docker-compose.yaml

@@ -29,5 +29,96 @@ services:
     ports:
       - "8080:8080"
 
+  grafana:
+    image: grafana/grafana:10.3.1
+    container_name: grafana
+    ports:
+      - "3000:3000"
+    depends_on:
+      prometheus:
+        condition: service_healthy
+      loki:
+        condition: service_healthy
+      tempo:
+        condition: service_started
+      alloy:
+        condition: service_started
+    environment:
+      GF_AUTH_ANONYMOUS_ENABLED: true
+      GF_SECURITY_ADMIN_USER: admin
+      GF_SECURITY_ADMIN_PASSWORD: admin
+      GF_METRICS_ENABLED: "true"
+    volumes:
+      - ./infrastructure/grafana/provisioning:/etc/grafana/provisioning
+      - ./infrastructure/grafana/dashboards:/var/lib/grafana/dashboards
+    healthcheck:
+      test: [ "CMD", "wget", "--spider", "http://localhost:3000/api/health" ]
+      interval: 10s
+      timeout: 5s
+      retries: 10
+
+  prometheus:
+    image: prom/prometheus:latest
+    container_name: prometheus
+    ports:
+      - "9090:9090"
+    volumes:
+      - ./infrastructure/prometheus/prometheus.yml:/etc/prometheus/prometheus.yml
+    healthcheck:
+      test: [ "CMD", "wget", "--spider", "http://localhost:9090/-/healthy" ]
+      interval: 10s
+      timeout: 5s
+      retries: 5
+
+  loki:
+    image: grafana/loki:2.9.2
+    container_name: loki
+    ports:
+      - "3100:3100"
+    volumes:
+      - ./infrastructure/loki/loki-config.yaml:/etc/loki/loki-config.yaml:ro
+    healthcheck:
+      test: [ "CMD", "wget", "--spider", "-q", "http://localhost:3100/ready" ]
+      interval: 10s
+      timeout: 5s
+      retries: 5
+
+  tempo:
+    image: grafana/tempo:2.4.1
+    container_name: tempo
+    command: [ "-config.file=/etc/tempo/tempo.yaml" ]
+    volumes:
+      - ./infrastructure/tempo/tempo.yaml:/etc/tempo/tempo.yaml
+      - tempo_data:/var/tempo
+    ports:
+      - "3200:3200"
+
+  alloy:
+    image: grafana/alloy:latest
+    container_name: alloy
+    user: root
+    ports:
+      - "9080:9080" # HTTP interface
+      - "4317:4317" # OTLP gRPC
+      - "4318:4318" # OTLP http
+    volumes:
+      - ./infrastructure/alloy/config.alloy:/etc/alloy/config.alloy:ro
+      - /var/lib/docker/containers:/var/lib/docker/containers:ro
+      - /var/run/docker.sock:/var/run/docker.sock
+      - /var/log:/var/log:ro
+    environment:
+      GRAFANA_LOKI_URL: http://loki:3100/loki/api/v1/push
+    command:
+      - run
+      - --server.http.listen-addr=0.0.0.0:9080
+      - --storage.path=/var/lib/alloy/data
+      - /etc/alloy/config.alloy
+
 volumes:
-  pgdata:
+  pgdata:
+  artifacts:
+  tempo_data:
+  loki_data:
+  loki_chunks:
+  loki_index:
+  loki_rules:

infrastructure/alloy/config.alloy

@@ -0,0 +1,74 @@
+loki.write "local" {
+  endpoint {
+    url = "http://loki:3100/loki/api/v1/push"
+  }
+}
+
+discovery.docker "containers" {
+  host             = "unix:///var/run/docker.sock"
+  refresh_interval = "5s"
+}
+
+discovery.relabel "containers" {
+  targets = discovery.docker.containers.targets
+
+  rule {
+    source_labels = ["__meta_docker_container_name"]
+    regex         = "/(.*)"
+    target_label  = "container"
+  }
+
+  rule {
+    target_label = "job"
+    replacement  = "docker"
+  }
+
+  rule {
+    target_label = "host"
+    replacement  = constants.hostname
+  }
+
+  rule {
+    source_labels = ["__meta_docker_container_name"]
+    regex         = "/(.*)"
+    target_label  = "service"
+  }
+
+  rule {
+    source_labels = ["__meta_docker_container_log_stream"]
+    target_label  = "stream"
+  }
+}
+
+loki.source.docker "containers" {
+  host          = "unix:///var/run/docker.sock"
+  targets       = discovery.docker.containers.targets
+  relabel_rules = discovery.relabel.containers.rules
+  forward_to    = [loki.write.local.receiver]
+}
+
+otelcol.receiver.otlp "default" {
+  grpc {
+    endpoint = "0.0.0.0:4317"
+  }
+  http {
+    endpoint = "0.0.0.0:4318"
+  }
+  output {
+    traces = [otelcol.exporter.otlp.tempo.input]
+    metrics = [otelcol.exporter.prometheus.default.input]
+  }
+}
+
+otelcol.exporter.otlp "tempo" {
+  client {
+    endpoint = "tempo:4317"
+    tls {
+      insecure = true
+    }
+  }
+}
+
+otelcol.exporter.prometheus "default" {
+  forward_to = []
+}

infrastructure/grafana/dashboards/postgres-overview.json

@@ -0,0 +1,43 @@
+{
+  "title": "Postgres Overview",
+  "timezone": "browser",
+  "schemaVersion": 39,
+  "version": 1,
+  "refresh": "10s",
+  "panels": [
+    {
+      "type": "gauge",
+      "title": "pg_up",
+      "targets": [{ "expr": "pg_up", "refId": "A" }],
+      "gridPos": { "x": 0, "y": 0, "w": 4, "h": 6 }
+    },
+    {
+      "type": "timeseries",
+      "title": "Active sessions by state",
+      "targets": [
+        { "expr": "sum by (state,instance) (pg_stat_activity_count)", "refId": "A" }
+      ],
+      "gridPos": { "x": 4, "y": 0, "w": 20, "h": 6 }
+    },
+    {
+      "type": "timeseries",
+      "title": "Transactions per second (commits+rollbacks)",
+      "targets": [
+        { "expr": "sum(rate(pg_stat_database_xact_commit[1m])) + sum(rate(pg_stat_database_xact_rollback[1m]))", "refId": "A" }
+      ],
+      "gridPos": { "x": 0, "y": 6, "w": 12, "h": 8 }
+    },
+    {
+      "type": "timeseries",
+      "title": "Tuples fetched/returned/inserted/updated/deleted",
+      "targets": [
+        { "expr": "sum(rate(pg_stat_database_tup_fetched[1m]))", "legendFormat": "fetched", "refId": "A" },
+        { "expr": "sum(rate(pg_stat_database_tup_returned[1m]))", "legendFormat": "returned", "refId": "B" },
+        { "expr": "sum(rate(pg_stat_database_tup_inserted[1m]))", "legendFormat": "inserted", "refId": "C" },
+        { "expr": "sum(rate(pg_stat_database_tup_updated[1m]))", "legendFormat": "updated", "refId": "D" },
+        { "expr": "sum(rate(pg_stat_database_tup_deleted[1m]))", "legendFormat": "deleted", "refId": "E" }
+      ],
+      "gridPos": { "x": 12, "y": 6, "w": 12, "h": 8 }
+    }
+  ]
+}

infrastructure/grafana/datasources/datasources.yaml

@@ -0,0 +1,20 @@
+apiVersion: 1
+
+datasources:
+  - name: Prometheus
+    type: prometheus
+    access: proxy
+    url: http://prometheus:9090
+    isDefault: true
+  - name: Loki
+    type: loki
+    access: proxy
+    url: http://loki:3100
+    isDefault: false
+    editable: true
+  - name: Tempo
+    type: tempo
+    access: proxy
+    url: http://tempo:3200
+    isDefault: false
+    editable: true

infrastructure/grafana/provisioning/dashboards/dashboards.yml

@@ -0,0 +1,11 @@
+apiVersion: 1
+
+providers:
+  - name: 'default'
+    folder: 'Individuals-api'
+    type: file
+    disableDeletion: false
+    editable: true
+    options:
+      path: /var/lib/grafana/dashboards
+      foldersFromFilesStructure: true

infrastructure/grafana/provisioning/datasources/datasources.yaml

@@ -0,0 +1,20 @@
+apiVersion: 1
+
+datasources:
+  - name: Prometheus
+    type: prometheus
+    access: proxy
+    url: http://prometheus:9090
+    isDefault: true
+  - name: Loki
+    type: loki
+    access: proxy
+    url: http://loki:3100
+    isDefault: false
+    editable: true
+  - name: Tempo
+    type: tempo
+    access: proxy
+    url: http://tempo:3200
+    isDefault: false
+    editable: true

infrastructure/loki/loki-config.yaml

@@ -0,0 +1,56 @@
+auth_enabled: false
+
+server:
+  http_listen_post: 3100
+
+ingester:
+  lifecycler:
+    address: 0.0.0.0
+    ring:
+      kvstore:
+        store: summary
+      replication_factor: 1
+    final_sleep: 0s
+  chunk_idle_period: 5m
+  chunk_retain_period: 30s
+  max_transfer_retries: 0
+
+schema_config:
+  configs:
+    - from: 2018-04-15
+      store: boltdb
+      object_store: filesystem
+      schema: v11
+      index:
+        prefix: index_
+        period: 168h
+
+storage_config:
+  boltdb:
+    directory: /loki/index
+  filesystem:
+    directory: /loki/chunks
+
+limits_config:
+  enforce_metric_name: false
+  reject_old_samples: true
+  reject_old_samples_max_age: 168h
+  ingestion_rate_mb: 8
+  ingestion_burst_size_md: 10
+
+chunk_storage_config:
+  max_look_back_period: 0s
+
+table_manager:
+  chunk_tables_provisioning:
+    inactive_read_throughput: 0
+    inactive_write_throughput: 0
+    provisioned_read_throughput: 0
+    provisioned_write_throughput: 0
+  index_tables_provisioning:
+    inactive_read_throughput: 0
+    inactive_write_throughput: 0
+    provisioned_read_throughput: 0
+    provisioned_write_throughput: 0
+  retention_deletes_enabled: true
+  retention_period: 672h

infrastructure/prometheus/prometheus.yml

@@ -0,0 +1,53 @@
+global:
+  scrape_interval: 15s
+  scrape_timeout: 10s
+  evaluation_interval: 15s
+
+
+scrape_configs:
+  - job_name: prometheus
+    static_configs:
+      - targets: [ "prometheus:9090" ]
+
+  - job_name: api
+    metrics_path: /actuator/prometheus
+    static_configs:
+      - targets: [ 'api:8091' ]
+
+  - job_name: persons-api
+    metrics_path: /actuator/prometheus
+    static_configs:
+      - targets: [ 'persons-api:8092' ]
+
+  - job_name: keycloak
+    metrics_path: /metrics
+    static_configs:
+      - targets: [ 'keycloak:9000' ]
+
+  - job_name: grafana
+    metrics_path: /metrics
+    static_configs:
+      - targets: [ 'grafana:3000' ]
+
+  - job_name: alloy
+    metrics_path: /metrics
+    static_configs:
+      - targets: [ 'alloy:9080' ]
+
+  - job_name: loki
+    metrics_path: /metrics
+    static_configs:
+      - targets: [ 'loki:3100' ]
+
+  - job_name: tempo
+    metrics_path: /metrics
+    static_configs:
+      - targets: [ 'tempo:3200' ]
+
+  - job_name: person-postgres
+    static_configs:
+      - targets: [ 'person-postgres-exporter:9187' ]
+
+  - job_name: keycloak-postgres
+    static_configs:
+      - targets: [ 'keycloak-postgres-exporter:9187' ]

infrastructure/tempo/tempo.yaml

@@ -0,0 +1,22 @@
+server:
+  http_listen_port: 3200
+
+distributor:
+  receivers:
+    otlp:
+      protocols:
+        grpc:
+        http:
+
+ingester:
+  trace_idle_period: 10s
+
+compactor:
+  compaction:
+    compacted_block_retention: 48h
+
+storage:
+  trace:
+    backend: local
+    local:
+      path: /var/tempo/traces

src/main/resources/logback.xml

@@ -0,0 +1,21 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE configuration>
+
+<configuration scan="true">
+
+    <statusListener class="ch.qos.logback.core.status.NopStatusListener"/>
+
+    <appender name="console" class="ch.qos.logback.core.ConsoleAppender">
+        <encoder>
+            <pattern>{"timestamp":"%d{yyyy-MM-dd'T'HH:mm:ss.SSSXXX}", "trace_id": "%X{trace_id}",
+                "span_id":"%X{span_id}", "level":"%level", "thread":"%thread",
+                "logger":"%logger{36}","message":"%msg"}%n
+            </pattern>
+        </encoder>
+    </appender>
+
+    <root level="INFO">
+        <appender-ref ref="console"/>
+    </root>
+
+</configuration>