版本
架构
springboot
logback
grafana alloy
grafana loki
grafana
实施
springboot
<!-- 可观测 开始 -->
<!-- 可观测 logs -->
<dependency>
<groupId>net.logstash.logback</groupId>
<artifactId>logstash-logback-encoder</artifactId>
<version>5.2</version>
</dependency>
<dependency>
<groupId>ch.qos.logback</groupId>
<artifactId>logback-classic</artifactId>
<version>1.2.3</version>
</dependency>
<dependency>
<groupId>ch.qos.logback</groupId>
<artifactId>logback-core</artifactId>
<version>1.2.3</version>
</dependency>
<dependency>
<groupId>ch.qos.logback</groupId>
<artifactId>logback-access</artifactId>
<version>1.2.3</version>
</dependency>
<!-- 可观测 结束 -->
<?xml version="1.0" encoding="UTF-8"?>
<configuration scan="true" scanPeriod="60 seconds" debug="false">
<appender name="CONSOLE_JSON" class="ch.qos.logback.core.ConsoleAppender">
<encoder class="net.logstash.logback.encoder.LogstashEncoder">
<fieldNames>
<timestamp>timestamp</timestamp>
<message>message</message>
<logger>logger</logger>
<thread>thread</thread>
<level>level</level>
<levelValue>[ignore]</levelValue>
</fieldNames>
<includeMdc>true</includeMdc>
<throwableConverter class="net.logstash.logback.stacktrace.ShortenedThrowableConverter">
<maxDepthPerThrowable>50</maxDepthPerThrowable>
<rootCauseFirst>true</rootCauseFirst>
</throwableConverter>
</encoder>
</appender>
<root level="INFO">
<appender-ref ref="CONSOLE_JSON" />
</root>
<logger name="org.springframework" level="WARN"/>
<logger name="org.apache" level="WARN"/>
<logger name="com.netflix" level="WARN"/>
<logger name="reactor" level="WARN"/>
<logger name="org.eclipse.jetty" level="WARN"/>
<logger name="com.ibatis" level="DEBUG"/>
<logger name="java.sql" level="DEBUG"/>
<logger name="dao" level="DEBUG"/>
<logger name="com.gh" level="DEBUG"/>
<logger name="com.ghjl" level="DEBUG"/>
</configuration>
alloy
discovery.kubernetes "pods" {
role = "pod"
namespaces {
names = ["n1", "n2", "n3"]
}
}
discovery.relabel "pod_logs" {
targets = discovery.kubernetes.pods.targets
rule {
source_labels = ["__meta_kubernetes_pod_label_item_name"]
target_label = "item_name"
}
rule {
source_labels = ["__meta_kubernetes_pod_label_env_type"]
target_label = "env_type"
}
rule {
source_labels = ["__meta_kubernetes_pod_label_app_name"]
target_label = "app_name"
}
rule {
source_labels = ["__meta_kubernetes_pod_node_name"]
target_label = "node_name"
}
rule {
source_labels = ["__meta_kubernetes_pod_name"]
target_label = "pod_name"
}
rule {
source_labels = ["__meta_kubernetes_namespace"]
target_label = "pod_namespace"
}
rule {
source_labels = ["__meta_kubernetes_pod_container_name"]
target_label = "container_name"
}
}
loki.source.kubernetes "pod_logs" {
targets = discovery.relabel.pod_logs.output
forward_to = [loki.process.pod_logs.receiver]
}
loki.process "pod_logs" {
stage.cri {}
stage.json {
expressions = {
level = "level",
message = "message",
logger = "logger",
thread = "thread",
trace_id = "trace_id",
span_id = "span_id",
stackTrace = "stack_trace",
}
}
stage.template {
source = "message"
template = "{{ if .trace_id }}trace_id={{ .trace_id }} {{ end }}{{ .Value }}"
}
stage.labels {
values = {
level = "",
item_name = "item_name",
env_type = "env_type",
app_name = "app_name",
pod_name = "pod_name",
pod_namespace = "pod_namespace",
node_name = "node_name",
}
}
forward_to = [loki.write.default.receiver]
}
loki.write "default" {
endpoint {
name = "default"
url = "http://loki-gateway/loki/api/v1/push"
headers = {
"X-Scope-OrgID" = "tenant1",
}
}
external_labels = {
cluster = "kubernetes",
}
}
otelcol.receiver.otlp "default" {
grpc {
endpoint = "0.0.0.0:4317"
}
http {
endpoint = "0.0.0.0:4318"
}
output {
traces = [otelcol.processor.batch.default.input]
}
}
otelcol.processor.batch "default" {
timeout = "1s"
send_batch_size = 8192
send_batch_max_size = 8192
output {
traces = [otelcol.exporter.otlp.tempo.input]
}
}
otelcol.exporter.otlp "tempo" {
client {
endpoint = "http://tempo-distributed-distributor.monitoring.svc.cluster.local:4317"
tls {
insecure = true
}
}
timeout = "10s"
}
loki
# loki-values-retention.yaml
# Loki 完整 Helm values —— 基于现有部署追加保留策略与 Compactor 自动清理
# 用法: helm upgrade loki -n monitoring grafana/loki --repo https://grafana.github.io/helm-charts -f ./loki-values-retention.yaml
# 2026-07-15
# ============================================================
# 部署模式与镜像
# ============================================================
deploymentMode: Distributed
global:
imageRegistry: harbor.xxx.com/public
# ============================================================
# 后端组件（Distributed 模式下使用 microservices，backend 不启用）
# ============================================================
backend:
replicas: 0
read:
replicas: 0
write:
replicas: 0
singleBinary:
replicas: 0
# ============================================================
# Bloom 组件（当前未启用）
# ============================================================
bloomBuilder:
replicas: 0
bloomGateway:
replicas: 0
bloomPlanner:
replicas: 0
# ============================================================
# 微服务副本数
# ============================================================
distributor:
replicas: 3
maxUnavailable: 2
ingester:
replicas: 3
zoneAwareReplication:
enabled: false
querier:
replicas: 3
maxUnavailable: 2
queryFrontend:
replicas: 2
maxUnavailable: 1
queryScheduler:
replicas: 2
indexGateway:
replicas: 2
maxUnavailable: 1
compactor:
replicas: 1
# ============================================================
# Chunk 缓存 (memcached)
# ============================================================
chunksCache:
resources:
requests:
cpu: 500m
memory: 1Gi
limits:
memory: 2Gi
# ============================================================
# 查询结果缓存 (memcached)
# ============================================================
resultsCache:
resources:
requests:
cpu: 500m
memory: 1229Mi
limits:
memory: 1229Mi
# ============================================================
# S3 存储密钥（CEPH RGW secret）
# ============================================================
defaults:
extraEnvFrom:
- secretRef:
name: loki-bucket-claim
# ============================================================
# Loki 核心配置
# ============================================================
loki:
# -------- Schema --------
schemaConfig:
configs:
- from: "2024-04-01"
index:
period: 24h
prefix: loki_index_
object_store: s3
schema: v13
store: tsdb
# -------- S3 存储后端 (Ceph RGW) --------
storage:
type: s3
bucketNames:
admin: loki-data
chunks: loki-data
ruler: loki-data
s3:
endpoint: http://rook-ceph-rgw-ceph-objectstore.rook-ceph.svc.cluster.local:80
region: us-east-1
s3ForcePathStyle: true
# -------- 日志保留策略 --------
limits_config:
# 查询限制
max_cache_freshness_per_query: 10m
query_timeout: 300s
split_queries_by_interval: 15m
volume_enabled: true
# 写入保护：拒绝时间戳超过 7 天的旧日志
reject_old_samples: true
reject_old_samples_max_age: 168h
# ======== 核心：日志保留 30 天 ========
retention_period: 720h
# 可选：按标签定义差异化保留策略
retention_stream:
# # 示例：特定命名空间的日志保留 7 天
# - selector: '{namespace="test"}'
# priority: 1
# period: 168h
# # 示例：重要应用的日志保留 90 天
# - selector: '{app_name="payment-service"}'
# priority: 2
# period: 2160h
# 每个租户最大活跃流数，防止标签爆炸
max_global_streams_per_user: 100000
# -------- Ingester 内存控制 --------
ingester:
# Chunk 空闲 30 分钟后刷写到存储
chunk_idle_period: 30m
# Chunk 存活超过 2 小时强制刷写
max_chunk_age: 2h
# 刷写后在内存中保留 30 分钟供查询
chunk_retain_period: 30m
# WAL 在关机时刷写
wal:
flush_on_shutdown: true
# -------- Compactor 自动清理 --------
compactor:
# 启用保留期自动删除
retention_enabled: true
# 标记删除后延迟 2 小时再物理删除（避免误删）
retention_delete_delay: 2h
# 并行删除 worker 数
retention_delete_worker_count: 2
# 删除请求持久化到 S3，重启不丢失
delete_request_store: s3
QA
注意springboot 2.0.5.RELEASE的依赖版本