业务请求：客户端 → 网关 → 应用服务 → 数据库 / 缓存 / MQ
                         │ Agent 采集与上报
                         ▼
                    OAP Server
                    │       ▲
              写入/查询     │ 查询 API
                    ▼       │
                  存储 ← OAP ← UI ← 浏览器

install -d -m 0755 /data/skywalking/config
install -d -m 0750 /data/skywalking/elasticsearch
chown 1000:0 /data/skywalking/elasticsearch
cd /data/skywalking
sysctl vm.max_map_count

if [ "$(sysctl -n vm.max_map_count)" -lt 262144 ]; then
  printf 'vm.max_map_count=262144\n' > /etc/sysctl.d/90-skywalking-es.conf
  sysctl -p /etc/sysctl.d/90-skywalking-es.conf
fi

name: skywalking
services:
  elasticsearch:
    image: docker.elastic.co/elasticsearch/elasticsearch:8.17.4
    restart: unless-stopped
    environment:
      discovery.type: single-node
      xpack.security.enabled: "false"
      ES_JAVA_OPTS: "-Xms1g -Xmx1g"
    volumes:
      - /data/skywalking/elasticsearch:/usr/share/elasticsearch/data
    healthcheck:
      test:
        - CMD-SHELL
        - "curl -fsS 'http://localhost:9200/_cluster/health?wait_for_status=yellow&timeout=5s' >/dev/null"
      interval: 10s
      timeout: 10s
      retries: 18
      start_period: 40s
    logging:
      driver: json-file
      options:
        max-size: "20m"
        max-file: "3"

  oap:
    image: apache/skywalking-oap-server:10.4.0
    restart: unless-stopped
    depends_on:
      elasticsearch:
        condition: service_healthy
    environment:
      SW_STORAGE: elasticsearch
      SW_STORAGE_ES_CLUSTER_NODES: elasticsearch:9200
      SW_STORAGE_ES_INDEX_REPLICAS_NUMBER: "0"
      SW_HEALTH_CHECKER: default
      SW_TELEMETRY: prometheus
      SW_CORE_RECORD_DATA_TTL: "3"
      SW_CORE_METRICS_DATA_TTL: "7"
      JAVA_OPTS: "-Xms1g -Xmx1g"
    volumes:
      - /data/skywalking/config:/skywalking/ext-config:ro
    ports:
      - "127.0.0.1:11800:11800"
      - "127.0.0.1:12800:12800"
      - "127.0.0.1:1234:1234"
    logging:
      driver: json-file
      options:
        max-size: "20m"
        max-file: "3"

  ui:
    image: apache/skywalking-ui:10.4.0
    restart: unless-stopped
    depends_on:
      - oap
    environment:
      SW_OAP_ADDRESS: http://oap:12800
    ports:
      - "127.0.0.1:8080:8080"
    logging:
      driver: json-file
      options:
        max-size: "20m"
        max-file: "3"

cd /data/skywalking
docker compose config -q
docker compose pull
docker compose up -d
docker compose ps
docker compose logs --tail=100 elasticsearch oap ui
curl -fsS http://127.0.0.1:12800/healthcheck

curl -fsS http://127.0.0.1:12800/graphql \
  -H 'Content-Type: application/json' \
  --data '{"query":"{checkHealth{score details}}"}'

docker compose exec -T elasticsearch \
  curl -fsS 'http://localhost:9200/_cluster/health?pretty'
docker compose exec -T elasticsearch \
  curl -fsS 'http://localhost:9200/_cat/indices?v'
curl -fsS http://127.0.0.1:1234/metrics -o /tmp/oap-metrics.txt
head -n 20 /tmp/oap-metrics.txt

ssh -p 22022 -N -L 18080:127.0.0.1:8080 ops@skywalking.example.com

install -d /data/skywalking/downloads
cd /data/skywalking/downloads
curl -fLO https://archive.apache.org/dist/skywalking/java-agent/9.4.0/apache-skywalking-java-agent-9.4.0.tgz
curl -fLO https://archive.apache.org/dist/skywalking/java-agent/9.4.0/apache-skywalking-java-agent-9.4.0.tgz.sha512
sha512sum -c apache-skywalking-java-agent-9.4.0.tgz.sha512
install -d /opt
tar -xzf apache-skywalking-java-agent-9.4.0.tgz -C /opt
test -f /opt/skywalking-agent/skywalking-agent.jar

export SW_AGENT_NAME='prod::orders-api'
export SW_AGENT_COLLECTOR_BACKEND_SERVICES='127.0.0.1:11800'
export SW_LOGGING_OUTPUT=CONSOLE
java -javaagent:/opt/skywalking-agent/skywalking-agent.jar \
  -jar /opt/orders-api/app.jar

java -javaagent:/opt/skywalking-agent/skywalking-agent.jar \
  -Dskywalking.agent.service_name='prod::orders-api' \
  -Dskywalking.collector.backend_service=127.0.0.1:11800 \
  -jar /opt/orders-api/app.jar

export SW_AGENT_NAME='prod::web-tomcat'
export SW_AGENT_COLLECTOR_BACKEND_SERVICES='127.0.0.1:11800'
export CATALINA_OPTS="${CATALINA_OPTS:-} -javaagent:/opt/skywalking-agent/skywalking-agent.jar"

FROM eclipse-temurin:21-jre
WORKDIR /app
COPY skywalking-agent/ /opt/skywalking-agent/
COPY app.jar /app/app.jar
ENV SW_LOGGING_OUTPUT=CONSOLE
ENTRYPOINT ["java", "-javaagent:/opt/skywalking-agent/skywalking-agent.jar", "-jar", "/app/app.jar"]

apiVersion: apps/v1
kind: Deployment
metadata:
  name: orders-api
  namespace: apps
spec:
  replicas: 2
  selector:
    matchLabels:
      app: orders-api
  template:
    metadata:
      labels:
        app: orders-api
    spec:
      containers:
        - name: app
          image: registry.example.com/orders-api:1.0.0
          ports:
            - name: http
              containerPort: 8080
          env:
            - name: SW_AGENT_NAME
              value: "prod::orders-api"
            - name: SW_AGENT_INSTANCE_NAME
              valueFrom:
                fieldRef:
                  fieldPath: metadata.name
            - name: SW_AGENT_COLLECTOR_BACKEND_SERVICES
              value: skywalking-oap.observability.svc:11800
            - name: SW_LOGGING_OUTPUT
              value: CONSOLE
          resources:
            requests:
              cpu: "250m"
              memory: "512Mi"
            limits:
              memory: "1Gi"
---
apiVersion: v1
kind: Service
metadata:
  name: orders-api
  namespace: apps
spec:
  selector:
    app: orders-api
  ports:
    - name: http
      port: 8080
      targetPort: http

kubectl -n apps get pods -l app=orders-api
kubectl -n apps logs deployment/orders-api --tail=100
kubectl -n observability get svc skywalking-oap
kubectl -n observability get endpointslices \
  -l kubernetes.io/service-name=skywalking-oap

python3 -m venv .venv
. .venv/bin/activate
python -m pip install 'apache-skywalking==1.2.0'
export SW_AGENT_NAME='prod::orders-python'
export SW_AGENT_COLLECTOR_BACKEND_SERVICES='127.0.0.1:11800'
export SW_AGENT_PROTOCOL=grpc
sw-python run python manage.py runserver 127.0.0.1:8000 --noreload

sw-python run -p gunicorn mysite.wsgi:application \
  --workers 2 --bind 0.0.0.0:8000

import _ "github.com/apache/skywalking-go"

go get github.com/apache/skywalking-go@v0.6.0
go mod tidy
go build -toolexec=/opt/skywalking-go-agent -a -o orders-api .
export SW_AGENT_NAME='prod::orders-go'
export SW_AGENT_REPORTER_GRPC_BACKEND_SERVICE='127.0.0.1:11800'
./orders-api

events {
    worker_connections 1024;
}

http {
    lua_package_path "/opt/skywalking-nginx-lua/lib/?.lua;;";
    lua_shared_dict tracing_buffer 20m;

    init_worker_by_lua_block {
        local buffer = ngx.shared.tracing_buffer
        buffer:set("serviceName", "prod::edge-gateway")
        buffer:set("serviceInstanceName", "edge-gateway-01")
        buffer:set("includeHostInEntrySpan", false)
        require("skywalking.util").set_randomseed()
        require("skywalking.client"):startBackendTimer("http://127.0.0.1:12800")
        skywalking_tracer = require("skywalking.tracer")
    }

    upstream orders_backend {
        server 127.0.0.1:18081;
    }

    server {
        listen 18080;

        location / {
            rewrite_by_lua_block {
                skywalking_tracer:start("orders-api")
            }
            proxy_pass http://orders_backend;
            proxy_set_header Host $host;
            proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;

            body_filter_by_lua_block {
                if ngx.arg[2] then
                    skywalking_tracer:finish()
                end
            }
            log_by_lua_block {
                skywalking_tracer:prepareForReport()
            }
        }
    }
}

openresty -t
openresty -s reload
curl -fsS http://127.0.0.1:18080/orders

<dependency>
    <groupId>org.apache.skywalking</groupId>
    <artifactId>apm-toolkit-logback-1.x</artifactId>
    <version>9.4.0</version>
</dependency>

<configuration>
    <appender name="STDOUT" class="ch.qos.logback.core.ConsoleAppender">
        <encoder class="ch.qos.logback.core.encoder.LayoutWrappingEncoder">
            <layout class="org.apache.skywalking.apm.toolkit.log.logback.v1.x.TraceIdPatternLogbackLayout">
                <Pattern>%d{yyyy-MM-dd HH:mm:ss.SSS} [%tid] [%thread] %-5level %logger{36} - %msg%n</Pattern>
            </layout>
        </encoder>
    </appender>
    <root level="INFO">
        <appender-ref ref="STDOUT"/>
    </root>
</configuration>

environment:
  SW_OTEL_RECEIVER: default
  SW_OTEL_RECEIVER_ENABLED_HANDLERS: "otlp-traces,otlp-metrics,otlp-logs"
  SW_RECEIVER_ZIPKIN: default
  SW_ZIPKIN_KAFKA_COLLECTOR_ENABLED: "false"
  SW_QUERY_ZIPKIN: default

environment:
  SW_ZIPKIN_ADDRESS: http://oap:9412

default:
  rate: 1000
  duration: 2000
services:
  - name: "prod::payment-api"
    rate: 10000
    duration: 1000

environment:
  SW_CORE_RECORD_DATA_TTL: "3"
  SW_CORE_METRICS_DATA_TTL: "7"

每日 Trace 原始数据量
≈ 每秒完成的请求数 × 86400 × 实际采样率 × 每条 Trace 平均字节数

所需磁盘
≈ 每日索引实际增长量 × 保留天数 × (1 + 副本数) + 索引/合并余量

rules:
  service_latency_rule:
    expression: sum(service_resp_time > 1000) >= 3
    period: 10
    silence-period: 10
    message: 'Service {name}: average response time exceeded 1000 ms in at least 3 minute buckets during the last 10 minutes'
    tags:
      level: WARNING
    hooks:
      - dingtalk.ops

  service_success_rate_rule:
    expression: sum((service_sla / 100) < 99) >= 3
    period: 10
    silence-period: 10
    message: 'Service {name}: success rate below 99% in at least 3 minute buckets during the last 10 minutes'
    tags:
      level: CRITICAL
    hooks:
      - dingtalk.ops

hooks:
  dingtalk:
    ops:
      text-template: |-
        {"msgtype":"text","text":{"content":"SkyWalking Alarm: %s"}}
      webhooks:
        - url: "https://oapi.dingtalk.com/robot/send?access_token=YOUR_TOKEN"
          secret: "YOUR_SIGNING_SECRET"

cd /data/skywalking
docker compose restart oap
docker compose logs --tail=100 oap

scrape_configs:
  - job_name: skywalking-oap
    scrape_interval: 15s
    static_configs:
      - targets: ["oap:1234"]

OAP:        SW_AUTHENTICATION
Java Agent: SW_AGENT_AUTHENTICATION

SW_STORAGE_ES_CLUSTER_NODES=es-01.internal:9200,es-02.internal:9200
SW_STORAGE_ES_HTTP_PROTOCOL=https
SW_ES_USER=skywalking
SW_ES_PASSWORD=<由密钥管理系统注入>
SW_STORAGE_ES_SSL_JKS_PATH=/skywalking/certs/es-truststore.jks
SW_STORAGE_ES_SSL_JKS_PASS=<由密钥管理系统注入>

cd /data/skywalking
docker compose ps
docker compose logs --since=10m --tail=200 oap
curl -fsS http://127.0.0.1:12800/healthcheck
ss -lnt | grep -E ':(11800|12800|8080|1234)\b'
docker compose exec -T elasticsearch \
  curl -fsS 'http://localhost:9200/_cluster/health?pretty'
docker compose exec -T elasticsearch \
  curl -fsS 'http://localhost:9200/_cat/allocation?v'