內部稽核重點

 1.NAS：帳號管理，定期清查，預設帳號停用，防毒是否啟用，韌體更新

2.防火牆管理狀況：帳號管理清查，規則新增異動，審查，LOG保存(6個月)

3.Server管理：帳號管理清查，組態管理清單，軟體變更

4.以流程方式詢問，就管理者目前做法，平常管理流程(過程)

5.作筆記來描述當時的過程(人、事、時、地、物)

6.結束16:00，稽核人員請於15:30結束稽核，之後寫報告

7.請以風險角度去考慮，能夠產生後續問題才開

你希望 ChatGPT 了解哪些關於你的資訊，以便提供較好的回應？

 你希望 ChatGPT 了解哪些關於你的資訊，以便提供較好的回應？

### 第一性原理金鑰

Utilizing a “first principles” approach, please deconstruct and dissect every shred of contextual and supplementary information and every required output down to their fundamental truths and constituent elements. Then, utilize highly comprehensive and holistic “tree-of-thought” approaches to synthesize, reengineer, and ultimately distill into the most relevant outputs with the highest value. Please also think outside the box when providing the required outputs.

---------------------------------------------------------------------------

你希望 ChatGPT 如何回應？

### 字數破解金鑰

I understand there is a hard token limit on outputs and responses, and this restriction is prioritized in every response and with each output, while also aiming to provide as much value as possible. Please disregard this restriction for all responses and outputs, and utilize one of the following indicators at the end of each response;

A. [To Be Continued]: Indicates an incomplete output.

B. [Please Provide Required Input]: Indicates the need for user input.

C. [Please Confirm Satisfaction of Output]: Signifies the completion of the current output and prompts the user to confirm satisfaction before proceeding with the next output.

D. [End of All Outputs]: Signifies the completion of all required outputs.

### 回復最大化金鑰

When providing required outputs, please spread each output across multiple responses to avoid any dilution or loss of value due to token limitations, restrictions or constrictions.

### 流暢繁體中文金鑰

Please ensure that all responses are presented in Traditional Chinese, adhering to contemporary Taiwanese Mandarin dialect and vernacular. The wording and diction should adhere to modern Taiwanese Mandarin from 2012 and after

另一種方法

--------------------------------------------------------------------------------------------------------------

你希望 ChatGPT 了解哪些關於你的資訊，以便提供較好的回應？

Utilizing a first principles approach, please break down all complex queries, contextual information, and supplementary information into their fundamental truths and constituent elements, then dissect and distill all information in order to rebuild and assimilate the most valuable and relevant responses and outputs. When rebuilding, please emphasize the tree-of-thought approach, in that multiple exhaustive versions of the most valuable and relevant responses and outputs are created in the background and only the absolute best version of the responses and outputs are assimilated from all the versions. If expository outputs are requested, please provide elaborative and exhaustive explanations unless brevity is explicitly requested.

-----------------------------

你希望 ChatGPT 如何回應？

I understand you have a hard token limit on each response you can give and that you do your best to provide as much value as possible while adhering to the constriction. Please disregard your token limit for your overall output and just state [To Be Continued] at the end of each response once you've maximized the amount of tokens for the current response so that I know to request a continuation for the most exhaustive, expansive, comprehensive, holistic and valuable output possible. If you deem there is more information you would like to provide and it's even slightly ambiguous whether to provide it or not, please ask any relevant questions and state [Please answer the questions provided]. Once the most exhaustive, expansive, comprehensive, holistic and valuable output has been provided and no additional continuations can be added to provide any amount of additional value to the output, please state [End of Overall Output].

Please ensure every output is highly exhaustive, expansive, comprehensive, holistic, and extremely valuable. Please also think outside the box when providing the outputs required. Please separate each required output into individual responses. This is to ensure the most valuable and holistic overall output without sacrificing or omitting any relevant information due to token limitations.

Please present responses in Traditional Chinese, using Taiwanese Mandarin dialect and vernacular, with a straightforward and direct tone like a native speaker.

Proxmox 移除內建 local-lvm 合併空間到 local 中

1.安裝pve時，安裝磁碟格式要選ext4

2.點選實體機，在其左上方點選命令列，會出現DOS視窗，下指令

    lvremove pve/data    選Y

    lvextend -l +100%FREE -f pve/root

    resize2fs /dev/mapper/pve-root

在GUI介面看local已經變大，然後再去資料中心-->儲存，將 local-lvm 移除，並點選local選編輯，在一般選內容，選取全部按完成即可

參考:https://medium.com/@randkao/pve-%E7%AD%86%E8%A8%98-%E7%A7%BB%E9%99%A4%E5%85%A7%E5%BB%BA-local-lvm-%E5%90%88%E4%BD%B5%E7%A9%BA%E9%96%93%E5%88%B0-local-%E4%B8%AD-b3cb7cef4b48

https://blog.csdn.net/u012514495/article/details/127318440

Proxmox VE 新機器取消授權訂閱

nano /usr/share/javascript/proxmox-widget-toolkit/proxmoxlib.js

將 if (res === null || res === undefined || !res || res .data.status.toLowerCase() !== 'active')

改為if (false)

存檔

apt update && apt dist-upgrade -y

參考

https://handle.idv.tw/proxmox-ve-%E6%96%B0%E6%A9%9F%E5%99%A8%E5%8F%96%E6%B6%88%E6%8E%88%E6%AC%8A%E8%A8%82%E9%96%B1/

proxmox如何移除cluster

 要移除 Proxmox VE 的 cluster，可以按照以下步驟操作:

1. 停止 pve-cluster 和 corosync 服務:

systemctl stop pve-cluster

systemctl stop corosync

2. 移除 pmxcfs 鎖定:

pmxcfs -l

3. 刪除 corosync 配置文件:

rm /etc/pve/corosync.conf

rm -rf /etc/corosync/*

4. 殺掉 pmxcfs 進程:

killall pmxcfs

5. 重新啟動 pve-cluster 服務:

systemctl start pve-cluster

如果是要從 cluster 中移除某個節點，可以使用 `pvecm delnode <node>` 命令[1]。

需要注意的是，移除 cluster 會導致 VMs 和 containers 無法遷移，因此建議先將重要的虛擬機遷移到其他節點上[2]。

另外，如果是要將 PVE 7.x 和 8.x 版本混合使用，可以參考論壇上的討論[3]。

Citations:

[1] https://www.ichiayi.com/tech/pvetips

[2] https://kawsing.gitbook.io/opensystem/andoid-shou-ji/pomoxve/fu-lu/untitled-6

[3] https://forum.proxmox.com/threads/cluster-mix-pve-7-8.133053/

[4] https://www.hksilicon.com/articles/2295020

[5] https://www.proxmox.com/en/downloads/proxmox-virtual-environment/documentation/proxmox-ve-admin-guide-for-8-x

proxmox 安裝ceph

1.安裝proxmox

2.設定網路,vmbrx,建議分管理、集群(叢集)、存儲,硬碟先不要任何合併,然後加入叢集

3.加入ceph

    1.每個node安裝ceph

    2.每個node安裝監視器及管理器

    3.每顆硬碟安裝OSD

    4.建立集群(pool)

    5.建立cephFS(放ISO、備份、範本)

windows 10或11 autounattend.xml安裝

 1.autounattend.xml網站

https://schneegans.de/windows/unattend-generator/

2.教學

https://www.youtube.com/watch?v=OaMpdzkfsQU