> For the complete documentation index, see [llms.txt](https://advwacloud.gitbook.io/rd-newbie-guide/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://advwacloud.gitbook.io/rd-newbie-guide/kai-fa-huan-jing/k8s_datahub_dev.md).

# \[開發階段使用] datahub k8s deploy

## Intro

* 介紹如何使用helm chart來部署datahub到k8s
  * chart: 包含了k8s各個資源的template(deployment template/service template/ingress template...), 部署時會配合values.yaml將值注入template (values.yaml也會有template)
  * helm: 部署/更新/查詢 chart
  * advantage: 參數化部屬/一次部署全部相關資源/一次刪除全部相關資源
* 注意: 本篇chart username/password我都挖掉了, 要自己填上去

## Prerequisite

* helm v3.x
* kubectl v1.x
* docker
* harbor
  * <https://harbor.arfa.wise-paas.com/harbor>

## Flow (6 Steps)

1. 修改代碼, 重做image並推上harbor
2. 使用本機helm註冊harbor上的chart repo (只有第一次需要)
3. 如果chart也有修改, 也需要把它推上harbor
4. 去harbor上拿到預設的 chart values.yaml內容, 依據不同環境修改內容
5. 搭配values.yaml, 使用helm install將app部署上k8s
6. 下指令檢查部署狀態

## Step 1 - 每次改code(或commit新東西)都要重新做image上harbor, 我都在linux做 (因為個人電腦防毒不能裝docker)

\[linux] /home/wadev3/roy/buildspace/datahub-portal

\[linux] /home/wadev3/roy/buildspace/datahub-dataworker

改完code (或pull完), 直接在專案根目錄跑指令, **指令可以一鍵做image+上傳到harbor**:

第一次push前, 一定要先docker login

```
sudo docker login https://harbor.arfa.wise-paas.com/
```

**portal**

```
flag="harbor.arfa.wise-paas.com/datahub-dev/datahub-portal:2.0.1" \
&& sudo docker build -t=$flag --no-cache=true -f="./dockerfiles/localbuild.dockerfile" . \
&& sudo docker push $flag
```

**worker**

```
flag="harbor.arfa.wise-paas.com/datahub-dev/datahub-dataworker:2.0.1" \
&& sudo docker build -t=$flag --no-cache=true -f="./dockerfiles/localbuild.dockerfile" . \
&& sudo docker push $flag
```

## Step 2 - 在本機helm註冊chart repo, 此步驟跑過一次就好

```
helm repo add --username harborUser --password harborPwd datahub-dev https://harbor.arfa.wise-paas.com/chartrepo/datahub-dev
```

* datahub-dev是helm在本機註冊的別名
* helm repo ls

## Step 3 - 如果chart也有修改, 也需要把它推上harbor (若無修改請略過)

helm push datahub-chart/ datahub-dev

## Step 4 - 準備helm install file

**標準流程:**&#x20;

* **去harbor找到要用的chart, 複製values內容, 修改值(例如env.\*)**
* 需要修改的地方有
  * portal/worker的image細節 (repo/tag)
    * **開發階段請將pullPolicy改成Always, 因為production預設值為IfNotPresent**
  * global.database.secretName
  * * 多合一的secret
  * global.url.host
    * 規則是 .$namespace-name.$clustername.en.internal
  * ingress.hosts.host
    * external url裡可讓開發者修改的部分, 可包含dash, 預設是portal-datahub, 如果space已經有, 請改別的名子, 避免url衝突, 例如portal-datahub-dev
    * 例如我要佈一個我自己測試的datahub portal, 這欄位的值我會換成portal-datahub-roytest

&#x20;**url 內外部映射範例**&#x20;

**example 1**

* url.host: .datahub.eks008.en.internal
* ingress.hosts.host: portal-datahub
* external:[https://portal-datahub-datahub-eks008.sa.wise-paas.com](https://portal-datahub-datahub-eks008.sa.wise-paas.com/)

**example 2**

* url.host: .datahub.eks008.en.interna
* ingress.hosts.host: portal-datahub-roytest
* external: [https://portal-datahub-roytest-datahub-eks008.sa.wise-paas.com](https://portal-datahub-datahub-eks008.sa.wise-paas.com/)

***強烈建議***

因為要自己組外部url太麻煩了, 各個站點的domain也都不同, 如果該站點route-api有work, 可直接call api拿外部url

流程為

* kubectl get ingress datahub
* 拿到internal url: portal-datahub.ifactory.eks002.en.internal
* 呼叫 GET [https://api-router-ensaas.sa.wise-paas.com/v1/routers/domain/INGRESS-INTERNAL-URL/external](https://api-router-ensaas.sa.wise-paas.com/v1/routers/domain/portal-datahub.datahub.ews001.internal/external)
  * ex. <https://api-router-ensaas.sa.wise-paas.com/v1/routers/domain/portal-datahub.ifactory.eks002.en.internal/external>
* 返回結果裡的data就是外部url

![](/files/-M4wc-wcK7pvsxBqJtdV)

## Step 5 - 佈到k8s上 helm install

**重要請看這裡**

* 你可能會不知道要用datahab或datahub-dev裡的chart, 這裡提供一個簡單的準則
* 當你開發時沒動到chart, 你可以配合release chart和測試的image, 但如果你測試需要動到chart, 那就自己手動推chart到datahub-dev, 並使用它來測試

helm install CHART\_NAME\_ON\_SERVER datahub-dev/datahub --version 2.0.0 -f values.yaml

> helm install datahub-roy datahub-dev/datahub --version 2.0.0 -f slave04-scada-scada-install.yaml

CHART\_NAME\_ON\_SERVER請佈datahub以外的名子 (共用的datahub我取這個名子)

## Step 6 - 下一些指令檢查狀態

```
## 列出所有佈到k8s上的helm chart
helm ls

## 列出某一個install完的chart狀態
helm CHART_NAME status

## 列出所有pod (可透過-n指定namespace, 或是直接在k8s config指定namespace)
kubectl get pod

## 列出pod的log
kubectl log POD_NAME
```


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://advwacloud.gitbook.io/rd-newbie-guide/kai-fa-huan-jing/k8s_datahub_dev.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.