提交 13f9af8d 编写于 作者: C chenyujiahw

Merge remote-tracking branch 'upstream/master' into addBanner

文件已删除
...@@ -3,3 +3,4 @@ resources ...@@ -3,3 +3,4 @@ resources
resources/_gen resources/_gen
.idea .idea
.DS_Store
\ No newline at end of file
...@@ -162,14 +162,14 @@ paginate = 10 ...@@ -162,14 +162,14 @@ paginate = 10
[[menu.main]] [[menu.main]]
identifier = "Advisories" identifier = "Advisories"
name = "Security Advisories" name = "Security Advisories"
url = "https://cve.openeuler.org/cve#/home?lang=en" url = "https://cve.openeuler.org/#/home?lang=en"
weight = 2 weight = 2
parent = "security" parent = "security"
[[menu.main]] [[menu.main]]
identifier = "CVE" identifier = "CVE"
name = "CVE" name = "CVE"
url = "https://cve.openeuler.org/cve#/CVE?lang=en" url = "https://cve.openeuler.org/#/CVE?lang=en"
weight = 3 weight = 3
parent = "security" parent = "security"
...@@ -354,14 +354,14 @@ paginate = 10 ...@@ -354,14 +354,14 @@ paginate = 10
[[languages.zh.menu.main]] [[languages.zh.menu.main]]
identifier = "Advisories" identifier = "Advisories"
name = "安全公告" name = "安全公告"
url = "https://cve.openeuler.org/cve#/home?lang=zh" url = "https://cve.openeuler.org/#/home?lang=zh"
weight = 2 weight = 2
parent = "security" parent = "security"
[[languages.zh.menu.main]] [[languages.zh.menu.main]]
identifier = "CVE" identifier = "CVE"
name = "CVE" name = "CVE"
url = "https://cve.openeuler.org/cve#/CVE?lang=zh" url = "https://cve.openeuler.org/#/CVE?lang=zh"
weight = 3 weight = 3
parent = "security" parent = "security"
...@@ -574,7 +574,7 @@ paginate = 10 ...@@ -574,7 +574,7 @@ paginate = 10
[params.copyright] [params.copyright]
enable = true enable = true
text = "Copyright © openEuler 2020. All rights reserved.<br><a href=\"/en/brand.html\">Brand</a>&nbsp;&nbsp;<a href=\"/en/legal.html\">Legal Notice</a>" text = "Copyright © openEuler 2020. All rights reserved.<br><a href=\"/en/brand.html\">Brand</a>&nbsp;&nbsp;<a href=\"/en/legal.html\">Private Policy</a>&nbsp;&nbsp;<a href=\"/en/legal.html\">Legal Notice</a>"
[languages.zh.params.copyright] [languages.zh.params.copyright]
enable = true enable = true
......
+++
title = "The Future of Cloud in a COVID-19 Era"
date = "2020-06-24"
tags = ["cloud","COVID-19","Open Hardware"]
archives = "2020-06"
author = "ByBryan Che"
summary = "How was Huawei Cloud able to grow so quickly over such a short time? One of the things that distinguishes Huawei is that we believe in the power of open innovation."
+++
# The Future of Cloud in a COVID-19 Era
## **Three Years of Astonishing Cloud Growth**
Almost exactly three years ago in June 2017, Huawei launched its public cloud business, [Huawei Cloud](https://www.huaweicloud.com/en-us/), in China. At the time, China’s public cloud market had just exceeded **US$1 billion in the 1H 2017**.
The top vendors in the space were:
<img src="https://blog.huawei.com/wp-content/uploads/2020/06/1-3.png" />
Source: [China Internet Watch](https://www.chinainternetwatch.com/22936/public-cloud-market-share-h1-2017/) / **Public Cloud Market Share in H1 2017 (IaaS)**
Three years later, and Canalys has just released its latest figures for the China public cloud market, as of June 2020:
<img src="https://blog.huawei.com/wp-content/uploads/2020/06/2-3.png" />
Several things are noteworthy from a comparison between 2017 and 2020:
- Over 3 years, China’s public cloud infrastructure market has grown about 800%, from $1B over 6 months in H1 2017 to US$3.9 billion over 3 months in Q1 2020.
- China’s public cloud market is coalescing around its larger technology leaders, BAT and Huawei.
- During these 3 years, Huawei Cloud grew and gained share the fastest, going from non-existent to now the #2 public cloud in China.
On this 3-year anniversary of Huawei Cloud, I’d like to take the opportunity to reflect on a few key trends in cloud computing: the effects of COVID-19 on cloud, the evolution of global cloud industries, and the importance of being open.
## **COVID-19 and the Acceleration of Digital Trends**
COVID-19 has been a terrible blight upon the world, inflicting both human suffering and economic hardship. As people have been staying at home and moving interactions online, the use of digital services and connections has spiked dramatically.
From events to meetings to media to shopping, everything that can be online is moving online. Of course, in China, many of these activities had already moved online and onto smartphones years ago—and at much higher adoption rates than in other countries. Ubiquitous app-based food delivery, restaurants with QR-code-based menus, cashless spending, and many other digital services are so normal here, they are the rule rather than the exception.
Even with all these digital services, though, China’s total spending on public cloud infrastructure is still relatively small both for its economy size and also when compared with other markets. **So, it is remarkable that even in China, COVID-19 has brought such a dramatic increase in spending on cloud computing services**.
<center class="half">
<img src="https://blog.huawei.com/wp-content/uploads/2020/06/shutterstock_495697993.jpg" width="200" height="200"/><img src="https://blog.huawei.com/wp-content/uploads/2020/06/3.jpg" width="200" height="200"/><img src="https://blog.huawei.com/wp-content/uploads/2020/06/shutterstock_1268311450-1024x869.jpg" width="200" height="200"/>
</center>
As evidenced by a 67% increase in Q1 cloud spending against a massive global slowdown, COVID-19 has demonstrated to Chinese companies that in times of abrupt need, only cloud can truly deliver the necessary scale, agility, and depth of services required. Thus, even after this COVID-19 crisis has passed, the acceleration towards cloud in China will continue.
## **The Evolution of Global Cloud Industries**
As China and the world accelerate their adoption of cloud, what will this look like? Different regions will evolve differently, but they will share many similar characteristics. For example, consider both China and Europe.
## **China and New Infrastructure**
To help address the economic fallout from COVID-19, the Chinese government recently announced to accelerate a multi-trillion yuan investment plan towards **New Infrastructure**, the goal of which is to provide ubiquitous and high quality connectivity and computing. Target technologies and industries for New Infrastructure spending include cloud, AI, 5G, blockchain, and industrial internet.
<img src="https://blog.huawei.com/wp-content/uploads/2020/06/shutterstock_741922060-1024x683.jpg" />
So, not only did COVID-19 bring a spike in cloud consumption in China while everyone stayed home, it is also leading to a massive new investment across the country in digital technologies. This in turn will only further accelerate the adoption of cloud services in China.
# Europe and Digital Sovereignty
Like China, the European Commission has recently introduced a massive €750 billion stimulus plan to deal with the COVID-19 fallout, *Next Generation EU*. And, like with China’s New Infrastructure, Europe’s **Next Generation EU** calls for spending heavily on projects towards strengthening for the digital age:
<video autoplay="" loop="" muted="" src="https://blog.huawei.com/wp-content/uploads/2020/06/581695-v2-3.mp4" style="max-width: 100%; width: 720px; margin: 10px 0px; object-fit: cover;"></video>
- Investing in more and better connectivity, especially in the rapid deployment of 5G networks.
- A stronger industrial and technological presence in strategic sectors, including artificial intelligence, cybersecurity, supercomputing and cloud.
- Building a real data economy as a motor for innovation and job creation.
- Increased cyber resilience.
In addition, though, as Europe invests in these technologies, it is also doing so with an eye towards digital sovereignty. And so, also this past month, [GAIA-X](https://www.data-infrastructure.eu/GAIAX/Navigation/EN/Home/home.html), a project to build a federated digital infrastructure for Europe, officially announced their new European foundation:
> *With GAIA-X, representatives from politics, business and science from France and Germany, together with other European partners, create a proposal for the next generation of a data infrastructure for Europe: a secure, federated system that meets the highest standards of digital sovereignty while promoting innovation. This project is the cradle of an open, transparent digital ecosystem, where data and services can be made available, collated and shared in an environment of trust.*
>
> [Data Infrastructure EU, GAIA-X](https://www.data-infrastructure.eu/GAIAX/Navigation/EN/Home/home.html)
According to a recent Gartner study, the adoption rate of public cloud for many European countries is similar to that of China and also lags many other markets:
<img src="https://blog.huawei.com/wp-content/uploads/2020/06/15.png" />
With projects like Next Generation EU and GAIA-X, however, European adoption of cloud and digital technologies should rapidly accelerate over the next few years. And just as in China and around the world, we look forward to sharing our experience and expertise and to working together openly and securely to build out this next generation of digital services.
# **The Importance of Being Open**
How was Huawei Cloud able to grow so quickly over such a short time? One of the things that distinguishes Huawei is that we believe in the power of open innovation.
This is how we view our Huawei Cloud business:
<img src="https://blog.huawei.com/wp-content/uploads/2020/06/QQ%E6%8B%BC%E9%9F%B3%E6%88%AA%E5%9B%BE%E6%9C%AA%E5%91%BD%E5%90%8D.png" />
## **Open Hardware**
First, Huawei believes that as we move towards next generation cloud services, AI, energy efficiency, edge computing, and many additional factors are going to drive cloud towards heterogeneous architectures. To support that, Huawei offers a variety of advanced chipsets optimized for different types of use cases, including x86 CPU, Arm CPU, GPU, NPU etc.
Unlike other cloud providers, though, we share our hardware and chipsets openly with other vendors to implement their own systems and clouds. We do not restrict them only to Huawei Cloud.
## **Open Source Software**
Huawei Cloud is built on open source software. From OpenStack at the IaaS layer to Kubernetes and other CNCF projects at the container and PaaS layer to projects like our own open source AI framework [MindSpore](https://www.mindspore.cn/) for innovative services, Huawei Cloud is arguably the most open cloud in the industry.
## **Open, Collaborative Innovation**
Huawei is not just consuming open source software, though. Indeed, we are a top-level member and top contributor across all the major open source foundations and projects, including Linux Foundation, CNCF, OpenStack Foundation, Apache Foundation, and Eclipse Foundation. It is by deeply participating in the open source ecosystem that we are able to bring such fast innovation to market and grow at such a high rate.
It’s not just in open source communities that we collaborate with others. This is why we have such a big partner focus at Huawei Cloud and look to enable them to be successful as well. When they do well, our customers do well, and we all do well together.
# **To Shared Recovery and Triumph**
I hope that during this period of deep trouble, a spirit of open, collaborative innovation can help us all work together for our shared recovery. Cloud computing works because it shares resources across many different users. Huawei Cloud has succeeded because we share in the open development and benefits of cloud.
Europe, China, and the world will all have different paths forward out of our current situation. But, may the next three years and beyond bring for everyone a hopeful and triumphant future.
\ No newline at end of file
+++
title = "openEuler Becomes a Member of the CNA Program"
date = "2020-06-24"
tags = ["CVE", "CNA", "Security"]
archives = "2020-06-24"
author = "liujingang09, openEuler Security Committee"
summary = "openEuler Becomes a Member of the CNA Program"
+++
### openEuler Becomes a Member of the CNA Program
The openEuler community attaches great importance to the community version security. To quickly respond to and handle security issues related to the openEuler, the community has developed a complete vulnerability management policy. On June 24, 2020, openEuler joins the CVE Numbering Authority (CNA) Program. Currently, openEuler is entitled to assign and manage CVEs related to the openEuler community. By joining the CNA Program, openEuler applies mature vulnerability management standards in the industry to promote the community cyber security.
The security committee of openEuler community is responsible for building community security engineering and improving vulnerability response capabilities. We hope that security experts and enthusiasts who are interested in openEuler can join our hands to enhance the openEuler community security.
#### Vulnerability management policy:
https://openeuler.org/en/security.html
#### What is CVE?
+ CVE is an international, community-based effort that maintains a community-driven, open data registry of vulnerabilities.
+ The CVE IDs assigned through the registry enable program stakeholders to rapidly discover and correlate vulnerability information used to protect systems against attacks.
+ The CVE List is built by CVE Numbering Authorities (CNAs). Every CVE Entry added to the list is assigned by a CNA.
+ The CVE List feeds the U.S. National Vulnerability Database (NVD).
#### CVE Value:
+ CVE enables two or more people or tools to refer to a vulnerability and know they are talking about the same thing, resulting in significant time and cost savings.
#### CVE is Community Driven:
+ The CVE Program relies on the community (vendors, end users, researchers, and more) to discover and register vulnerabilities.
+ CVE IDs are assigned by CVE Numbering Authorities (CNAs), which are operated on a voluntary basis by participating organizations.
+ The CVE Board, which drives the direction of the CVE Program, consists of industry, academic, and government representatives from around the world.
+ CVE Working Groups develop the program’s policies (approved by the CVE Board) and are open to the community.
#### Sponsored by:
+ [The CVE Program](https://cve.mitre.org/) is sponsored by the Cybersecurity and Infrastructure Security Agency (CISA, [https://www.cisa.gov/](https://www.cisa.gov/)of the U.S. Department of Homeland Security (DHS) and is operated by [the MITRE Corporation](https://www.mitre.org/) in close collaboration with international industry, academic, and government stakeholders
#### What are CNAs (CVE Numbering Authorities)
+ CNAs are organizations authorized by the CVE Program to assign CVE IDs to vulnerabilities affecting products within their distinct, agreed-upon scope.
...@@ -61,7 +61,7 @@ And if you are going to post a blog in Chinese, the content/zh/blog is your work ...@@ -61,7 +61,7 @@ And if you are going to post a blog in Chinese, the content/zh/blog is your work
Assume that you are preparing a English blog. Assume that you are preparing a English blog.
``` ```
cd content/en/post cd content/en/blog
mkdir <your-gitee-id> mkdir <your-gitee-id>
cd <your-gitee-id> cd <your-gitee-id>
touch YEAR-MONTH-DAY-title.md touch YEAR-MONTH-DAY-title.md
......
+++
listname = "Crystal-ci"
emailaddress = "crystal-ci@openeuler.org"
description = "Mail list used for the research and development of project crystal-ci."
websitelink = "https://mailweb.openeuler.org/postorius/lists/crystal-ci.openeuler.org/"
archivelink = "https://mailweb.openeuler.org/hyperkitty/list/crystal-ci@openeuler.org"
weight = 15
+++
+++
listname = "QA"
emailaddress = "qa@openeuler.org"
description = " Main mail list for discussion of testing and quality assurance of openEuler."
websitelink = "https://mailweb.openeuler.org/postorius/lists/qa.openeuler.org/"
archivelink = "https://mailweb.openeuler.org/hyperkitty/list/qa@openeuler.org"
weight = 13
+++
+++
listname = "Sig-ai-bigdata"
emailaddress = "sig-ai-bigdata@openeuler.org"
description = "Mail list in openEuler, used for the discussions about the information of ai and bigdata software ecosystem."
websitelink = "https://mailweb.openeuler.org/postorius/lists/sig-ai-bigdata.openeuler.org/"
archivelink = "https://mailweb.openeuler.org/hyperkitty/list/sig-ai-bigdata@openeuler.org"
weight = 14
+++
...@@ -6,7 +6,7 @@ ...@@ -6,7 +6,7 @@
### 回答 ### 回答
这个问题保护了如下几种场景: 这个问题包括了如下几种场景:
#### 1. 我想找到当前存在的项目组有哪些,以及他们维护的项目 #### 1. 我想找到当前存在的项目组有哪些,以及他们维护的项目
......
## 本文目的
本文面向的读者是技术委员会委员、SIG Maintainer 或社区会议的组织者,目的是指导和帮助会议组织者在开源社区中组织高效的会议。
## 什么是开源社区会议?
因为开源社区存在异步和远程协作特点,开源社区的会议多指在线会议 (Online Conference) 。开源社区会议的作用是通过会议提升社区协作效率,加强社区成员之间的了解程度提升社区凝聚力,同时通过开放的社区会议吸引社区参与者。
会议内容通常包括:
* 讨论解决社区中遇到的管理问题
* 同步项目编码进展、讨论 Milestone、分析 Issue 等技术问题
开源社区会议最大的特点是 “开放、平等和互助” ,即对所有有兴趣参与的成员是开放参加、会议的文字、语音和视频存档是公开可访问;在会议过程中所有参与者都是平等的,有发表看法的权利和自由;参与会议的社区成员相互帮助,而不是抱怨。
开源社区会议可以有如下几种形式:
* 文字会议方式,常见方式是通过 IRC [1] 频道进行文字会议。
* 语音或视频会议
* Welink [2] - 可录制视频到本地
* Zoom [3] - 可录制视频到本地,同时提供云存储服务
* meet.jit.si [4] - 可录制视频,只能保存到 Dropbox 服务
* Google Meeting [5] - 可录制视频,只能保存到 Google Drive 服务
## 会议组织指导
### 开会前的准备
1. 与团队成员确定会议时间,需要考虑时区、工作时间等因素
2. 与团队确定会议方式,采用文字、语音或者视频
3. 确定会议议程,可以设置几个固定议题,再收集临时议题
4. 确定会议记录方式,可以使用etherpad.openeuler.org、Wiki等工具。推荐Etherpad,因为不需要权限控制,方便大家各自填写信息
5. 发送会议通知,建议用 Calendar 发送会议时间和议题
### 会议过程
1. 根据会议议程分配好会议时间,按照时间安排进行议题,避免拖堂。
2. 会议开始后,如果希望录音或者录视频,提醒并得到与会人员的认可。
3. 点名(可选)。如果涉及到决议,比如技术委员会开会,则为了保证结果的合法性,需要点名,确保有足够的有投票权的人员参加了会议。
4. 请与会人员登记姓名以记录与会人。如果在IRC开会,则不需要这一步骤;如果是其他方式,邀请大家在Etherpad或者Wiki等里面自行登记自己的姓名或者ID。
5. 按照议程逐个进行。
6. 针对每一个会议结论,通过文字或者语音与与会人员确认,确保对会议结论理解一致。
### 开会后的工作
1. 整理会议纪要,发送会议纪要。
2. 记录会议遗留问题。
### 其他
1. 会议取消。因为节假日或者其他原因需要取消会议时,提前在Calendar里面发送取消通知给对应的Mail list。
2. 会议礼仪
1. 考虑到Mail list的接收方很广泛,为了避免大邮件,在邮件中*不要使用图片、附件*,如果需要,则通过将图片、附件等放到可以公开访问的云上,在邮件中引用链接。
2. 感谢。感谢大家参加会议,感谢大家对会议议程的贡献,由衷、及时地表达谢意。
3. 鼓励与会人发言。开会的目的是充分交流,需要经常鼓励与会人发表意见。
4. 照顾非母语的与会人。比如以英语开会时,考虑到非英语母语的与会人参与会有困难,在过程中要耐心确认对方是否理解,并及时将关键信息以文字形式写到Etherpad或者IRC中,帮助其理解。
5. 开放地听取他人的观点。无论是否认同其观点,都要保持开放心态充分理解其观点。
## 参考信息
[1] https://zh.wikipedia.org/wiki/IRC
[2] https://welink.huaweicloud.com
[3] https://zoom.us
[4] https://meet.jit.si
[5] https://meet.google.com
[6] https://opensource.com/open-organization/16/1/community-calls-will-increase-participation-your-open-organization
[7] https://osf.dev/about/four-opens/
---
# 作者
[Fred Li](https://gitee.com/zerodefect)
# 协作者
1. [Eli Ma](https://gitee.com/genedna)
...@@ -8,4 +8,5 @@ headless: true ...@@ -8,4 +8,5 @@ headless: true
- [openEuler的发布节奏和生命周期政策是怎么样的?]({{< relref "./docs/lifecycle.md" >}}) - [openEuler的发布节奏和生命周期政策是怎么样的?]({{< relref "./docs/lifecycle.md" >}})
- [我怎样才能找到感兴趣的项目组?]({{< relref "./docs/findprojects.md" >}}) - [我怎样才能找到感兴趣的项目组?]({{< relref "./docs/findprojects.md" >}})
- [SIG Maintainer如何增删项目?]({{< relref "./docs/modifyswinsig.md" >}}) - [SIG Maintainer如何增删项目?]({{< relref "./docs/modifyswinsig.md" >}})
- [使用邮件列表经常有问题,怎么办?]({{< relref "./docs/maillistfailure.md" >}}) - [使用邮件列表经常有问题,怎么办?]({{< relref "./docs/maillistfailure.md" >}})
\ No newline at end of file - [如何在开源社区组织一个会议?]({{< relref "./docs/howtomanagemeetings.md" >}})
\ No newline at end of file
此差异已折叠。
+++
title = "openEuler 社区成立 Ceph SIG"
date = "2020-06-16"
tags = ["openEuler 社区", "XSKY", "Ceph"]
archives = "2020-06"
author = "genedna"
summary = "openEuler 社区正式成立Ceph SIG"
+++
## openEuler 社区成立 Ceph SIG
openEuler 开源社区(以下简称“openEuler社区”)正式成立 Ceph SIG(Special Interest Group 特别兴趣小组),完善 openEuler 操作系统在分布式存储场景的适配。 openEuler 社区参与者和用户可获得由专业存储团队维护的稳定的 Ceph 版本构建分布式存储系统,并通过 openEuler 社区获得技术支持。
Ceph 是一个具备高性能、可靠性和可扩展性的分布式存储解决方案,可以同时提供块、文件和对象三种服务。 目前,Ceph 已经成为最主流的开源分布式存储系统。为了进一步丰富 openEuler 环境下的存储解决方案,经过 openEuler 技术委员会讨论,正式成立 Ceph SIG, 当前 Ceph SIG 的第一位 Maintainer 为软件定义存储厂商 XSKY 。XSKY 将会把 Ceph 集成到 openEuler 的发行版本,使得 openEuler 具备分布式存储场景的适配能力。根据计划,在今年Q3前,openEuler 社区将完成 Ceph 最新主流发行版本的所有适配工作,并适时启动自动化 QA 框架的搭建。
XSKY 公司作为全球领先的 Ceph 开源社区推动者之一,源代码贡献居全球前列。一直以来,XSKY 主导并维护了 Ceph 社区的多个模块,也是 Ceph 基金会的全球创始顶级会员。
openEuler 社区技术委员会委员熊伟表示:“成立 Ceph SIG 将进一步增强 openEuler 面向存储场景的适配能力。XSKY 公司作为项目组的 Maintainer ,能够加速 Ceph 在 openEuler 的落地,并提供更加快速的版本迭代计划和问题修复能力,结合 openEuler 和 Ceph 版本发布计划为社区提供最新发行版本。
XSKY公司资深架构师池信泽表示:“openEuler 致力于提供一个开放、多元和架构包容的软件生态系统,帮助用户快速构建 5G、边缘、云场景等多元化的应用环境。我们非常高兴能够成为 openEuler 社区 Ceph SIG 的 Maintainer,有机会推动Ceph和openEuler两个社区的连接。未来,我们将和众多开发者一起为 openEuler 社区提供存储领域丰富的企业特性。”
### 关于XSKY
[XSKY](https://www.xsky.com) 是中国软件定义存储的领导厂商,致力于以中国技术力量影响开放平台生态系统,创建自主可控的底层设施,以主流的、先进的技术和产品为客户创造价值。 IDC 将 XSKY 定义为“2019 中国软件定义存储创新者”。2020 年,XSKY 入选 《Gartner Peer Insights‘Voice of the Customer’:Distributed File Systems and Object Storage》 报告,并位列第二象限。 根据 IDC 《2019Q4 China SDS Market Overview》显示:截至2019年,在中国软件定义存储(SDS)整体市场份额排名中,XSKY 位居第四;其中对象存储细分市场排名第一(22.2%),已连续11个季度占据国内市场第一位,块存储细分市场排名第三(14.1%)。
+++
title = "openEuler正式成为CNA,获得CVE颁发资质"
date = "2020-06-24"
tags = ["CVE", "CNA", "Security"]
archives = "2020-06-24"
author = "liujingang09, openEuler Security Committee"
summary = "openEuler正式成为CNA,获得CVE颁发资质"
+++
### openEuler正式成为CNA,获得CVE颁发资质
openEuler非常重视社区版本的安全性,制定了一套完整的漏洞管理策略,快速的响应和处理openEuler相关的安全问题。2020年6月24日openEuler顺利通过CNA(CVE Numbering Authorities)加入程序,成为CVE编号授权机构,openEuler同时也拥有颁发和管理openEuler社区相关CVE编号的资质。openEuler通过加入CNA,遵循业界成熟的漏洞管理标准,以促进openEuler社区安全的持续发展。
openEuler安全委员会一直致力于提升社区安全和隐私保护能力,欢迎关注openEuler的安全专家和爱好者加入openEuler安全委员会,与我们一起构建openEuler社区安全。
#### openEuler漏洞管理策略:
https://openeuler.org/zh/security.html
### 关于CVE
全称是 Common Vulnerabilities & Exposures 通用漏洞披露,建立于 1999年9月。 CVE是一个漏洞字典表,每个漏洞都拥有一个唯一的CVE编码。用户可以通过唯一的CVE编码在漏洞数据库或安全工具中快速的找到漏洞影响范围和修补信息,以便快速的确认系统受漏洞影响情况和获取解决方案。
### 关于CNA
全称是 CVE Numbering Authority,即“CVE编号授权机构”,CNA包括供应商、开源项目、漏洞研究人员、国家/行业CERT等,成为CNA将可以在授权范围内分配和管理CVE编号。
此差异已折叠。
+++
title = "A-Tune让系统调优更智能"
date = "2020-05-08"
tags = ["A-Tune","调优","智能"]
archives = "2020-05"
author = "openEuler Blog Maintainer"
summary = "A-Tune让系统调优更智能"
+++
# A-Tune让系统调优更智能
2020年1月1日,openEuler开源社区如约而至。2020年3月27日,在华为开发者大会(HDC 2020)上,openEuler首个长期支持版openEuler LTS 20.03正式推出。自此,我们的计算世界将变得更为精彩。
那么,openEuler社区除了提供openEuler这个开源Linux操作系统的标准组件之外,还有哪些厉害的子项目呢? A-Tune和iSula便是答案。
接下来,就让我们来揭开A-Tune的神秘面纱吧。
## A-Tune是什么
让我们先从名字来认识下A-Tune吧。“Tune”代表了**性能调优**;“A”则承载了Tune的愿景与定位,其蕴含了3层寓意Aware、Adaptive、Auto,即
**自感知(Aware)**- 自动感知负载和业务模型;
**自适应(Adaptive)**- 动态监测负载特征变化,适时调整资源;
**自调节(Auto)**- 通过感知负载模型,自动匹配优化库。
顾名思义,A-Tune就是一款基于openEuler开发的,自动化、智能化性能调优引擎。它利用人工智能,对运行在操作系统上的业务建立精准模型,感知并推理出业务特征,根据业务特征及负载情况给出最佳的参数配置组合,从而使业务处于最佳系统性能状态下运行。
## 为什么要开发A-Tune
系统调优一直是一个门槛高的系统性工程,强依赖工程师的技能和经验。如,一个简单的应用,除了自身代码外,支撑其运行的环境,如硬件平台、操作系统、数据库等都可能是影响性能的因素。如何在众多因素中找到性能瓶颈,需要工程师们熟悉大量参数的含义、配置方法,以及业务场景,并不断积累经验,才能对系统进行快速精准调优。
随着各行各业数字化智能化转型,新应用、新技术、新计算架构层出不穷,传统系统调优更是面临以下挑战:
- 涉及上万的能力组合,挑战人工调优极限
运行在操作系统上的业务类型成百上千,形态千差万别,从硬件到操作系统,到中间件再到上层应用,涉及的调优参数组合上万+。随着业务复杂度和调优对象的增加,调优组合和调优所需成本指数级增长,大大超过了工程师的能力范围。
<img src="/zh/blog/shanshishi/2020-05-08-atune-tuning-smarter-01.png" style="width:100%;">
- 应用场景剧增,实验室测试无法穷举
实际业务场景成千上万,计算、网络、存储等硬件配置层出不穷,实验室无法遍历所有的应用和业务场景,不可能完全模拟动态负载,无法对不同的场景组合进行穷举测试。
为了降低调优门槛、提升调优效率,A-Tune应运而生。
## A-Tune的优势
A-Tune利用大数据和人工智能技术,通过对各种类型的业务进行大数据分析,建立精准业务模型,并制定相应的调优策略。A-Tune具有如下优势:
- **简化系统调优**
A-Tune尽可能的屏蔽了硬件和操作系统底层细节,工程师无需感知底层细节,就可以快速上手调优。因此,降低了对工程师的技能要求,简化了系统调优。
- **提升调优效率**
A-Tune利用大数据和人工智能技术,从CPU、IO、网络、内存等多个维度抽象刻画不同业务类型系统行为的特征,并对收集到的结果进行大数据分析和机器学习,识别有效的调优参数,快速匹配多种配置组合,从而找到最佳配置。对于业务复杂、配置项多,能够大大提升调优效率。
- **持续积累调优经验**
A-Tune能够利用人工智能技术自动分析业务特征,寻找优化配置,建立优化模型库。针对具体的业务场景,结合历史经验,训练并持续优化业务模型,积累和优化模型库,满足多业务多场景下的调优。
## A-Tune的架构
A-Tune分层架构如下图所示,主要包括智能决策层、系统画像层、交互系统层以及优化模式库。
<img src="/zh/blog/shanshishi/2020-05-08-atune-tuning-smarter-02.png" style="width:100%;">
- **智能决策层**:包含感知系统和决策系统。感知系统获取环境资源使用数据,利用人工智能技术推理判断业务类型;决策系统根据感知子系统获得的数据进行调优决策。
- **系统画像层**:包含数据平台、标注系统和学习系统。数据平台用于存放软硬件数据;标注系统用于业务环境数据采集和业务模型的聚类;学习系统用于业务模型的学习和分类。
- **交互系统层**:用于各类系统资源的监控和配置,执行调优策略。
- **优化模式库**:构建并积累各种业务模型对应的优化能力和配置。
## A-Tune应用场景
当前,A-Tune根据不同的负载情况,在如下场景进行了测试
| **应用类型** | **已支持应用** |
| ------------------- | ----------------------------------- |
| https应用 | Nginx |
| 数据库 | MongoDB、MySQL、PostgreSQL、MariaDB |
| 大数据 | Hadoop、Spark |
| 内存密集型应用 | SPECjbb2015 |
| 计算+网络密集型应用 | Redis |
| 计算密集型应用 | SPECCPU2006 |
| 网络密集型应用 | Dubbo |
实验结果如下图所示:
<img src="/zh/blog/shanshishi/2020-05-08-atune-tuning-smarter-03.png" style="width:100%;" >
注:实验室测试数据,结果在不同环境存在偏差
## 关注A-Tune
当前A-Tune还在不断的开发和完善当中,如你对A-Tune感兴趣,欢迎关注A-Tune项目(扫描二维码,点击star),我们将在此发布开发路标并进行讨论,期待你的加入。
https://gitee.com/openeuler/A-Tune
<img src="/zh/blog/shanshishi/2020-05-08-atune-tuning-smarter-04.png" style="width:20%;" >
+++
title = "openEuler 跑到树莓派上"
date = "2020-05-25"
tags = ["Raspberry Pi", "summer2020", "操作系统移植"]
archives = "2020-05"
author = "woqidaideshi"
summary = "openEuler Raspberry Pi 版本镜像的构建、使用"
+++
# 概要
树莓派由英国“Raspberry Pi 基金会”开发,类似微型计算机,价格便宜,体积小,具有电脑的所有基本功能。自2012年以来,几乎每年都有新的版本发布,而且配置有了很大提升。树莓派从最初只做主板,到现在已经慢慢增加了周边的设备,包括显示器、摄像头以及一些传感器等。树莓派40针接口作为一个开放接口,提供了很大开发空间。作为开源硬件领域的一个较为高阶的硬件产品,树莓派近年来得到了越来越的开发者和专业人士的追捧。
为了降低 openEuler 使用门槛,促进用户推广,让大家尽快体验并使用 openEuler,本文将介绍如何将 openEuler 移植到树莓派。
本文主要包括三部分:
- [刷写镜像](#刷写镜像)
- [使用树莓派](#使用树莓派)
- [构建镜像](#构建镜像)
本文内容参考自 [openEuler-RaspberryPi](https://gitee.com/openeuler/raspberrypi)。如有疑问,欢迎大家在本文提交评论或者在 [openEuler-RaspberryPi](https://gitee.com/openeuler/raspberrypi) 提交 issue。
# 刷写镜像
## 实验环境
- Windows10/Linux/Mac
- 树莓派 3B/3B+/4B
- 16G 以上的 Micro SD 卡
## 获取树莓派 img 镜像
下载适配树莓派的最新 openEuler 镜像,也可以参照 [构建镜像](#构建镜像) 定制自己的 openEuler 树莓派镜像,下面以 `XXX.img` 表示该镜像。
当前最新的 openEuler 20.03 LTS 的内测版本镜像,[下载](https://isrc.iscas.ac.cn/EulixOS/repo/dailybuild/1/isos/20200508/openEuler_20200508151847.img.xz)
该镜像的基本信息:
- [更新日志](https://gitee.com/openeuler/raspberrypi/blob/master/documents/changelog.md)
- 发布时间:2020-05-11
- 大小:245 MiB
- 操作系统版本:openEuler 20.03 LTS
- 内核版本:4.19.90-2003.4.0.0036
- 固件来源:[firmware](https://github.com/raspberrypi/firmware)[bluez-firmware](https://github.com/RPi-Distro/bluez-firmware)[firmware-nonfree](https://github.com/RPi-Distro/firmware-nonfree)
- 构建文件系统的源仓库:[openEuler-20.03-LTS](http://repo.openeuler.org/openEuler-20.03-LTS/everything/aarch64/)
- 镜像内置源仓库:[openEuler 20.03 LTS 源仓库](https://gitee.com/openeuler/raspberrypi/blob/master/scripts/config/openEuler-20.03-LTS.repo)
发布的镜像更新情况参见 [openEuler-RaspberryPi](https://gitee.com/openeuler/raspberrypi/blob/master/README.md)
## 刷写 SD 卡
可以在 Windows、Linux、Mac 环境下的刷写 SD 卡,详见 [刷写镜像](https://gitee.com/openeuler/raspberrypi/blob/master/documents/刷写镜像.md)
下面我们以 Windows 环境为例,介绍如何将镜像刷写到 SD 卡。
### 格式化 SD 卡
下载应用:SDFormater,用于格式化 SD 卡。
若 SD 卡之前未安装过镜像,盘符正常只有一个,选择 SD 卡对应盘符,直接格式化即可;
若 SD 卡之前安装过镜像,盘符会有三个,选择 SD 卡对应盘符(图例中为 E: 盘),格式化即可:
<img src="/zh/blog/woqidaideshi/images/2020-05-25-raspberrypi-userguide-disk.jpg">
<img src="/zh/blog/woqidaideshi/images/2020-05-25-raspberrypi-userguide-disk-beforeformat.jpg">
<img src="/zh/blog/woqidaideshi/images/2020-05-25-raspberrypi-userguide-disk-afterformat.jpg">
<img src="/zh/blog/woqidaideshi/images/2020-05-25-raspberrypi-userguide-disk-after.jpg">
### 写入 SD 卡
下载应用:Win32 Disk Imager。
右键选择“以管理员身份运行”,打开 Win32 Disk Imager,选择镜像 img 文件和待写入的 SD 卡,点击【写入/Write】。
<img src="/zh/blog/woqidaideshi/images/2020-05-25-raspberrypi-userguide-writeSD.jpg">
写入完成即可。
写入成功后,SD 卡自动分区后的盘符为:
<img src="/zh/blog/woqidaideshi/images/2020-05-25-raspberrypi-userguide-disk.jpg">
# 使用树莓派
## 启用树莓派
版本要求:树莓派 3B/3B+/4B。
默认用户名:root,密码:openeuler。
将刷好的 SD 卡插入树莓派,通电启用。树莓派正常启动,还需连接网线至局域网。
由于使用树莓派时,大多都使用 ssh 远程连接;在树莓派启动联网时,无法得知其 IP 地址。
有以下两种方式:
1. 本地登录
将树莓派连接显示器(树莓派视频输出接口为 Micro HDMI)、键盘、鼠标后,启动树莓派,可以看到树莓派启动日志输出到显示器上。待树莓派启动成功,输入用户名(root)和密码(openeuler)登录。目前,openEuler 系统无桌面环境,添加桌面支持相关进度可关注 [期望增加桌面支持](https://gitee.com/openeuler/raspberrypi/issues/I1F21F),欢迎大家参与。
登录成功后,树莓派即显示本机相关信息,包括本机 IP。如下图(ssh 登录也显示这些信息):
<img src="/zh/blog/woqidaideshi/images/2020-05-25-raspberrypi-userguide-loginPi.jpg">
1. ssh 远程登录
如果树莓派连接已知路由器,可登录路由器管理,新增的 IP 即为树莓派 IP:
<img src="/zh/blog/woqidaideshi/images/2020-05-25-raspberrypi-userguide-getIP.jpg">
从上图看到,树莓派对应 IP 为:192.168.1.102,使用命令 `ssh root@192.168.1.102` 后输入密码 `openeuler`,即可远程登录树莓派。
## 根目录分区扩展
默认根目录分区空间比较小,在使用之前,需要对分区进行扩容。详情参见 [根目录分区扩展
](https://gitee.com/openeuler/raspberrypi/blob/master/documents/树莓派使用.md#根目录分区扩展)
## wifi 连接
wifi 使用详情参见 [wifi 连接
](https://gitee.com/openeuler/raspberrypi/blob/master/documents/树莓派使用.md#wifi-连接)
## 音频
音频使用详情参见 [音频
](https://gitee.com/openeuler/raspberrypi/blob/master/documents/树莓派使用.md#音频)
## 蓝牙
蓝牙使用详情参见 [蓝牙
](https://gitee.com/openeuler/raspberrypi/blob/master/documents/树莓派使用.md#蓝牙)
## GPIO
GPIO 使用详情参见 [GPIO
](https://gitee.com/openeuler/raspberrypi/blob/master/documents/树莓派使用.md#GPIO)
# 构建镜像
可以根据需要定制内核,构建镜像。详细内容请参考 [openEuler 镜像的构建](https://gitee.com/openeuler/raspberrypi/blob/master/documents/openEuler镜像的构建.md)
主要过程包括:
1. 准备环境
2. 编译内核
3. 准备树莓派固件和应用
4. 制作 openEuler 的 rootfs
5. 制作镜像
## 准备环境
- 操作系统:openEuler;
- 架构:AArch64;
- 硬盘存储不低于 50G;
- 内存不低于 2G;
- 可访问外网。
可以通过以下方式获取 AArch64 架构的运行环境:
- 使用 AArch64 架构的主机,例如刷写 openEuler 镜像的树莓派 3B/3B+/4B
- 使用 [QEMU](https://www.qemu.org/) 模拟器搭建 AArch64 运行环境
## 编译内核
除了使用 AArch64 架构的 openEuler 或 Centos 7/8 运行环境,也可以采用交叉编译的方式编译内核,文档详见 [交叉编译内核](https://gitee.com/openeuler/raspberrypi/blob/master/documents/交叉编译内核.md)
用户可以根据需要定制自己的内核,内核编译过程详见 [编译内核](https://gitee.com/openeuler/raspberrypi/blob/master/documents/openEuler镜像的构建.md#编译内核)
## 准备树莓派固件和应用
过程详见 [树莓派固件和应用](https://gitee.com/openeuler/raspberrypi/blob/master/documents/openEuler镜像的构建.md#树莓派固件和应用)
树莓派固件:
- [firmware](https://github.com/raspberrypi/firmware)
- [bluez-firmware](https://github.com/RPi-Distro/bluez-firmware)
- [firmware-nonfree](https://github.com/RPi-Distro/firmware-nonfree)
树莓派应用:
- [pi-bluetooth](https://github.com/RPi-Distro/pi-bluetooth)
## 制作 openEuler 的 rootfs
过程详见 [制作 openEuler 的 rootfs](https://gitee.com/openeuler/raspberrypi/blob/master/documents/openEuler镜像的构建.md#制作-openeuler-的-rootfs)
使用 rpm 命令制作 openEuler 的 rootfs,并为 rootfs 安装 NetworkManager、openssh-server、 passwd 等必要软件,设置网络等。
rootfs 的基本信息:
- 操作系统版本:openEuler 20.03 LTS
- 构建文件系统的源仓库:[openEuler-20.03-LTS](http://repo.openeuler.org/openEuler-20.03-LTS/everything/aarch64/)
- 镜像内置源仓库:[openEuler 20.03 LTS 源仓库](https://gitee.com/openeuler/raspberrypi/blob/master/scripts/config/openEuler-20.03-LTS.repo)
## 制作镜像
过程详见 [制作镜像](https://gitee.com/openeuler/raspberrypi/blob/master/documents/openEuler镜像的构建.md#制作镜像)
主要过程包括:
1. 创建镜像文件
1. 创建镜像后分区
2. 格式化分区后挂载其 boot 和根目录分区
2. 完善根目录分区,需要拷贝的文件
1. 制作的 rootfs
2. 编译的内核模块
3. 树莓派部分固件和应用
3. 完善 boot 分区,需要拷贝的文件
1. 树莓派引导文件
2. 编译的内核和设备树文件
注意,完善根目录分区时还需要根据实际情况设置 fstab 等内容,完善 boot 分区时需要根据实际需要设置 config.txt 等内容。
至此,openEuler 的树莓派镜像制作完成。发布的镜像更新情况参见 [openEuler-RaspberryPi](https://gitee.com/openeuler/raspberrypi/blob/master/README.md)
\ No newline at end of file
...@@ -16,6 +16,10 @@ id = "mails" ...@@ -16,6 +16,10 @@ id = "mails"
5. 回复您收到的电子邮件以确认您的订阅。 5. 回复您收到的电子邮件以确认您的订阅。
6. 最后您将收到来自一封来自邮件列表的欢迎邮件。 6. 最后您将收到来自一封来自邮件列表的欢迎邮件。
注:建议您在订阅前把邮箱的‘答复邮件上的邮件头使用英语’的相关设置打开:1)打开邮箱的选项界面。2)点击【高级】——并找到【国际选项】——勾选‘答复或转发邮件上的邮件头和转发通知使用英语’。
<img src="/img/mailimg.png" style="width:65%" >
{{< mails "列表名称" "邮件地址" "描述" >}} {{< mails "列表名称" "邮件地址" "描述" >}}
### 如何发送邮件到邮件列表? ### 如何发送邮件到邮件列表?
......
此差异已折叠。
+++
title = "【手把手带你玩转openEuler】系列直播进行中"
date = "2020-07-07"
tags = ["openEuler", "live"]
banner = "img/banners/banner-20200707-openeluer-live.png"
author = "openEuler"
summary = "【手把手带你玩转openEuler】系列直播进行中"
+++
***【******手把手带你玩转openEuler******】******系列直播进行中***
为了让更多人更深入认识openEuler并积极参与进来,社区将通过一系列直播和大家近距离接触,邀请openEuler开源的重要参与者、SIG组maintainer等资深专家来进行持续分享,从带你认识openEuler,到教你如何安装、如何使用openEuler,如何参与到openEuler社区贡献,都有很详细的解读。
本系列直播于7月份每***周四、周日晚2******0******:0******0***开播,欢迎来B站openEuler直播间和社区讲师互动!
***已******播出******:1、初识o******penEuler*** 讲师:朱延朋
​ 视频回顾链接:https://www.bilibili.com/video/BV1it4y197KQ
***2******、安装 o******penEuler*** 讲师:冯涛
​ 视频回顾链接:https://www.bilibili.com/video/BV1vK4y1s7QG
***七月后续直播计划:***
<img src="/img/events/20200707-openeluer-live-01.png" width="100%" style="max-width:583px"/>
\ No newline at end of file
+++
listname = "Crystal-ci"
emailaddress = "crystal-ci@openeuler.org"
description = "用于crystal-ci的项目研发。"
websitelink = "https://mailweb.openeuler.org/postorius/lists/crystal-ci.openeuler.org/"
archivelink = "https://mailweb.openeuler.org/hyperkitty/list/crystal-ci@openeuler.org"
weight = 15
+++
+++
listname = "QA"
emailaddress = "qa@openeuler.org"
description = "openEuler社区测试和质量保证讨论列表。"
websitelink = "https://mailweb.openeuler.org/postorius/lists/qa.openeuler.org/"
archivelink = "https://mailweb.openeuler.org/hyperkitty/list/qa@openeuler.org"
weight = 13
+++
+++
listname = "Sig-ai-bigdata"
emailaddress = "sig-ai-bigdata@openeuler.org"
description = "openEuler中sig-ai-bigdata邮件列表,讨论openeuler上ai和bigdata软件栈生态相关信息。"
websitelink = "https://mailweb.openeuler.org/postorius/lists/sig-ai-bigdata.openeuler.org/"
archivelink = "https://mailweb.openeuler.org/hyperkitty/list/sig-ai-bigdata@openeuler.org"
weight = 14
+++
...@@ -194,4 +194,4 @@ other = "ISO" ...@@ -194,4 +194,4 @@ other = "ISO"
other = "Installation Guide" other = "Installation Guide"
[main_cards_summer2020_link] [main_cards_summer2020_link]
other = "https://isrc.iscas.ac.cn/summer2020" other = "/zh/events/20200607.html"
\ No newline at end of file \ No newline at end of file
...@@ -193,5 +193,5 @@ other = "安装指南" ...@@ -193,5 +193,5 @@ other = "安装指南"
[main_cards_summer2020_link] [main_cards_summer2020_link]
other = "https://isrc.iscas.ac.cn/summer2020" other = "/zh/events/20200607.html"
...@@ -22,12 +22,12 @@ ...@@ -22,12 +22,12 @@
<div class="row"> <div class="row">
<div class="col-md-9"> <div class="col-md-8">
<section class="post"> <section class="post">
{{range .Data.Pages}} {{range .Data.Pages}}
<h2><a href="{{ .RelPermalink }}">{{ .Title }}</a></h2> <h3 style="font-family: 方正兰亭中黑;"><a href="{{ .Permalink }}">{{ .Title }}</a></h3>
<h5 class="author-category"> <h5 class="author-category">
{{ if isset .Params "author" }} {{ if isset .Params "author" }}
{{ .Params.author }} {{ .Params.author }}
...@@ -60,8 +60,8 @@ ...@@ -60,8 +60,8 @@
</div> </div>
<div class="col-md-3"> <div class="col-md-4">
<h3 style="overflow:hidden"><a href="/zh/blogguidance.html"><img src="{{ "/svg/edit.svg" }}" alt="edit" > <h3 style="overflow:hidden;font-family: 方正兰亭中黑;"><a href="/zh/blogguidance.html"><img src="{{ "/svg/edit.svg" }}" alt="edit" >
{{i18n "main_postablog_source"}}</a></h3> {{i18n "main_postablog_source"}}</a></h3>
{{partial "widgets/tags.html" . }} {{partial "widgets/tags.html" . }}
{{partial "widgets/archives.html" . }} {{partial "widgets/archives.html" . }}
......
此差异已折叠。
此差异已折叠。
此差异已折叠。
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册