Total Pageviews

Saturday, 1 June 2024

ET-BERT

 ET-BERT, a network traffic classification model on encrypted traffic. The work has been accepted as The Web Conference (WWW) 2022 accepted paper. 

ET-BERT is a method for learning datagram contextual relationships from encrypted traffic, which could be directly applied to different encrypted traffic scenarios and accurately identify classes of traffic. First, ET-BERT employs multi-layer attention in large scale unlabelled traffic to learn both inter-datagram contextual and inter-traffic transport relationships. Second, ET-BERT could be applied to a specific scenario to identify traffic types by fine-tuning the labeled encrypted traffic on a small scale.

The framework of ET-BERT

The work is introduced in the 31st The Web Conference:

Xinjie Lin, Gang Xiong, Gaopeng Gou, Zhen Li, Junzheng Shi and Jing Yu. 2022. ET-BERT: A Contextualized Datagram Representation with Pre-training Transformers for Encrypted Traffic Classification. In Proceedings of The Web Conference (WWW) 2022, Lyon, France. Association for Computing Machinery.

Note: this code is based on UER-py. Many thanks to the authors.

Table of Contents


Requirements

  • Python >= 3.6
  • CUDA: 11.4
  • GPU: Tesla V100S
  • torch >= 1.1
  • six >= 1.12.0
  • scapy == 2.4.4
  • numpy == 1.19.2
  • shutil, random, json, pickle, binascii, flowcontainer
  • argparse
  • packaging
  • tshark
  • SplitCap
  • scikit-learn
  • For the mixed precision training you will need apex from NVIDIA
  • For the pre-trained model conversion (related with TensorFlow) you will need TensorFlow
  • For the tokenization with wordpiece model you will need WordPiece
  • For the use of CRF in sequence labeling downstream task you will need pytorch-crf

Datasets

The real-world TLS 1.3 dataset is collected from March to July 2021 on China Science and Technology Network (CSTNET). For privacy considerations, we only release the anonymous data (see in CSTNET-TLS 1.3).

Other datasets we used for comparison experiments are publicly available, see the paper for more details. If you want to use your own data, please check if the data format is the same as datasets/cstnet-tls1.3/ and specify the data path in data_process/.


Using ET-BERT

You can now use ET-BERT directly through the pre-trained model or download via:

wget -O pretrained_model.bin https://drive.google.com/file/d/1r1yE34dU2W8zSqx1FkB8gCWri4DQWVtE/view?usp=sharing

After obtaining the pre-trained model, ET-BERT could be applied to the spetic task by fine-tuning at packet-level with labeled network traffic:

python3 fine-tuning/run_classifier.py --pretrained_model_path models/pre-trained_model.bin \
                                   --vocab_path models/encryptd_vocab.txt \
                                   --train_path datasets/cstnet-tls1.3/packet/train_dataset.tsv \
                                   --dev_path datasets/cstnet-tls1.3/packet/valid_dataset.tsv \
                                   --test_path datasets/cstnet-tls1.3/packet/test_dataset.tsv \
                                   --epochs_num 10 --batch_size 32 --embedding word_pos_seg \
                                   --encoder transformer --mask fully_visible \
                                   --seq_length 128 --learning_rate 2e-5

The default path of the fine-tuned classifier model is models/finetuned_model.bin. Then you can do inference with the fine-tuned model:

python3 inference/run_classifier_infer.py --load_model_path models/finetuned_model.bin \
                                          --vocab_path models/encryptd_vocab.txt \
                                          --test_path datasets/cstnet-tls1.3/packet/nolabel_test_dataset.tsv \
                                          --prediction_path datasets/cstnet-tls1.3/packet/prediction.tsv \
                                          --labels_num 120 \
                                          --embedding word_pos_seg --encoder transformer --mask fully_visible

Reproduce ET-BERT

Pre-process

To reproduce the steps necessary to pre-train ET-BERT on network traffic data, follow the following steps:

  1. Run vocab_process/main.py to generate the encrypted traffic corpus or directly use the generated corpus in corpora/. Note you'll need to change the file paths and some configures at the top of the file.
  2. Run main/preprocess.py to pre-process the encrypted traffic burst corpus.
       python3 preprocess.py --corpus_path corpora/encrypted_traffic_burst.txt \
                             --vocab_path models/encryptd_vocab.txt \
                             --dataset_path dataset.pt --processes_num 8 --target bert
    
  1. Run data_process/main.py to generate the data for downstream tasks if there is a dataset in pcap format that needs to be processed. This process includes two steps. The first is to split pcap files by setting splitcap=True in datasets/main.py:54 and save as npy datasets. Then the second is to generate the fine-tuning data. If you use the shared datasets, then you need to create a folder under the dataset_save_path named dataset and copy the datasets here.

Pre-training

To reproduce the steps necessary to finetune ET-BERT on labeled data, run pretrain.py to pre-train.

   python3 pre-training/pretrain.py --dataset_path dataset.pt --vocab_path models/encryptd_vocab.txt \
                       --output_model_path models/pre-trained_model.bin \
                       --world_size 8 --gpu_ranks 0 1 2 3 4 5 6 7 \
                       --total_steps 500000 --save_checkpoint_steps 10000 --batch_size 32 \
                       --embedding word_pos_seg --encoder transformer --mask fully_visible --target bert

Fine-tuning on downstream tasks

To see an example of how to use ET-BERT for the encrypted traffic classification tasks, go to the Using ET-BERT and run_classifier.py script in the fine-tuning folder.

Note: you'll need to change the path in programes. 

from https://github.com/linwhitehat/ET-BERT

 

logto

 The better identity infrastructure for developers and the open-source alternative to Auth0.

logto.io

discord checks release core coverage cloud gitpod render

Logto1 is an Auth0 alternative designed for modern apps and SaaS products. It offers a seamless developer experience and is well-suited for individuals and growing companies.

🧑‍💻 Comprehensive frontend-to-backend identity solution

  • Enables OIDC-based authentication with Logto SDKs.
  • Supports passwordless sign-in, along with various options like email, phone number, username, Google, Facebook, and other social sign-in methods.
  • Offers beautiful UI components with customizable CSS to suit your business needs.

📦 Out-of-the-box infrastructure

  • Includes a ready-to-use Management API, serving as your authentication provider, thus eliminating the need for extra implementation.
  • Provides SDKs that seamlessly integrate your apps with Logto across multiple platforms and languages, tailored to your development environment.
  • Offers flexible connectors that can be scaled with community contributions and customized with SAML, OAuth, and OIDC protocols.

💻 Enterprise-ready solutions

  • Implements role-based access control (RBAC) for scalable role authorization, catering to a wide range of use cases.
  • Facilitates user management and provides audit logs for understanding identity-related user information and maintaining security.
  • Enables single sign-on (SSO) and multi-factor authentication (MFA) without extra coding.
  • Leverages Logto Organizations to build multi-tenancy apps with ease.

In a more approachable way, we refer to this solution as "Customer Identity Access Management (CIAM)" or simply, the "Customer Identity Solution."

Subscribe to us now to stay updated with the latest information about the Logto Cloud (SaaS) and receive feature updates in real-time.

Get started

  • Visit our 🎨 website for a brief introduction if you are new to Logto.
  • A step-by-step guide is available on 📖 docs.logto.io.

Interactive demo

  • Try Logto Cloud to have the same dev experience and zero deployment overhead.

  • If you launch Logto via GitPod, please wait until you see the message like App is running at https://3002-...gitpod.io in the terminal, press Cmd (or Ctrl on Windows) and click the URL starts with https://3002- to continue your Logto journey.

Launch Logto

Docker Compose

Docker Compose CLI usually comes with Docker Desktop.

curl -fsSL https://raw.githubusercontent.com/logto-io/logto/HEAD/docker-compose.yml | \
docker compose -p logto -f - up

npm-init

Requires Node.js ^20.9.0 + PostgreSQL ^14.0.

npm init @logto
from https://github.com/logto-io/logto
---------------------------------------------------

(self-hosted)

GitPod

To start an online GitPod workspace for Logto, click here. Wait a few moment, you'll see the message like:

Gitpod is running

Logto uses port 3001 for its core service and port 3002 for the interactive Admin Console by default.

To continue your Logto journey, press Ctrl (or Cmd) and click the link that starts with https://3002-.... Enjoy!

Local

Prerequisites

Higher versions usually work but are not guaranteed.

We recommend using a new empty database which is dedicated for Logto, while it's not a hard requirement.

Download and start

In your terminal:

npm init @logto@latest

Once you complete the init process and start Logto, you will see the message like:

Core app is running at http://localhost:3001
Core app is running at https://your-domain-url
Admin app is running at http://localhost:3002
Admin app is running at https://your-admin-domain-url

Heading to http://localhost:3002/ to continue your Logto journey. Enjoy!

from https://docs.logto.io/docs/tutorials/get-started/

 
 

习近平在台湾立法院占据桥头堡了吗?


 VOA 20240529

台湾立法院在通过国会改革法案后,支持执政党民进党的民众聚集立法院外,放出“拒绝中国干政”的气球。(2024年5月28日)


美国之音 黄丽玲 台北 - 台湾立法院星期二(5月28日)三读通过五大职权修法,但扩权争议引发的朝野对峙和公民抗争才刚揭开序幕。除了内部政争效应外,台湾舆论也聚焦外来的“中国因素”,尤其关注中国流亡作家袁红冰早自2月就提出的警告:“习近平要在台湾立法院占领制高点”或桥头堡,从内部分化台湾社会,为武统铺路。这样的论述虽不易核实,但近日引发立院场外抗议群众的热议却是不争的事实。

台湾总统赖清德上任第一周,两大在野阵营国民党(蓝)和民众党(白)就急着联手通过五大职权修法,是实现国会改革正义的时机凑巧?还是有意制约赖政府的政争手段?背后有无中共的介入?在台湾抗议群众中引发热议。

今年2月初,袁红冰向海外网媒《看中国》揭露、中国红二代泄漏给他的国台办密件《占领台湾立法院制高点的统战战略要点》尤其引发台湾公民团体的关注,不少人热议,台湾立院局势似乎正照着这个剧本走,因为蓝白合在立院扩权后,不仅凌驾行政权之上,也凌驾司法调查权等之上,恐让北京大开渗透的方便之门。

对此,在澳大利亚的袁红冰在接受美国之音采访时表示,台湾立法院在国民党的主导下,打着国会改革的旗号,实质扩权,都是照着中共统战部的意志在行事,因为习近平已指示对台单位要抓紧蓝白合在立院占多数的机遇,以发挥对台政治效应。

台湾立院扩权给赖政府戴上脚镣和手铐

袁红冰说:“这个政治效应主要表现在立法院的扩权上,目的有两个:一是通过立法院扩权实际上给赖清德的行政权戴上脚镣和手铐,这是国台办内部的一个指示性意见。”

至于第二个目的,袁红冰说:“通过扩权让立法权凌驾在台湾的司法权之上,来抵消绿营对司法权的掌控和影响,然后下一步,就是要废除《反渗透法》。”

他说,台湾2019年通过的《反渗透法》阻断了中共和台湾内部通共分子的管道,因此,若此法废除,中共才能重新打开对台渗透的通道。

袁红冰说,中共政权的运作都带有政治目的,绝不可能为了套交情,就和国民党人往来。他认为,中共近期接连会晤夏立言、马英九,到傅崐萁率领的17人访团,都是收买和分化台湾社会的迹象。

他还说,中共体制内的良知人士向他透露,武统台湾是习近平最重要的战略想定之一,但军事入侵台湾前,中共得清扫两大障碍,一是透过在台收买的协作者,散布恐共/媚共情绪,以瓦解台湾自卫的战斗意志;二是利用蓝白合在立院的多数席位,让中共“在台湾权力结构中占领一个桥头堡,用习近平的话讲,叫做在台湾立法院占领制高点”。

袁红冰呼吁,台湾社会对中共的分化企图提高警觉,虽然他也对台湾公民社会辨别是非的力量深具信心。

国民党斥袁红冰是“造谣专家”

对于袁红冰揭露的说法,国民党发言人杨智伃5月24日接受记者采访时批评袁红冰是“造谣专家”。她质疑,袁红冰2009年出版《台湾大劫难》一书时就直指中共收买民进党新潮流派系,若按此逻辑,出身新潮流派系的赖清德早就被中共收买,恐更加证明台湾需要国会改革,通过立法院监督行政权。

民众党主席柯文哲则分别于5月22ㄖ和27日透过自己的油管节目《柯P揪时在》澄清该党的改革初衷,并一一破解民进党的所谓“谣言”。他说,中国虽然时常对台湾发动认知战,但他发现,民进党在此次修法过程中也制造了很多假新闻,才是造谣,试图改变台湾年轻人的认知。

柯文哲说,这次修法的内容多是民进党过去做不到的主张,包括总统国会报告,接受即问即答是民进党早于2012年就提出的。另外,针对黑箱指控,他说,立院审议前,已进行过逾1500个小时的公听会等讨论,绝无黑箱作业等。

他还呼吁民进党与其侧翼不要抹黑抹红,将不同意见的人全部打成“中共的同路人”。

针对这份中国国台办密件的真伪,美国之音三度致电中国国台办新闻局正副局长的办公室电话,希望核实或取得置评,但两通电话无人接听,另一通则一接听后,随即遭挂断。

支持执政党民进党的民众聚集台湾立法院外,举着“昨日香港,今日台湾”的标语牌。(2024年5月28日)
支持执政党民进党的民众聚集台湾立法院外,举着“昨日香港,今日台湾”的标语牌。(2024年5月28日)

台湾抗议者:中国介入乃合理推测

虽然国台办密件的真实性难辨,但好几位在立法院外抗议的公民和年轻人都说,蓝白合联手快速修法,有中国因素介入应是合理推测。

在台北的政治大学大二生游博安告诉美国之音:“我不会那么的直接去连接到说,这个是中国的指示。但国民党一直以来就被认为是一个相对对中国比较友善的政党,刚好是太阳花(公民抗争运动)10周年,我觉得大家有这个联想很合理......大家认为,他们是所谓中共同路人,这是个很合理的推测。”

游博安说,尤其国民党立院党团总召傅崐萁近期才刚访中,也难免让人有更深的联想。但他对民众党遭中共收买持保留意见,因为民众党是小党,立场比较模糊,可能视议题,与任一大党合作都有可能。

但原是小党时代力量和黄国昌支持者的游博安说,他和许多人对黄国昌此次倒向国民党的表现非常失望,因为黄国昌10年前打弊案时还批评“国民党连当在野党都不配”,对他如今游走各党的投机行为不以为然。

原为柯的支持者、在立法院外抗议的梁先生也说,民众党这次推动的修法是脱序行为,违宪外,也恐被中国所利用。

梁先生告诉美国之音:“我更期待你(柯文哲)是在遵守中华民国台湾的宪法架构下来进行国会改革,而不是造成国会乱象,甚至极可能在这过程中被我们敌对势力中国政府所利用,来迫害我们的国家。”

来自中台湾嘉义的郭女士则支持袁红冰的观察,她说,因为520前后立法院就出现黑箱法案,不利台湾民主,又适逢傅崐萁才刚访中回来,中共的角色难免启人疑窦。

郭女士告诉美国之音:“台湾跟中国,一边一国,互不隶属,请中国不要干涉台湾的内政。”

黄兆年:蓝白是中国统战的吸纳对象

在台北的政治大学国家发展研究所副教授黄兆年认为,此次立委扩权修法中的中国因素确实值得台湾警戒。

他说,台湾立院扩权后,让立委有调查权,也有权取得政府、军队、民企、公民团体的机敏资料,甚至一般人的隐私,若中共进一步获得蓝白立委的配合,确实与其现阶段的战略目标契合。

不过,他说,国民党和民众党是否配合中共对台渗透,尚待观察,也未有确切证据。

但中共从对台统战的战略目标和手段上,的确乐见蓝白联手扩权。

黄兆年说,中国不放弃武统台湾,因此,台湾立院扩权若反成泄漏台湾军事、国安情报的管道,有助于中共的情蒐。另外,中国面对经济下行也致力推动产业升级,因此早就透过并购、商业间谍或骇客在各国窃取得高科技技术,来提升科技和经济实力,若台湾立院也是取得高科技业相关研发情资的管道之一,自然为北京开方便之门。

至于从手段上,他说,中共统战思维一直立基在联合次要敌人,打击主要敌人,因此,国民党和民众党都是其统战吸纳的对象,以对付民进党,达到反独促统的目标。尤其国民党失去丰沛的党政资源可下放给地方势力,亦即其过去与地方的“恩庇主”和侍从角色已不再,这让北京有机可趁,绕过台湾的中央政府,甚至取代国民党的“恩庇主”角色,以农业或观光等经济利益换取台湾地方势力的政治妥协,甚至绑架国民党,走向更亲中的倾向。

至于民众党,黄兆年说,虽和北京有较少的利益纠葛,但蓝白联手诉诸民粹、民主表象推法,可扩大其权力基础。

黄兆年告诉美国之音:“这三方一拍即合,就是说,国民党目前是被地方派系所绑架的一个政党,然后北京,再加上民粹政客所组成的白色阵营,所以所谓的国会改革法案,变成是中共统一战线工程的一环,是有这个可能性。”

台灣人的危機感出來了,團結對抗中共的代理人


 李志銘  / 思想坦克 


回顧台灣近代整個民主化過程,舉凡自1990年推動國會全面改選的「野百合學運」、2013年洪仲丘事件引發的白衫軍運動、2014年反服貿黑箱的「太陽花學運」、2019年「拒絕紅色媒體,守護台灣民主」大遊行,乃至2024年524反國會擴權、黑箱審查法案的「青鳥行動」,其中幾乎所有妨礙改革的保守阻力都來自中國國民黨,這段期間當然還包含了中共持續不斷的文攻武嚇及在國際上的打壓。

遺憾的是,在(2016年)馬英九政府下台後,當全世界都跟你說中共是敵人的時候,唯有國民黨明顯從過去的「反共」轉變為「親共」,並與極權中共控制的官媒風向及立場趨於一致,甚至和對岸「同聲一氣」不斷威嚇台灣人民,漸漸與民主世界的價值觀形成對立。

據我所知,這次大選有不少民眾為了政黨平衡、避免一黨獨大,於是採取「分裂投票」:也就是說總統票投民進黨,政黨票及區域立委票則是投給了國民黨或民眾黨,至於那些總統票投給柯的年輕小草其實也都是對台灣未來的民主政治有所寄望。沒想到國民黨在立法院取得多數(席位)權力、乃至於該黨總召傅崐萁不顧花蓮震災,執意帶立委團去中國「進京面聖」回來之後,隨即以黑箱闖關強推「藐視國會罪」等國會擴權法案。

種種倒行逆施的行徑,儼然印證了中研院學者吳叡人5月24、28兩天在立法院外的街頭演講所陳述:北京的焦慮、國民黨對久違已久的權力的饑渴,還有新興民粹政客的野心,三者醜陋的慾望,共同交織出了這一齣一邊演一邊寫的劇本的立法權政變的爛戲醜劇。


與此同時,再加上本地媒體不斷廣傳袁紅冰日前披露中共今年2月早已預言要「全面搞亂滲透台灣,戰場就在立法院」的報導文章,無疑也更加深了過去有不少缺乏「敵我意識」的台灣人、這次終於也能直接感受到一股切身的「危機感」,才因此激起十萬人在5月24這天走上街頭抗爭的強烈意志。


「去核心化」的街頭民主教室,遍地開花的公民肥皂箱

相較於十年前的318「太陽花學運」,十年後的521、524、528「青鳥行動」毋寧更像是進化版的「太陽花2.0」。舉例來說,十年前那些曾經參與「太陽花」佔領立法院的大學生和高中生們,後來都已經長大成為中年上班族或學校老師,如今卻又再度回到了街上,不僅跟許多新生代的年輕人一同遊行抗爭,同時也因為這些「中生代」參與者本身經濟能力的成長,以至於紛紛出現了過去「太陽花時代」前所未見─能夠提供相關物資、交通費、住宿費用等後勤支援的「課金組」角色,讓那些經濟條件較為匱乏的年輕人或是中南部的青年學子也能來台北參加陳抗表達立場。

另外,或許是受到了2019年香港「反送中」運動採取「Be Water」策略及核心精神所影響(如水聚散,意指不被形式所拘,要像水一樣流動,既柔軟又剛強,既能適應萬物,又能匯聚為強大的力量),這次「青鳥行動」亦不再有明顯的某個學運明星或社會菁英出來作為領頭代表,而是更具有一種「去核心化」、「人人平等」、「不分你我」共同參與的庶民特質,比如有許多民眾便曾在現場看到網紅名人鄭家純低調地出現在人群中協助人流管制,甚至等到大家慢慢散場後一個人默默地幫忙收拾椅子。

除青島東路以外,其他鄰近中山南路和濟南路各地皆設有多處「公民肥皂箱」,任何一個民眾都可以上台抒發己見,台下幾個朋友圍成一圈席地而坐就可以開始彼此討論。現場不僅有高中生以生澀的口吻念著手中的講稿,也有初次上街頭的大學生靦腆地講述自己為什麼要來現場,若是走累了、口渴了就到附近的濟南教會喝水吃粥休息一番。舉目所見,整個街道就像是一個大型民主教室,彷彿有種「遍地開花」的嘉年華氛圍,接近傍晚時甚至還能看到台灣早期(80年代)黨外街頭運動時期著名的「民主戰車」重現江湖!

當然最引人注目的,無論是521、524或528的抗議現場,都能明顯感受到大多數民眾對於「背骨仔」黃國昌的一股強烈憤怒!簡直成了全台灣「仇恨值最高」的政治人物,街頭上幾乎隨處可見民眾各種痛罵、嘲諷黃國昌的標語手板及宣傳海報,不僅立法院外眾聲怒斥「黃國昌偽君子」、「花蓮王的『萁娃娃』」,亦有民眾製作舉牌「豬哥亮沒有死,死的是2014的黃國昌」,旁邊一個大概十歲的小朋友拿著手板標語「我有欠你嗎?」(黃國昌的口頭禪),還有人在街上到處散發他的「遺照」,現場一堆流動小型靈堂,前面貼著「昌容宛在」、「國會垃圾」、「敬悼國會最噁爛的臉龐」等紙條。


網路上則是迅速瘋傳公民自製的「蔥師表」大酸黃國昌!舉凡歌劇、佛經、聖歌、布袋戲配音、搖滾、鋼琴等各種版本不斷推陳出新,另也有花蓮民眾自編「傅王賦」怒批傅崑萁,意外掀起了全民二創風潮。

危機便是轉機:持續堅持的「民主防衛戰」與公民運動

當你走上街頭,赫然發覺每個人似乎都變成了文案高手,在國家有難時,發揮無限智慧與創意來表達抗議,吶喊出內心的悲憤及不滿。這段期間,雖然是讓許多台灣人感到憤怒的時刻,卻也是最療癒的時刻。

台灣的民主不是天下掉下來的,是每一世代勇敢的台灣人爭取來的!

從解嚴前到現在,台灣人已經在街上捍衛自由民主快40年了,須知這一切成果皆來之不易。如今面臨中共惡鄰長期不斷透過武力威脅,以及「國共滲透台灣,藍白裡應外合」之下,若想要擁有獨立自主的生活,就必須有一代又一代的人民願意勇敢站出來反抗。

儘管全台灣各地都有抗議、立法院前面集結數十萬人,同時還有30名來自各國的專家學者在美國著名的《外交家》雜誌(The Diplomat)發表連署信,以「國際學者對台灣的立法改革敲響警鐘」(International Scholars Sound the Alarm Over Legislative Reforms Proposed in Taiwan)為標題,反對台灣在野黨強行通過「國會改革法案」。然而令人遺憾的是,在國眾兩黨完全無視於民意的強制動員下,這樁違反民主程序、侵害人民權力的「國會擴權法案」終究還是在528這天全案表決(三讀)通過了。

現今抗爭尚未結束,接下來的「民主防衛戰」還有很多仗要打。星星之火,燎灼於島嶼各處。此後的應對之道,即是由台灣社會的公民力量開始嘗試推動各種多元化、彈性化的長期抗爭模式和重點議題,比如行政權覆議、大法官釋憲、公投複決、罷免立委等,設法讓「青鳥行動」的社會能量能夠延續至年底可以進行罷免立委的時間點。

每當感到絕望和困頓的時候,腦海中總是不禁想起了早年那些黨外民主前輩經常傳唱的台灣民謠《補破網》:「今日若將這來放,是永遠無希望,為著前途弄破網,找傢私補破網」,說到激動處、「公民肥皂箱」講台上的吳叡人忍不住哽咽道出香港2019年反送中運動抗爭名言:「不是因希望才堅持,而是堅持才有希望」。

所谓"兑现承诺"

 在古老东方帝国泱泱数千年的文化巨典里,有孙子兵法,有三十六计,有出其不意,有韬光养晦,有老奸巨猾,有能屈能伸,有见风使舵,有审时度势,有阳奉阴违,有欲盖弥彰,有暗渡陈仓,有瞒天过海,有偷天换日,有尔虞我诈,有口蜜腹剑,有佛口蛇心,有虚与委蛇,有首鼠两端,有笑里藏刀,有两面三刀,有厚黑学,有两头吃,有墙头草,……,就没见过一次"兑现承诺"这一条。


不但我们土著没有见过,洋人同样没有见过。如果有谁见过一次"兑现承诺",那么,不说远了,只说近的,最近三百年来的所有外来战争都会归零。

譬如两次鸦片战争、中法战争、甲午战争、庚子战争、和最近的日中战争,等等。即便被揍得鼻青脸肿,依然在东方文化巨典里看不见"兑现承诺",巨典的每一页都只写着"落后就要挨打"。

虽没有"兑现",但"承诺"还是有的。不但有承诺,而且是天下第一。譬如不拿一针一线、吃苦在前享乐在后、先富带动后富、半条裤子,尤其写有"人民"二字的地方,全都是满满的承诺,不但看得见,而且刷在墙上、刻在大理石上、挂在庄严巍峨的正门上,谁看谁都会热泪盈眶。

前几天我看见一个视频,一个老奶奶推着轮椅,轮椅上是她的老伴,老奶奶将轮椅推到了江边,不知道是一条什么江。两位老人相互拥抱了一下还说了几句什么,老奶奶用力将轮椅推到了江水里,接着老奶奶顺势也扑到了轮椅上,两位老人瞬息就消失在微波荡漾的江水中。我紧紧盯着视频中的江面,朝远方看,再朝远方看,我看见了江对面的山坡上有一处没有长草的石岩,石岩上面还斑驳地残存着石灰水刷写的承诺"***个好,***养老"。那一刻,我的眼角不知不觉就热泪盈眶。

活在今天,你必须是一个非常优秀的机会主义者

陈丹青先生所言的勇者:“活在今天,你必须是一个非常优秀的机会主义者,你得足够无耻,足够勇敢,我不带一丝丝贬义。”“这块土地上的人的唯一信仰就是,去他妈的,活下去再说。”

写到这里,我不得不,向每一位年轻的跳者深深地鞠一躬。之后,还是久久地沉默,久久地发呆,久久地听着窗外鸟语花香。我不想说什么,也说不出什么。
去年四月,四个年轻人在天门山绝尘而去时,我写了一篇《我看见他们转身离去,任凭眼泪在雨中纷飞》。去年七月,三个年轻人在广东出租屋烧炭而去时,我写了一篇《人生这出悲剧,不是人人能演完》。今年,我看见一个接着一个的跳者,我连眼泪也没有了。甚至,我都不认为每一个跳者是一种悲剧。因为古希腊哲学家伊比鸠鲁早在2300多年前就说“死,不是死者的不幸,而是生者的不幸。”
也许,我们大多数,也许,是绝大多数,不,应该是所有,都认为那些跳河跳江的年轻人是因为穷得活不下去了,所以不得不跳了。
但我,可以斩钉截铁地说,他们不是。
陈丹青先生讲得很清楚,想活下去太简单了,只要足够无耻,如果不足够,那就再勇敢地无耻。像我们绵延两千多年的老祖宗一样,亲亲相食、易子而烹、大义灭亲、埋儿奉母、吃观音土、把村里的树皮草根全拔光、挨村挨户乞讨,……,看,我们一代又一代的老祖宗不是活下来了么,并且都还生了一大窝一大窝的地活下来了。在今天想活下去,远比我们老祖宗更简单,朝阳、枫桥、网评员、跟踪异见人士、拦截告状者、踢太婆摊子、……,都需要大量的人,尤其需要年轻人,怎么都能混一碗饭吃。
在今天,如果一个年轻的女孩子想活下去,远比男孩子更简单,更简单,更简单。
4月19日,镇江丹阳大桥上那个16岁的姑娘,妈妈告诉她“没有钱就出去打工”。她最终没有听妈妈的话,而是回头一瞥最后的人间,16岁的她毅然决然选择与无情的江水流向远方。如果,如果,如果她真的出去打工,她一定能活下去。今天连985名校生和众多海归生都在美团抢单度日,一个16岁的姑娘出去打工又是一条什么样的活路呢?我不敢想,我也不愿想。我不敢说,我也不愿说。所以,我摘取法国象征派诗人兰波《致奥菲利亚》中的句子向那位16岁姑娘沉重道别:“我永恒的灵魂注视着你的心,纵然,黑夜孤寂,白昼如焚。如果我没有记错,你的生命原本是一场盛宴!去吧、在那里,所有的心灵全都敞开,所有的美酒纷纷溢出来。”
致敬吧,我们每一位活着的,都应该向今天的年轻人深深鞠一躬。他们,不是穷的活不下去才跳跳跳。他们,只是不想再像我们绵延两千多年的老祖宗那样“顽强”地活下去,他们不愿像牲口一样地活下去。他们都还是一个个孩子,但他们比我们和我们老祖宗都门儿清,他们活着就必须养一帮孙子。所以,他们跳了。他们跳下去的是一枚枚柔弱的肉身,但他们留下的是一颗颗坚硬的灵魂。

NetBox

 

The premier source of truth powering network automation. Open source under Apache 2. Public demo: https://demo.netbox.dev

netboxlabs.com/oss/netbox/ 

The cornerstone of every automated network

Latest release License Contributors GitHub stars Languages supported CI status

NetBox exists to empower network engineers. Since its release in 2016, it has become the go-to solution for modeling and documenting network infrastructure for thousands of organizations worldwide. As a successor to legacy IPAM and DCIM applications, NetBox provides a cohesive, extensive, and accessible data model for all things networked. By providing a single robust user interface and programmable APIs for everything from cable maps to device configurations, NetBox serves as the central source of truth for the modern network.

NetBox's Role | Why NetBox? | Getting Started | Get Involved | Project Stats | Screenshots

NetBox user interface screenshot

NetBox's Role

NetBox functions as the source of truth for your network infrastructure. Its job is to define and validate the intended state of all network components and resources. NetBox does not interact with network nodes directly; rather, it makes this data available programmatically to purpose-built automation, monitoring, and assurance tools. This separation of duties enables the construction of a robust yet flexible automation system.

Reference network automation architecture

The diagram above illustrates the recommended deployment architecture for an automated network, leveraging NetBox as the central authority for network state. This approach allows your team to swap out individual tools to meet changing needs while retaining a predictable, modular workflow.

Why NetBox?

Comprehensive Data Model

Racks, devices, cables, IP addresses, VLANs, circuits, power, VPNs, and lots more: NetBox is built for networks. Its comprehensive and thoroughly inter-linked data model provides for natural and highly structured modeling of myriad network primitives that just isn't possible using general-purpose tools. And there's no need to waste time contemplating how to build out a database: Everything is ready to go upon installation.

Focused Development

NetBox strives to meet a singular goal: Provide the best available solution for making network infrastructure programmatically accessible. Unlike "all-in-one" tools which awkwardly bolt on half-baked features in an attempt to check every box, NetBox is committed to its core function. NetBox provides the best possible solution for modeling network infrastructure, and provides rich APIs for integrating with tools that excel in other areas of network automation.

Extensible and Customizable

No two networks are exactly the same. Users are empowered to extend NetBox's native data model with custom fields and tags to best suit their unique needs. You can even write your own plugins to introduce entirely new objects and functionality!

Flexible Permissions

NetBox includes a fully customizable permission system, which affords administrators incredible granularity when assigning roles to users and groups. Want to restrict certain users to working only with cabling and not be able to change IP addresses? Or maybe each team should have access only to a particular tenant? NetBox enables you to craft roles as you see fit.

Custom Validation & Protection Rules

The data you put into NetBox is crucial to network operations. In addition to its robust native validation rules, NetBox provides mechanisms for administrators to define their own custom validation rules for objects. Custom validation can be used both to ensure new or modified objects adhere to a set of rules, and to prevent the deletion of objects which don't meet certain criteria. (For example, you might want to prevent the deletion of a device with an "active" status.)

Device Configuration Rendering

NetBox can render user-created Jinja2 templates to generate device configurations from its own data. Configuration templates can be uploaded individually or pulled automatically from an external source, such as a git repository. Rendered configurations can be retrieved via the REST API for application directly to network devices via a provisioning tool such as Ansible or Salt.

Custom Scripts

Complex workflows, such as provisioning a new branch office, can be tedious to carry out via the user interface. NetBox allows you to write and upload custom scripts that can be run directly from the UI. Scripts prompt users for input and then automate the necessary tasks to greatly simplify otherwise burdensome processes.

Automated Events

Users can define event rules to automatically trigger a custom script or outbound webhook in response to a NetBox event. For example, you might want to automatically update a network monitoring service whenever a new device is added to NetBox, or update a DHCP server when an IP range is allocated.

Comprehensive Change Logging

NetBox automatically logs the creation, modification, and deletion of all managed objects, providing a thorough change history. Changes can be attributed to the executing user, and related changes are grouped automatically by request ID.

Note

A complete list of NetBox's myriad features can be found in the introductory documentation.

Getting Started

NetBox Cloud
Looking for a managed solution? Check out NetBox Cloud or NetBox Enterprise!

Get Involved

from  https://github.com/netbox-community/netbox

----- 

Installation

NetBox Cloud

The instructions below are for installing NetBox as a standalone, self-hosted application. For a Cloud-delivered solution, check out NetBox Cloud by NetBox Labs.

The installation instructions provided here have been tested to work on Ubuntu 22.04 and CentOS 8.3. The particular commands needed to install dependencies on other distributions may vary significantly. Unfortunately, this is outside the control of the NetBox maintainers. Please consult your distribution's documentation for assistance with any errors.

The following sections detail how to set up a new instance of NetBox:

  1. PostgreSQL database
  2. Redis
  3. NetBox components
  4. Gunicorn or uWSGI
  5. HTTP server
  6. LDAP authentication (optional)

Requirements

Dependency Supported Versions
Python 3.10, 3.11, 3.12
PostgreSQL 12+
Redis 4.0+

Below is a simplified overview of the NetBox application stack for reference:

NetBox UI as seen by a non-authenticated user

Upgrading

If you are upgrading from an existing installation, please consult the upgrading guide.

from https://netboxlabs.com/docs/netbox/en/stable/installation/

https://netboxlabs.com/docs/netbox/en/stable/