1
2

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?

OpenSearch Service でログ分析入門!

Posted at

はじめに

OpenSearch Service でログ分析に入門しましょう!以下の構成を作成します。

Cloud9 はファイル編集の手間を削減するため利用しており EC2 で置き換え可能です。

overview.png

作業概要

  1. OpenSearch ドメイン作成
  2. mappings 登録
  3. Cloud9 インスタンス作成
  4. fluent-bit インストール&設定
  5. ログ出力&転送
  6. OpenSearch Dashboards ログ分析

OpenSearch ドメイン作成

以下設定に注意して OpenSearch ドメインを作成します。

簡単に入門できるよう、ネットワークや認証を簡易化しています。

  • 「マネージド型クラスター」から「ドメインの作成」をクリック
  • 「ネットワーク」で「パブリックアクセス」を選択
  • 「ドメインの作成方法」で「標準作成」を選択
  • 「きめ細やかなアクセスコントロール」で「マスターユーザーの作成」を選択
  • 「アクセスポリシー」で「きめ細やかなアクセスコントロールのみを仕様」を選択

es_create1.png
es_create2.png

mappings 登録

OpenSearch Dashboards URL をクリックしてログインします。Management > Dev Tools から Console を開いて mappings を登録します。mappings とはデータベースにおけるスキーマのようなものです。

mappings.txt
PUT /apache_logs
{
  "mappings": {
    "properties": {
      "@timestamp": { "type": "date" },
      "host": { "type": "ip" },
      "user": { "type": "keyword" },
      "code": { "type": "integer" },
      "size": { "type": "long" },
      "referrer": { "type": "keyword" },
      "agent": { "type": "text" },
      "method": { "type": "keyword" },
      "path": { "type": "keyword" }
    }
  }
}

mappings.png

Cloud9 インスタンス作成

Cloud9 インスタンスを作成します。

  • 「環境タイプ」で「新しい EC2 インスタンス」を選択
  • 「ネットワーク設定」で「AWS Systems Manager (SSM)」を選択

fluent-bit インストール&設定

Cloud9 インスタンスに fluent-bit をインストールし、ログ転送の設定をします。

fluent-bit.conf
[SERVICE]
    Flush        1
    Grace        10
    Log_Level    info
    Parsers_File parsers.conf

[INPUT]
    Name             tail
    Path             /var/log/apache2_test/access.log
    Tag              apache.access

[FILTER]
    Name             parser
    Match            apache.access
    Key_Name         log
    Parser           apache2
    Reserve_Data     On

[OUTPUT]
    Name             es
    Match            apache.access
    Host             search-xxxxxxxx.ap-northeast-1.es.amazonaws.com
    Port             443
    HTTP_User        xxxxxxxx
    HTTP_Passwd      xxxxxxxx
    Index            apache_logs
    Type             _doc
    Suppress_Type_Name On
    Time_Key         @timestamp
    tls              On
# fluent-bit をインストール
curl https://raw.githubusercontent.com/fluent/fluent-bit/master/install.sh | sh
# 設定を編集
sudo vi /etc/fluent-bit/fluent-bit.conf
# fluent-bit を再起動
sudo systemctl restart fluent-bit
# サービスの状態が acvtive (running) か確認
sudo systemctl status fluent-bit
# エラーが出ていないかログを確認
journalctl -u fluent-bit -e

ログ出力&転送

Apache のダミーログを出力します。今回は Go の自作プログラムを使います。

main.go
package main

import (
  "fmt"
  "math/rand"
  "os"
  "strconv"
  "strings"
  "time"
)

type WeightedItem struct {
  item   string
  weight int
}

type LogConfig struct {
  ipAddresses      []WeightedItem
  endpointStatuses []WeightedItem
  userAgents       []WeightedItem
}

var logConfig = LogConfig{
  ipAddresses: []WeightedItem{
    {item: "192.168.1.1", weight: 50},
    {item: "10.0.0.1", weight: 30},
    {item: "172.16.0.1", weight: 10},
    {item: "8.8.8.8", weight: 5},
    {item: "1.1.1.1", weight: 5},
    {item: "192.168.0.10", weight: 20},
    {item: "10.0.0.2", weight: 15},
    {item: "172.16.0.2", weight: 8},
    {item: "8.8.4.4", weight: 3},
    {item: "9.9.9.9", weight: 2},
    {item: "192.168.1.100", weight: 25},
    {item: "10.0.0.100", weight: 12},
    {item: "172.16.0.100", weight: 6},
    {item: "208.67.222.222", weight: 4},
    {item: "208.67.220.220", weight: 3},
  },
  endpointStatuses: []WeightedItem{
    {item: "/,200", weight: 40},
    {item: "/about,200", weight: 20},
    {item: "/contact,200", weight: 15},
    {item: "/products,200", weight: 10},
    {item: "/products,404", weight: 5},
    {item: "/services,200", weight: 8},
    {item: "/services,500", weight: 2},
    {item: "/blog,200", weight: 18},
    {item: "/blog/post1,200", weight: 12},
    {item: "/blog/post2,404", weight: 3},
    {item: "/faq,200", weight: 10},
    {item: "/support,200", weight: 7},
    {item: "/support,503", weight: 1},
    {item: "/login,200", weight: 25},
    {item: "/login,401", weight: 5},
    {item: "/register,200", weight: 15},
    {item: "/profile,200", weight: 10},
    {item: "/profile,403", weight: 2},
    {item: "/settings,200", weight: 8},
    {item: "/logout,302", weight: 6},
    {item: "/api/v1/users,200", weight: 15},
    {item: "/api/v1/products,200", weight: 12},
    {item: "/api/v1/orders,200", weight: 10},
  },
  userAgents: []WeightedItem{
    {item: "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36", weight: 50},
    {item: "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/14.1.1 Safari/605.1.15", weight: 30},
    {item: "Mozilla/5.0 (X11; Linux x86_64; rv:89.0) Gecko/20100101 Firefox/89.0", weight: 20},
    {item: "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/92.0.4515.107 Safari/537.36", weight: 40},
    {item: "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/92.0.4515.107 Safari/537.36", weight: 25},
    {item: "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:90.0) Gecko/20100101 Firefox/90.0", weight: 15},
    {item: "Mozilla/5.0 (iPhone; CPU iPhone OS 14_6 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/14.1.1 Mobile/15E148 Safari/604.1", weight: 35},
    {item: "Mozilla/5.0 (iPad; CPU OS 14_6 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) CriOS/91.0.4472.80 Mobile/15E148 Safari/604.1", weight: 20},
    {item: "Mozilla/5.0 (Android 11; Mobile; rv:68.0) Gecko/68.0 Firefox/88.0", weight: 15},
  },
}

func weightedRandomChoice(items []WeightedItem) string {
  totalWeight := 0
  for _, item := range items {
    totalWeight += item.weight
  }

  randomNumber := rand.Intn(totalWeight)
  for _, item := range items {
    randomNumber -= item.weight
    if randomNumber < 0 {
      return item.item
    }
  }

  return items[0].item // This should never happen, but it's here as a fallback
}

func generateLogEntry() string {
  ipAddress := weightedRandomChoice(logConfig.ipAddresses)
  endpointStatus := weightedRandomChoice(logConfig.endpointStatuses)
  userAgent := weightedRandomChoice(logConfig.userAgents)

  parts := strings.Split(endpointStatus, ",")
  endpoint := parts[0]
  statusCode := parts[1]

  timestamp := time.Now().Format("02/Jan/2006:15:04:05 -0700")
  method := "GET"
  protocol := "HTTP/1.1"
  bytesSent := rand.Intn(5000) + 500
  referer := "-"

  return fmt.Sprintf("%s - - [%s] \"%s %s %s\" %s %d \"%s\" \"%s\"",
    ipAddress, timestamp, method, endpoint, protocol, statusCode, bytesSent, referer, userAgent)
}

func main() {
  if len(os.Args) != 2 {
    fmt.Println("Usage: go run main.go <lines_per_second>")
    os.Exit(1)
  }

  linesPerSecond, err := strconv.Atoi(os.Args[1])
  if err != nil || linesPerSecond <= 0 {
    fmt.Println("Please provide a valid positive integer for lines per second")
    os.Exit(1)
  }

  ticker := time.NewTicker(time.Second)
  defer ticker.Stop()

  for range ticker.C {
    for i := 0; i < linesPerSecond; i++ {
      fmt.Println(generateLogEntry())
    }
  }
}
# 出力用のディレクトリ準備
sudo mkdir /var/log/apache2_test
sudo chmod 777 /var/log/apache2_test
# ダミーログ出力プログラムを作成
vi main.go
# Apache のダミーログを出力(引数は 1 秒あたりの出力行数)
go run main.go 5 >> /var/log/apache2_test/access.log

Dashboards を使ったログ分析

インデックスパターンを作成し、分析できるように準備します。

index_pattern.png

インデックスパターンが作成できたら、Discover からログの分析が可能です!いろんな検索を試してみましょう!

discover_top.png
search_code2.png
search_agent2.png

Visualize を利用すれば、わかりやすい可視化が可能です!

dashboard_large.png

まとめ

OpenSearch Service であれば、ログの可視化が簡単にできます。今まで見向きもしていなかったログから、なにか新しいインサイトを得れるかも…!?

1
2
0

Register as a new user and use Qiita more conveniently

  1. You get articles that match your needs
  2. You can efficiently read back useful information
  3. You can use dark theme
What you can do with signing up
1
2

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?