307](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2307) Docker garbage collection should not remove base or python37 images for any hopsworks version\r\n[HOPSWORKS-2302](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2302) Hide python pip search bar as underlying pip search is now deprecated\r\n[HOPSWORKS-2294](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2294) Installing a python library from git should enclose the url in quotes\r\n[HOPSWORKS-2283](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2283) Reading from the online feature store doesn't work on Databricks\r\n[HOPSWORKS-2249](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2249) provenance elasticsearch - count and aggregations should be separate methods\r\n[HOPSWORKS-2248](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2248) Python file path is not shown in Job details\r\n[HOPSWORKS-2238](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2238) Feature Group schema primary key info not set for partition keys\r\n[HOPSWORKS-2219](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2219) Remove NaN from statistics\r\n[HOPSWORKS-2202](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2202) Docker registry caches deleted manifests\r\n[HOPSWORKS-2199](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2199) Glassfish consul domain should be included in Kafka super.users\r\n[HOPSWORKS-2198](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2198) Cannot get the schema of a shared kafka topic\r\n[HOPSWORKS-2187](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2187) Add anaconda user to hadoop group only if the group exists. Create hops-system Anaconda env conditionally\r\n[HOPSWORKS-2182](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2182) Disable unattended-upgrades in Ubuntu\r\n[HOPSWORKS-2174](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2174) issues attaching xattr to shared files\r\n[HOPSWORKS-2164](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2164) Add API Key annotation  on the commits endpoint\r\n[HOPSWORKS-2159](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2159) ruby tests fix\r\n[HOPSWORKS-2158](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2158) Query constructor ignores more than 2 joining conditions\r\n[HOPSWORKS-2151](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2151) Tag validation doesn't work for key only tags\r\n[HOPSWORKS-2145](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2145) Flask \\(sklearn\\) race condition when reading pid file\r\n[HOPSWORKS-2102](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2102) Conversion of Jupyter notebooks to Python programs should consider notebook kernel\r\n[HOPSWORKS-1984](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1984) Exception when creating online feature group without primary key\r\n\r\n### Task\r\n\r\n[HOPSWORKS-2348](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2348) Disable storage and compute quota by default\r\n[HOPSWORKS-2337](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2337) Upgrade hudi to 0.7.0 version\r\n[HOPSWORKS-2327](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2327) Ui fixed for generating training datasets\r\n[HOPSWORKS-2324](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2324) Display alert if project is using older docker image\r\n[HOPSWORKS-2320](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2320) Set PARQUET as default for non-hudi offline feature groups\r\n[HOPSWORKS-2287](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2287) Bump up [docker.io](http:\u002F\u002Fdocker.io) version for ubuntu\r\n[HOPSWORKS-2275](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2275) Remove InodeEncodingStatus Bean\r\n[HOPSWORKS-2274](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2274) Feature store \"NOT FOUND\" errors should return 404\r\n[HOPSWORKS-2241](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2241) Consul security recipe should not try to get Hopsworks IP if not needed\r\n[HOPSWORKS-2206](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2206) HSFS profile to install with and without Hive dependencies\r\n[HOPSWORKS-2196](https:\u002F","2021-02-12T09:23:58",{"id":229,"version":230,"summary_zh":231,"released_at":232},90569,"v2.0.0","# Release notes - Hopsworks - Version 2.0.0\r\n\r\n### Task\r\n\r\n[HOPSWORKS-2287](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2287) Bump up [docker.io](http:\u002F\u002Fdocker.io) version for ubuntu\r\n[HOPSWORKS-2186](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2186) Remove installation of slack operator in Airflow cookbook\r\n[HOPSWORKS-2161](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2161) Option to disable just the Registration UI\r\n[HOPSWORKS-2156](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2156) Pin all TensorFlow projects in Docker image\r\n[HOPSWORKS-2126](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2126) Cleanup old hops-util\\(-py\\) feature store notebooks\r\n[HOPSWORKS-2125](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2125) Bump hops dependencies to 3.2.0.1-RC0 \r\n[HOPSWORKS-2109](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2109) Remove base\\_env from PythonDep\r\n[HOPSWORKS-2094](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2094) Setup GPU monitoring without enabling it even when it's not a GPU deployment\r\n[HOPSWORKS-2090](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2090) Bump up [docker.io](http:\u002F\u002Fdocker.io) version for Ubuntu\r\n[HOPSWORKS-2072](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2072) \\[hops-hadoop-chef\\] Recipe should check if Hops distro has been build with correct profiles\r\n[HOPSWORKS-2043](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2043) Upgrade base and airflow conda environment to python 3.7\r\n[HOPSWORKS-2042](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2042) Unmanaged Hopsworks Image\r\n[HOPSWORKS-2041](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2041) Bump Hops Hadoop version to 3.2.0.1-SNAPSHOT\r\n[HOPSWORKS-2040](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2040) Docker registry does not need to be versioned\r\n[HOPSWORKS-2039](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2039) Bump hops dependencies to 3.2.0.0-RC3\r\n[HOPSWORKS-2037](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2037) Rename map\\_fun and wrapper\\_fun to train\\_fn\r\n[HOPSWORKS-1945](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1945) Store AWS access\u002Fsecret key in the secret store.\r\n[HOPSWORKS-1944](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1944) Split feature store tests in several categories\r\n[HOPSWORKS-1387](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1387) Upgrade to NDB 8\r\n\r\n### Bug\r\n\r\n[HOPSWORKS-2202](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2202) Docker registry caches deleted manifests\r\n[HOPSWORKS-2182](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2182) Disable unattended-upgrades in Ubuntu\r\n[HOPSWORKS-2162](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2162) Airflow page does not open\r\n[HOPSWORKS-2151](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2151) Tag validation doesn't work for key only tags\r\n[HOPSWORKS-2150](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2150) S3 storage connector should not bother about secrets if iam role is configured\r\n[HOPSWORKS-2149](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2149) Disable generating and sharing ssh key freely\r\n[HOPSWORKS-2147](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2147) Add feature group api scope to statistics resource endpoint\r\n[HOPSWORKS-2146](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2146) Resolve namenode consul name before setting feature group location\r\n[HOPSWORKS-2145](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2145) Flask \\(sklearn\\) race condition when reading pid file\r\n[HOPSWORKS-2128](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2128) Create api key not adding the right scope \r\n[HOPSWORKS-2122](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2122) Backend not handling null TimeTravelFormat sent by client\r\n[HOPSWORKS-2116](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2116) Add --network option to docker build\r\n[HOPSWORKS-2095](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2095) Uninstalling conda libraries doesn't work in airgapped environments\r\n[HOPSWORKS-2092](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2092) Overwriting a feature group by dropping\u002Frecreating loses default value of appended features\r\n[HOPSWORKS-2081](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2081) Fix search for registration disabled\r\n[HOPSWORKS-2080](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2080) Hops services should wait for Hopsworks before they start\r\n[HOPSWORKS-2068](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2068) Singleton initialization failure\r\n[HOPSWORKS-2067](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2067) shared featuregroup creation\r\n[HOPSWORKS-2065](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2065) Restart systemd units after failure \r\n[HOPSWORKS-2064](https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-2064) UI fixes for data preview\r\n[HOPSWORKS-2062](https:","2021-02-01T09:23:23",{"id":234,"version":235,"summary_zh":236,"released_at":237},90570,"v1.4.1","\r\n        Release notes - Hopsworks - Version 1.4.0\r\n    \r\n\u003Ch2>        Improvement\r\n\u003C\u002Fh2>\r\n\u003Cul>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1314'>HOPSWORKS-1314\u003C\u002Fa>] -         It should be possible to start Jupyter with a Shared Dataset as base directory\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1655'>HOPSWORKS-1655\u003C\u002Fa>] -         Refactor FeatureGroupService for the new clients\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1657'>HOPSWORKS-1657\u003C\u002Fa>] -         Generate per system user X.509 certificates\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1752'>HOPSWORKS-1752\u003C\u002Fa>] -         Changning Deployment Repos for Enterprise Edition\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1762'>HOPSWORKS-1762\u003C\u002Fa>] -         Add support for the Spark-Tfrecord library\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1786'>HOPSWORKS-1786\u003C\u002Fa>] -         xattr refactor - tags and featurestore search\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1791'>HOPSWORKS-1791\u003C\u002Fa>] -         [ePipe] Add support for large XAttrs\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1804'>HOPSWORKS-1804\u003C\u002Fa>] -         Integrate Prometheus with Consul\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1815'>HOPSWORKS-1815\u003C\u002Fa>] -         manage kubernetes through kagent\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1816'>HOPSWORKS-1816\u003C\u002Fa>] -         Use consul for hive-site and metrics.conf configuration \r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1825'>HOPSWORKS-1825\u003C\u002Fa>] -         Connect to MySQLd's using 127.0.0.1 instead of private IP\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1829'>HOPSWORKS-1829\u003C\u002Fa>] -         Register Kibana with consul\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1834'>HOPSWORKS-1834\u003C\u002Fa>] -         Add XAttrs configurations to chef\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1852'>HOPSWORKS-1852\u003C\u002Fa>] -         Jupyter server admin page returns 500\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1854'>HOPSWORKS-1854\u003C\u002Fa>] -         Add datanode replacement upon failure  parameters \r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1883'>HOPSWORKS-1883\u003C\u002Fa>] -         Add JOB audience to feature group delete endpoint\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1892'>HOPSWORKS-1892\u003C\u002Fa>] -         TrainingDataset\u002FFeatureGroup create automatically increments version if no version provided by user\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1897'>HOPSWORKS-1897\u003C\u002Fa>] -         Add example for submitting Spark\u002FPySpark jobs remotely\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1899'>HOPSWORKS-1899\u003C\u002Fa>] -         Add user and project create functions in hops Python library\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1921'>HOPSWORKS-1921\u003C\u002Fa>] -         Show certificate password in the UI instead of sending it by email.\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1924'>HOPSWORKS-1924\u003C\u002Fa>] -         Add Kafka scope to API keys\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1925'>HOPSWORKS-1925\u003C\u002Fa>] -         Upgrade conda to version 2020.07\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1935'>HOPSWORKS-1935\u003C\u002Fa>] -         Show project ID in the project settings\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1948'>HOPSWORKS-1948\u003C\u002Fa>] -         Distribute HSFS in the base environment\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1954'>HOPSWORKS-1954\u003C\u002Fa>] -         Use hopsworks version as a tag for all docker images\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1972'>HOPSWORKS-1972\u003C\u002Fa>] -         Add new api_key_value to featurestore connect method\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1974'>HOPSWORKS-1974\u003C\u002Fa>] -         [DOC] large xattributes - usage and upgrade details\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1975'>HOPSWORKS-1975\u003C\u002Fa>] -         Add support for passing job arguments in the Airflow workflow builder\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1976'>HOPSWORKS-1976\u003C\u002Fa>] -         New cloud installer script\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1979'>HOPSWORKS-1979\u003C\u002Fa>] -         Pydoop in hops-util-py should be able to handle hopsfs scheme\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1996'>HOPSWORKS-1996\u003C\u002Fa>] -   ","2020-10-09T14:33:29",{"id":239,"version":240,"summary_zh":241,"released_at":242},90571,"v1.3.0","\r\n        Release notes - Hopsworks - Version 1.3.0\r\n    \r\n\u003Ch2>        Bug\r\n\u003C\u002Fh2>\r\n\u003Cul>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1382'>HOPSWORKS-1382\u003C\u002Fa>] -         Placeholders for auditing REST API requests\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1471'>HOPSWORKS-1471\u003C\u002Fa>] -         Increase MaxNoOfTriggers and MaxNoOfFiredTriggers to delete large no of FKs\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1545'>HOPSWORKS-1545\u003C\u002Fa>] -         Handle cases where Elasticsearch is down\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1546'>HOPSWORKS-1546\u003C\u002Fa>] -         ruby tests\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1556'>HOPSWORKS-1556\u003C\u002Fa>] -         Kafka controller fixes\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1557'>HOPSWORKS-1557\u003C\u002Fa>] -         Dataset menu hidden as browser is not scrollable\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1558'>HOPSWORKS-1558\u003C\u002Fa>] -         dataset - share with cluster - fails\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1559'>HOPSWORKS-1559\u003C\u002Fa>] -         Truncate conda commands before upgrading to 1.1.0\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1560'>HOPSWORKS-1560\u003C\u002Fa>] -         Fix migration of datasets (HOPSWORKS-295)\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1561'>HOPSWORKS-1561\u003C\u002Fa>] -         Restrict access to elasticsearch data directory to elastic user\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1563'>HOPSWORKS-1563\u003C\u002Fa>] -         Schema of a shared topic is shown in one line\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1565'>HOPSWORKS-1565\u003C\u002Fa>] -         Removing a shared dataset removes it also from the source project\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1566'>HOPSWORKS-1566\u003C\u002Fa>] -         provenance elastic client - index not found exception not handled\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1568'>HOPSWORKS-1568\u003C\u002Fa>] -         Installing Node.js must work for airgapped deployments\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1569'>HOPSWORKS-1569\u003C\u002Fa>] -         Can't deploy with disable password login flag set to true\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1578'>HOPSWORKS-1578\u003C\u002Fa>] -         epipe operation fail on missing index\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1579'>HOPSWORKS-1579\u003C\u002Fa>] -         Align services on the project service nav bar\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1581'>HOPSWORKS-1581\u003C\u002Fa>] -         [hops-hadoop-log] Touch hadoop.log with correct ownership and permissions\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1583'>HOPSWORKS-1583\u003C\u002Fa>] -         Incorrect path for PYSPARK_ARCHIVES_PATH in spark-env.sh\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1589'>HOPSWORKS-1589\u003C\u002Fa>] -         Cannot upload to a shared dataset\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1590'>HOPSWORKS-1590\u003C\u002Fa>] -         Kafka advertised listeners external port conflict among brokers\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1596'>HOPSWORKS-1596\u003C\u002Fa>] -         Incorrect sklearn logs format\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1599'>HOPSWORKS-1599\u003C\u002Fa>] -         Cannot start\u002Fstop services from admin UI due to missing CommunicationController property\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1600'>HOPSWORKS-1600\u003C\u002Fa>] -         Services Status enum missing INIT state of kagent\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1601'>HOPSWORKS-1601\u003C\u002Fa>] -         Append projectId to topic name in kafka tour to avoid clashes with other topics\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1603'>HOPSWORKS-1603\u003C\u002Fa>] -         Kagent JWTHelper should retry in case of hopsworks authentication failure\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1604'>HOPSWORKS-1604\u003C\u002Fa>] -         In hops python library, the s3 connector tries to read parquet files as type avro\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1609'>HOPSWORKS-1609\u003C\u002Fa>] -         [ePipe] Fix version conflicts in Elasticsearch\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1612'>HOPSWORKS-1612\u003C\u002Fa>] -         Can't view directories with name url encoded\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1616'>HOPSWORKS-1616\u003C\u002Fa>] -         Hopsworks ear download URL is incorrect whe","2020-06-03T17:52:57",{"id":244,"version":245,"summary_zh":246,"released_at":247},90572,"v1.2.0","\r\n        Release notes - Hopsworks - Version 1.2.0\r\n    \r\n\u003Ch2>        Bug\r\n\u003C\u002Fh2>\r\n\u003Cul>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1283'>HOPSWORKS-1283\u003C\u002Fa>] -         Zip does not work when there is a space in the file path\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1379'>HOPSWORKS-1379\u003C\u002Fa>] -         Timers should be rescheduled when failed and handle exceptions\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1382'>HOPSWORKS-1382\u003C\u002Fa>] -         Placeholders for auditing REST API requests\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1507'>HOPSWORKS-1507\u003C\u002Fa>] -         [ndb-chef] List only TCP and UDP sockets when starting MGM server\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1509'>HOPSWORKS-1509\u003C\u002Fa>] -         Allow all OU in nodes DN in elasticsearch configuration\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1511'>HOPSWORKS-1511\u003C\u002Fa>] -         pci-tools package not  available and torchvision broken for 0.4.1\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1515'>HOPSWORKS-1515\u003C\u002Fa>] -         airflow webserver , scheduler and sqoop should not be enabled when service.enabled is false\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1534'>HOPSWORKS-1534\u003C\u002Fa>] -         Kibana error when recreating the project enviroment\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1541'>HOPSWORKS-1541\u003C\u002Fa>] -         [ePipe] Skip non searchable fields for schemabased exmetadata\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1543'>HOPSWORKS-1543\u003C\u002Fa>] -         [ePipe] Retry in case of conflict when adding datasets and projects\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1545'>HOPSWORKS-1545\u003C\u002Fa>] -         Handle cases where Elasticsearch is down\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1550'>HOPSWORKS-1550\u003C\u002Fa>] -         Do not upgrade pip binary in tensorflow default recipe\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1553'>HOPSWORKS-1553\u003C\u002Fa>] -         provenance cleaner timer\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1556'>HOPSWORKS-1556\u003C\u002Fa>] -         Kafka controller fixes\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1559'>HOPSWORKS-1559\u003C\u002Fa>] -         Truncate conda commands before upgrading to 1.1.0\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1560'>HOPSWORKS-1560\u003C\u002Fa>] -         Fix migration of datasets (HOPSWORKS-295)\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1561'>HOPSWORKS-1561\u003C\u002Fa>] -         Restrict access to elasticsearch data directory to elastic user\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1563'>HOPSWORKS-1563\u003C\u002Fa>] -         Schema of a shared topic is shown in one line\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1564'>HOPSWORKS-1564\u003C\u002Fa>] -         Job args expat tries to set null instead of empty string\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1565'>HOPSWORKS-1565\u003C\u002Fa>] -         Removing a shared dataset removes it also from the source project\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1578'>HOPSWORKS-1578\u003C\u002Fa>] -         epipe operation fail on missing index\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1589'>HOPSWORKS-1589\u003C\u002Fa>] -         Cannot upload to a shared dataset\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1590'>HOPSWORKS-1590\u003C\u002Fa>] -         Kafka advertised listeners external port conflict among brokers\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1594'>HOPSWORKS-1594\u003C\u002Fa>] -         HOPSWORKS-1554 breaks conda search in Hopsworks\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1599'>HOPSWORKS-1599\u003C\u002Fa>] -         Cannot start\u002Fstop services from admin UI due to missing CommunicationController property\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1600'>HOPSWORKS-1600\u003C\u002Fa>] -         Services Status enum missing INIT state of kagent\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1601'>HOPSWORKS-1601\u003C\u002Fa>] -         Append projectId to topic name in kafka tour to avoid clashes with other topics\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1603'>HOPSWORKS-1603\u003C\u002Fa>] -         Kagent JWTHelper should retry in case of hopsworks authentication failure\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1610'>HOPSWORKS-1610\u003C\u002Fa>] -         Prometheus alerting-rules files owned by root\r\n\u003C\u002Fli>\r\n\u003C\u002Ful>\r\n            \r\n\u003Ch2>        Improvement\r\n\u003C\u002Fh2>\r\n\u003Cul>\r\n\u003Cli>[\u003Ca href='h","2020-02-10T12:36:44",{"id":249,"version":250,"summary_zh":251,"released_at":252},90573,"v1.1.0","\r\n        Release notes - Hopsworks - Version 1.1.0\r\n    \r\n\u003Ch2>        Bug\r\n\u003C\u002Fh2>\r\n\u003Cul>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1214'>HOPSWORKS-1214\u003C\u002Fa>] -         Airflow HopsworksLaunchOperator does not handle job state App Master Failed \r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1245'>HOPSWORKS-1245\u003C\u002Fa>] -         [ePipe] Fixes for xattrs handling to be work with the changes introduced in HOPS-1434 and HOPS-1451\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1246'>HOPSWORKS-1246\u003C\u002Fa>] -         [ePipe] Check for the existence of elastic fields before delete\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1296'>HOPSWORKS-1296\u003C\u002Fa>] -         [ePipe] log replay fails on xattr op\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1305'>HOPSWORKS-1305\u003C\u002Fa>] -         Fix broken Anaconda export yml\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1318'>HOPSWORKS-1318\u003C\u002Fa>] -         Featurestore should throw an error if you try to create the same featuregroup twice\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1323'>HOPSWORKS-1323\u003C\u002Fa>] -         [ndb-chef] Fix bug in NDB cluster-init.sh\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1325'>HOPSWORKS-1325\u003C\u002Fa>] -         Online Feature store tests fail when project name contains uppercase letters\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1342'>HOPSWORKS-1342\u003C\u002Fa>] -         Add parameter for s3 encryption type in core-site.xml\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1353'>HOPSWORKS-1353\u003C\u002Fa>] -         Flink timer should handle case when services are not installed\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1385'>HOPSWORKS-1385\u003C\u002Fa>] -         HA elastic does not have a synchronization point between starting nodes and creating indices\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1398'>HOPSWORKS-1398\u003C\u002Fa>] -         User activation date is irrational\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1408'>HOPSWORKS-1408\u003C\u002Fa>] -         AirflowManager kills Hopsworks deployments if Airflow directory does not exist\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1412'>HOPSWORKS-1412\u003C\u002Fa>] -         Fix create featuregroup with partition keys\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1413'>HOPSWORKS-1413\u003C\u002Fa>] -         AWS S3 does not accept empty Authorization header when downloading from a  public bucket\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1425'>HOPSWORKS-1425\u003C\u002Fa>] -         Update stack to handle hopsfs paths starting by hopsfs instead of hdfs\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1426'>HOPSWORKS-1426\u003C\u002Fa>] -         Avoid using 100% containers for yarn containers\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1428'>HOPSWORKS-1428\u003C\u002Fa>] -         hops_airflow fails to install pymssql 3.0.3, pin it to 2.1.4\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1431'>HOPSWORKS-1431\u003C\u002Fa>] -         Better handling of Jupyter\u002FGit failures\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1432'>HOPSWORKS-1432\u003C\u002Fa>] -         Elasticsearch node should always use its private ip\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1434'>HOPSWORKS-1434\u003C\u002Fa>] -         Scale jvm -Xmx for Spark and Jupyter based on allocated memory\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1442'>HOPSWORKS-1442\u003C\u002Fa>] -         Update statistics job shows app file as undefined\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1443'>HOPSWORKS-1443\u003C\u002Fa>] -         [featurestore] Spark returning Infinity values for mean statistics of string type features\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1451'>HOPSWORKS-1451\u003C\u002Fa>] -         Storage connector ignored when creating a training dataset from the UI \r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1456'>HOPSWORKS-1456\u003C\u002Fa>] -         [hops-hadoop-chef] Hopsworks security related properties are not templated correctly \r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1457'>HOPSWORKS-1457\u003C\u002Fa>] -         Correct systemd scripts \r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1458'>HOPSWORKS-1458\u003C\u002Fa>] -         spark should not bring any hops jar with it\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1461'>HOPSWORKS-1461\u003C\u002Fa>] -         The JWT signing key should be decoded first before using it to sign jwt\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHO","2020-02-10T12:32:34",{"id":254,"version":255,"summary_zh":256,"released_at":257},90574,"v1.0.0","\r\n        Release notes - Hopsworks - Version 1.0.0\r\n    \r\n\u003Ch2>        Bug\r\n\u003C\u002Fh2>\r\n\u003Cul>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-432'>HOPSWORKS-432\u003C\u002Fa>] -         Clean Kafka ACLs when unsharing a topic or deleting a project.\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1095'>HOPSWORKS-1095\u003C\u002Fa>] -         Localhost (127.0.0.1) cluster definition support\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1114'>HOPSWORKS-1114\u003C\u002Fa>] -         Rework password recovery\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1146'>HOPSWORKS-1146\u003C\u002Fa>] -         Kagent stops reporting if env creation fails with non default repo\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1152'>HOPSWORKS-1152\u003C\u002Fa>] -         Dataset browser not displaying contents of shared datasets with naming conflicts\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1172'>HOPSWORKS-1172\u003C\u002Fa>] -         Upload progress not visible in the beginning of uploading large files\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1175'>HOPSWORKS-1175\u003C\u002Fa>] -         yarnapp\u002Fuser attribute not propagated correctly to yarn-site.xml\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1176'>HOPSWORKS-1176\u003C\u002Fa>] -         Security risk for default admin password pops up even if the password has been changed\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1178'>HOPSWORKS-1178\u003C\u002Fa>] -         [airflow-chef] Add default timezone for Airflow\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1180'>HOPSWORKS-1180\u003C\u002Fa>] -         Chef upgrade mixes kafka brokers ids\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1185'>HOPSWORKS-1185\u003C\u002Fa>] -         Jupyter not working when Hopsworks runs behind a LB\u002Freverse proxy w\u002F TLS termination\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1190'>HOPSWORKS-1190\u003C\u002Fa>] -         [featurestore] missing argument when writing petastorm datasets\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1195'>HOPSWORKS-1195\u003C\u002Fa>] -         Bad handling of Job resource missing\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1197'>HOPSWORKS-1197\u003C\u002Fa>] -         Fix broken admin and agent tests\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1198'>HOPSWORKS-1198\u003C\u002Fa>] -         Cleanup hadoop sbin scripts\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1199'>HOPSWORKS-1199\u003C\u002Fa>] -         ErrorCode range mismatch and TestRestCodes not testing all ErrorCodes\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1203'>HOPSWORKS-1203\u003C\u002Fa>] -         [ePipe] Fixes for JSON handling in the XAttr table\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1207'>HOPSWORKS-1207\u003C\u002Fa>] -         Hopsworks broken on Ubuntu 18.04\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1209'>HOPSWORKS-1209\u003C\u002Fa>] -         filebeat_kagent does not start with kagent start-all services script\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1210'>HOPSWORKS-1210\u003C\u002Fa>] -         Jupyter now requires the python kernel.json to be under python{2, 3} dir.\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1215'>HOPSWORKS-1215\u003C\u002Fa>] -         Use OSProcessExecutor for Python library search\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1216'>HOPSWORKS-1216\u003C\u002Fa>] -         tensorflow-chef install recipe CentOS hiccups\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1218'>HOPSWORKS-1218\u003C\u002Fa>] -         Python service not working when searching or installing libraries containing 'http'\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1225'>HOPSWORKS-1225\u003C\u002Fa>] -         add attribute for url to download spark in spark-chef\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1228'>HOPSWORKS-1228\u003C\u002Fa>] -         Remove livy session name from sparkmagic conf file\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1233'>HOPSWORKS-1233\u003C\u002Fa>] -         [ePipe] hivecleaner fail during project removal\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1234'>HOPSWORKS-1234\u003C\u002Fa>] -         Remove sudo command from sudoers scripts\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1236'>HOPSWORKS-1236\u003C\u002Fa>] -         Project cleanup does not work after a failed project creation\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-1246'>HOPSWORKS-1246\u003C\u002Fa>] -         [ePipe] Check for the existence of elastic fields before delete\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbr","2019-10-28T21:32:25",{"id":259,"version":260,"summary_zh":261,"released_at":262},90575,"v0.10.0","\r\n        Release notes - Hopsworks - Version 0.10.0\r\n    \r\n\u003Ch2>        Bug\r\n\u003C\u002Fh2>\r\n\u003Cul>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-197'>HOPSWORKS-197\u003C\u002Fa>] -         Turning on and off tours is a terrible user experience\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-715'>HOPSWORKS-715\u003C\u002Fa>] -         Hopsworks singletons are not single\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-770'>HOPSWORKS-770\u003C\u002Fa>] -         Disable http port and TLS 1.0 on glassfish\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-849'>HOPSWORKS-849\u003C\u002Fa>] -         kagent - conda can get stuck\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-903'>HOPSWORKS-903\u003C\u002Fa>] -         Jupyter cleanup and timer bugfixes \r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-904'>HOPSWORKS-904\u003C\u002Fa>] -         Users should be able to specify version of Python library to install when behind PyPi proxy\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-905'>HOPSWORKS-905\u003C\u002Fa>] -         remote_material_references is not cleaned up during project deletion\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-913'>HOPSWORKS-913\u003C\u002Fa>] -         Feature store quota missing from admin UI in hopsworks\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-917'>HOPSWORKS-917\u003C\u002Fa>] -         Non HA admin UI when HA namenodes\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-920'>HOPSWORKS-920\u003C\u002Fa>] -         Job monitor thread gets stuck if job is killed while running\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-927'>HOPSWORKS-927\u003C\u002Fa>] -         TensorBoard could not bind to an unsupported address family\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-928'>HOPSWORKS-928\u003C\u002Fa>] -         [Featurestore] Bug: nested spark schemas are not automatically translatable to Hive\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-929'>HOPSWORKS-929\u003C\u002Fa>] -         [Featurestore] Don't allow hyphen in featuregroup names because Hive do not allow it\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-930'>HOPSWORKS-930\u003C\u002Fa>] -         [Featurestore] bug, in API retrieval code in feature-registry featurestore is undefined\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-932'>HOPSWORKS-932\u003C\u002Fa>] -         Jobs UI date selector should persist user-selected values\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-937'>HOPSWORKS-937\u003C\u002Fa>] -         Adjust Cgroup configuration to account for longer Cgroup deletion and make CPU utilization configurable\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-938'>HOPSWORKS-938\u003C\u002Fa>] -         Adding pyFiles to Jupyter does not add it to python path\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-941'>HOPSWORKS-941\u003C\u002Fa>] -         Fix dataset access request and related tests\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-945'>HOPSWORKS-945\u003C\u002Fa>] -         Zipping and unzipping is broken with Hops TLS\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-948'>HOPSWORKS-948\u003C\u002Fa>] -         [hops-hadoop-chef] Copy hadoop logs utility is not removed from crontab\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-949'>HOPSWORKS-949\u003C\u002Fa>] -         [airflow-chef] Install recipe does not create root hops installation directory\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-951'>HOPSWORKS-951\u003C\u002Fa>] -         sqoop and airflow bugfixes\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-953'>HOPSWORKS-953\u003C\u002Fa>] -         [kagent-chef] kagent falsely assumes service is dead when it restarts\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-955'>HOPSWORKS-955\u003C\u002Fa>] -         [airflow-chef] Use assigned roles in JWT to determine if a user is admin\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-958'>HOPSWORKS-958\u003C\u002Fa>] -         AirflowJWTManager randomly does not renew tokens\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-962'>HOPSWORKS-962\u003C\u002Fa>] -         copy_to_hdfs is broken in hops-util-py\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-965'>HOPSWORKS-965\u003C\u002Fa>] -         Certificate enddate should always be UTC\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-968'>HOPSWORKS-968\u003C\u002Fa>] -         Dataset browser  pagination and select all not working\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-970'>HOPSWORKS-970\u003C\u002Fa>] -         upgrade chef ulimit cookbook\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-972'>HO","2019-07-02T14:22:32",{"id":264,"version":265,"summary_zh":266,"released_at":267},90576,"v0.9.1","\r\n        Release notes - Hopsworks - Version 0.9.1\r\n    \r\n\u003Ch2>        Bug\r\n\u003C\u002Fh2>\r\n\u003Cul>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-197'>HOPSWORKS-197\u003C\u002Fa>] -         Turning on and off tours is a terrible user experience\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-903'>HOPSWORKS-903\u003C\u002Fa>] -         Jupyter cleanup and timer bugfixes \r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-913'>HOPSWORKS-913\u003C\u002Fa>] -         Feature store quota missing from admin UI in hopsworks\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-920'>HOPSWORKS-920\u003C\u002Fa>] -         Job monitor thread gets stuck if job is killed while running\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-927'>HOPSWORKS-927\u003C\u002Fa>] -         TensorBoard could not bind to an unsupported address family\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-928'>HOPSWORKS-928\u003C\u002Fa>] -         [Featurestore] Bug: nested spark schemas are not automatically translatable to Hive\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-929'>HOPSWORKS-929\u003C\u002Fa>] -         [Featurestore] Don't allow hyphen in featuregroup names because Hive do not allow it\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-930'>HOPSWORKS-930\u003C\u002Fa>] -         [Featurestore] bug, in API retrieval code in feature-registry featurestore is undefined\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-932'>HOPSWORKS-932\u003C\u002Fa>] -         Jobs UI date selector should persist user-selected values\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-938'>HOPSWORKS-938\u003C\u002Fa>] -         Adding pyFiles to Jupyter does not add it to python path\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-941'>HOPSWORKS-941\u003C\u002Fa>] -         Fix dataset access request and related tests\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-945'>HOPSWORKS-945\u003C\u002Fa>] -         Zipping and unzipping is broken with Hops TLS\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-949'>HOPSWORKS-949\u003C\u002Fa>] -         [airflow-chef] Install recipe does not create root hops installation directory\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-951'>HOPSWORKS-951\u003C\u002Fa>] -         sqoop and airflow bugfixes\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-953'>HOPSWORKS-953\u003C\u002Fa>] -         [kagent-chef] kagent falsely assumes service is dead when it restarts\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-955'>HOPSWORKS-955\u003C\u002Fa>] -         [airflow-chef] Use assigned roles in JWT to determine if a user is admin\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-958'>HOPSWORKS-958\u003C\u002Fa>] -         AirflowJWTManager randomly does not renew tokens\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-968'>HOPSWORKS-968\u003C\u002Fa>] -         Dataset browser  pagination and select all not working\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-974'>HOPSWORKS-974\u003C\u002Fa>] -         Blank padded day in rspec breaks tests\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-975'>HOPSWORKS-975\u003C\u002Fa>] -         [airflow-chef] Change restart policy of airflow scheduler\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-976'>HOPSWORKS-976\u003C\u002Fa>] -         Airflow file manager does not refresh secret directory when navigate to another project\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-977'>HOPSWORKS-977\u003C\u002Fa>] -         Pin tornado version to 5.1.1\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-978'>HOPSWORKS-978\u003C\u002Fa>] -         dela::install should create \u002Fsrv\u002Fhops\u002F before creating sub directories\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-979'>HOPSWORKS-979\u003C\u002Fa>] -         Cannot change default 'admin' password of admin@kth.se\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-987'>HOPSWORKS-987\u003C\u002Fa>] -         User admin to slow and user search returning duplicated results \r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-988'>HOPSWORKS-988\u003C\u002Fa>] -         Launching Jupyter server gradually gets slower and times out\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-990'>HOPSWORKS-990\u003C\u002Fa>] -         uploading big files from web ui can cause 'invalidated token'\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-992'>HOPSWORKS-992\u003C\u002Fa>] -         [kagent-chef] Invalid command in kagent's service status script\r\n\u003C\u002Fli>\r\n\u003Cli>[\u003Ca href='https:\u002F\u002Flogicalclocks.atlassian.net\u002Fbrowse\u002FHOPSWORKS-993'>HOPSWORKS-993\u003C\u002Fa>] -         [Feturestore] tf record schema infer for float arrays is no","2019-04-04T09:03:42"]