使用 RESTful API 串接 Open Policy Agent

上一篇『初探 Open Policy Agent 實作 RBAC (Role-based access control) 權限控管』介紹了如何透過 Go 語言直接嵌入 Open Policy Agent (簡稱 OPA)設定檔，並透過 Go 套件直接查詢使用者權限。由於目前 OPA 只有支援三種模式串接各種不同的 Application，一種是透過 Go 語言直接整合，詳細請看上一篇教學，另一種是透過 RESTful API，也就是本篇的教學，最後一種是透過 WebAssembly 讓其他 application 可以直接讀取。之後有機會再來寫 WebAssembly 教學。而本篇將帶您了解如何透過 RESTful API 方式來完成 RBAC 權限控管，其實我比較期待支援 gRPC 模式，但是看到這篇 issue 提到，OPA 現在已經支援 Plugin 模式，大家想擴充的，可以自行處理。

請求流程

可以看到上述圖片看到整個 OPA 系統流程，OPA 就是確保各種 API Request 的權限，首先第一步驟會帶著 Auuthorization Token 去跟單一個服務詢問，接著此服務就會將資料帶到 OPA 進行查詢此請求是否有權限可以存取，最後 OPA 會回傳結果，接著再由服務端決定要回送哪些訊息。那本篇的重點會在下圖部分

準備查詢資料

底下是 OPA 的系統流程圖

我們可以很清楚知道，要拿到最後的 Query Result (JSON) 需要有三個 Input 的，分別是

Query Input (JSON)
Data (JSON)
Policy (Rego)

就以 RBAC 為範例，第一個 Query Input 的資料，就需要帶入像是使用者所在的群組 (user)，使用者現在要執行的動作 (action)，使用者要對什麼資源做事情 (object)。這三個資料轉成 JSON 格式如下:

{
  "input": {
    "user": ["design_group_kpi_editor", "system_group_kpi_editor"],
    "action": "edit",
    "object": "design"
  }
}

第二項就是準備系統內建的 Data 資料給 OPA，上述資料可以看到 user 所在的群組資訊，但是這些群組能做哪些事情，是 OPA 沒辦法知道的，所以需要將這些資料整理成 JSON 格式，並且上傳到 OPA 系統內

{
  "group_roles": {
    "admin": ["admin"],
    "quality_head_design": ["quality_head_design"],
    "quality_head_system": ["quality_head_system"],
    "quality_head_manufacture": ["quality_head_manufacture"],
    "kpi_editor_design": ["kpi_editor_design"],
    "kpi_editor_system": ["kpi_editor_system"],
    "kpi_editor_manufacture": ["kpi_editor_manufacture"],
    "viewer": ["viewer"],
    "viewer_limit_ds": ["viewer_limit_ds"],
    "viewer_limit_m": ["viewer_limit_m"],
    "design_group_kpi_editor": ["kpi_editor_design", "viewer_limit_ds"],
    "system_group_kpi_editor": ["kpi_editor_system", "viewer_limit_ds"],
    "manufacture_group_kpi_editor": ["kpi_editor_manufacture", "viewer"],
    "project_leader": ["viewer_limit_ds", "viewer_limit_m"]
  },
  "role_permissions": {
    "admin": [
      {"action": "view_all", "object": "design"},
      {"action": "edit", "object": "design"},
      {"action": "view_all", "object": "system"},
      {"action": "edit", "object": "system"},
      {"action": "view_all", "object": "manufacture"},
      {"action": "edit", "object": "manufacture"}
    ],
    "quality_head_design": [
      {"action": "view_all", "object": "design"},
      {"action": "edit", "object": "design"},
      {"action": "view_all", "object": "system"},
      {"action": "view_all", "object": "manufacture"}
    ],
    "quality_head_system": [
      {"action": "view_all", "object": "design"},
      {"action": "view_all", "object": "system"},
      {"action": "edit", "object": "system"},
      {"action": "view_all", "object": "manufacture"}
    ],
    "quality_head_manufacture": [
      {"action": "view_all", "object": "design"},
      {"action": "view_all", "object": "system"},
      {"action": "view_all", "object": "manufacture"},
      {"action": "edit", "object": "manufacture"}
    ],
    "kpi_editor_design": [
      {"action": "view_all", "object": "design"},
      {"action": "edit", "object": "design"}
    ],
    "kpi_editor_system": [
      {"action": "view_all", "object": "system"},
      {"action": "edit", "object": "system"}
    ],
    "kpi_editor_manufacture": [
      {"action": "view_all", "object": "manufacture"},
      {"action": "edit", "object": "manufacture"}
    ],
    "viewer": [
      {"action": "view_all", "object": "design"},
      {"action": "view_all", "object": "system"},
      {"action": "view_all", "object": "manufacture"}
    ],
    "viewer_limit_ds": [
      {"action": "view_all", "object": "design"},
      {"action": "view_all", "object": "system"}
    ],
    "viewer_limit_m": [{"action": "view_l3_project", "object": "manufacture"}]
  }
}

上述 Data 可以知道 Group 跟 Role 的對應關係，以及 Role 可以做得相對應事情。最後一項就是撰寫 OPA Policy，要透過 Rego 語言來撰寫，其實沒有很難。

package rbac.authz

import data.rbac.authz.acl
import input

# logic that implements RBAC.
default allow = false

allow {
    # lookup the list of roles for the user
    roles := acl.group_roles[input.user[_]]

    # for each role in that list
    r := roles[_]

    # lookup the permissions list for role r
    permissions := acl.role_permissions[r]

    # for each permission
    p := permissions[_]

    # check if the permission granted to r matches the user's request
    p == {"action": input.action, "object": input.object}
}

上面就是先把 User 對應的 Group Role 找到之後，再將全部的 Role 權限拿出來進行最後的比對產生結果，回傳值就會是 allow 布林值。

RESTful API

上述步驟將檔案整理成底下三個

input.json
data.json
rbac.authz.rego

透過底下指令依序將資料上傳到 OPA Server 內，第一個先上傳 data

1
2

curl -X PUT http://localhost:8181/v1/data/rbac/authz/acl \
  --data-binary @data.json

接著上傳 Policy

1
2

curl -X PUT http://localhost:8181/v1/policies/rbac.authz \
  --data-binary @rbac.authz.rego

最後驗證 input 資料

1
2

curl -X POST http://localhost:8181/v1/data/rbac/authz/allow \
  --data-binary @input.json

用 bat tool 驗證

$ bat POST http://localhost:8181/v1/data/rbac/authz/allow < input.json
POST /v1/data/rbac/authz/allow HTTP/1.1
Host: localhost:8181
Accept: application/json
Accept-Encoding: gzip, deflate
Content-Type: application/json
User-Agent: bat/0.1.0

{"input":{"action":"edit","object":"design","user":["design_group_kpi_editor","system_group_kpi_editor"]}}

HTTP/1.1 200 OK
Content-Type: application/json
Date: Sat, 01 May 2021 08:43:30 GMT
Content-Length: 15

{
  "result": true
}

或者可以透過簡單的 Go 語言來驗證，用 Go 1.16 新的 embed package 來驗證。

package main

import (
    "bytes"
    _ "embed"
    "fmt"
    "io/ioutil"
    "net/http"
    "time"
)

//go:embed input.json
var input []byte

func main() {
    url := "http://localhost:8181/v1/data/rbac/authz/allow"
    method := "POST"

    payload := bytes.NewReader(input)

    client := &http.Client{
        Timeout: 5 * time.Second,
    }
    req, err := http.NewRequest(method, url, payload)

    if err != nil {
        fmt.Println(err)
        return
    }
    req.Header.Add("Content-Type", "application/json")

    res, err := client.Do(req)
    if err != nil {
        fmt.Println(err)
        return
    }
    defer res.Body.Close()

    body, err := ioutil.ReadAll(res.Body)
    if err != nil {
        fmt.Println(err)
        return
    }
    fmt.Println(string(body))
}

結論

本篇透過 RESTful 方式來更新 Data 及 Policy，此方式相對於前篇直接坎入在 Go 語言內，效率上來說會比較差些，因為會牽扯到你把 OPA 部署在哪個環境內，中間肯定會有一些 Latency，不過這就要看團隊怎麼使用 OPA 了，各有優缺點，團隊如果不是在寫 Go 語言的，肯定只能用 RESTful 方式來更新及查詢。程式碼可以在這邊找到。

golang open policy agent

請求流程

準備查詢資料

RESTful API

結論

See also